<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Aashok F&amp;C Advisory on Medium]]></title>
        <description><![CDATA[Stories by Aashok F&amp;C Advisory on Medium]]></description>
        <link>https://medium.com/@aashokfcadvisory?source=rss-a9a1dc867724------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*vGNU6QWgRQU1zg0w</url>
            <title>Stories by Aashok F&amp;amp;C Advisory on Medium</title>
            <link>https://medium.com/@aashokfcadvisory?source=rss-a9a1dc867724------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 22:11:57 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@aashokfcadvisory/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Insured on Paper, Exposed in Reality: Why Business Insurance Fails at Claim Time]]></title>
            <link>https://medium.com/@aashokfcadvisory/insured-on-paper-exposed-in-reality-why-business-insurance-fails-at-claim-time-6c1af8c21edd?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/6c1af8c21edd</guid>
            <category><![CDATA[business]]></category>
            <category><![CDATA[accounting]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[startup]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Wed, 18 Nov 2026 12:31:01 GMT</pubDate>
            <atom:updated>2026-11-18T12:31:01.982Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 30 OF 100 | MEDIUM EDITION | INSURANCE, SMALL BUSINESS, RISK MANAGEMENT, FINANCE, CYBERSECURITY</strong></p><p>Insured on Paper, Exposed in Reality: Why Business Insurance Fails at Claim Time</p><p><em>Mondelez claimed for NotPetya damage and was refused under a war exclusion. Most gaps are found only when the claim is cut.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>Most business owners buy insurance the way they sign software terms: quickly, and without reading. I don’t blame them — policy wording is hard going.</p><p>But I’ve seen claims cut by half because a building was under-insured and rejected because a condition wasn’t followed. The details vary by country and insurer; the pattern doesn’t.</p><p>In June 2017, a cyberattack known as NotPetya spread rapidly from Ukraine across the world, crippling computer systems at shipping lines, pharmaceutical companies and food manufacturers.</p><p>The US snack maker Mondelez International said the attack damaged thousands of laptops and servers and disrupted its operations. It claimed on its property insurance. Its insurer refused, pointing to an exclusion for “hostile or warlike action” — because governments had attributed the attack to a state actor.</p><p>The dispute went to court in the United States and was settled in 2022, years later. <strong>Mondelez had insurance. What it had not had was certainty about what that insurance would actually pay for.</strong></p><h3>The relief that turns into disbelief</h3><p>A fire damages a warehouse. The owner’s first thought is relief: “We are insured.”</p><p>Then the claim process begins.</p><p>The surveyor notes that the stock declared in the policy was far lower than the stock actually held, so the claim is reduced proportionately. Some goods were in a location not covered. A condition about fire safety equipment was not met. Business interruption — the profit lost while operations were down — was never insured at all.</p><p>The policy existed. <strong>The protection the owner believed in did not.</strong></p><h3>A lesson from history: insurance was born in a coffee house</h3><p>In the late 1600s, merchants, ship owners and underwriters gathered at Edward Lloyd’s coffee house in London to share shipping news and agree to cover the risks of voyages. From those conversations grew Lloyd’s of London.</p><p>Even then, everything depended on <strong>the wording of the promise</strong> and <strong>the honesty of the disclosure</strong>: what the ship carried, where it sailed, and what dangers were — and were not — covered. Three centuries later, most insurance disputes still turn on those same two things.</p><h3>The uncomfortable part</h3><p>Many businesses buy insurance the way they renew a subscription: automatically, at the lowest premium, with minimal reading.</p><p>The broker sends a renewal. The premium is compared with last year. The policy is filed.</p><p>Nobody checks whether the business has changed — new locations, more stock, new products, new risks, digital operations — since the policy was first written.</p><p><strong>Insurance is a promise with conditions. Most disputes are about the conditions.</strong></p><h3>Why claims are reduced or rejected</h3><p>1. <strong>Underinsurance.</strong> If the sum insured is lower than the value at risk, many property policies apply an <strong>average clause</strong>, reducing the claim in proportion. Insure ₹50 lakh of stock for ₹25 lakh, and a ₹10 lakh loss may pay only about ₹5 lakh.</p><p>2. <strong>Non-disclosure or misrepresentation.</strong> Wrong information about the business, location, safety measures or past claims can reduce or void cover.</p><p>3. <strong>Exclusions.</strong> Floods, earthquakes, terrorism, war, wear and tear, gradual deterioration, cyber events, pandemics, contractual liabilities — it depends on the wording.</p><p>4. <strong>Breached conditions.</strong> Fire extinguishers, alarms, storage standards and record-keeping requirements that were not met.</p><p>5. <strong>Late notification.</strong> Most policies require prompt intimation of loss.</p><p>6. <strong>Poor documentation.</strong> Without stock records, invoices, asset registers and financial statements, proving the loss is very difficult.</p><h3>The cover many businesses overlook</h3><h3>Business interruption (loss of profits)</h3><p>Property insurance may pay to rebuild a factory. It usually does not pay for the profit lost and fixed costs incurred while the factory is shut — unless you have business interruption cover. When severe floods hit Thailand in 2011, factories of many global manufacturers were shut for weeks or months; the resulting interruption losses were among the largest insured losses of that year. For many businesses, <strong>interruption costs more than the physical damage.</strong></p><h3>Public and product liability</h3><p>Claims from customers or third parties for injury, property damage or defective products.</p><h3>Professional indemnity</h3><p>For advisers, consultants, IT providers and professionals whose errors could cause financial loss to clients.</p><h3>Directors’ and officers’ (D&amp;O) liability</h3><p>Protection for directors and officers against claims arising from management decisions — increasingly relevant as regulatory scrutiny rises.</p><h3>Cyber insurance</h3><p>Ransomware, data breaches, business email compromise and outages can cost more than many physical losses. Cyber policies vary widely in what they cover and often require minimum security controls — and, as Mondelez learned, exclusions matter.</p><h3>Key-person insurance</h3><p>Protection against the financial impact of losing a founder or critical team member.</p><h3>Trade credit insurance</h3><p>Protection against customer non-payment — valuable for exporters and businesses with large customer concentrations.</p><h3>Marine and transit insurance</h3><p>For goods in movement, with clarity on which party bears risk under the trade terms used.</p><h3>The CLAIM Check</h3><p>Before every renewal, run <strong>the CLAIM Check</strong>:</p><p>· <strong>C — Cover what matters.</strong> List your key risks: assets, stock, people, liability, cyber, credit, interruption. Is each covered?</p><p>· <strong>L — Limits correct.</strong> Value assets on a reinstatement or replacement basis where appropriate; update stock and asset values annually; set interruption cover on realistic recovery periods.</p><p>· <strong>A — Assumptions disclosed.</strong> Tell the insurer about material changes: new premises, products, processes, stock increases.</p><p>· <strong>I — Instructions followed.</strong> Read the conditions and warranties — and make sure operations comply.</p><p>· <strong>M — Make records claim-ready.</strong> Asset registers, stock records, invoices, photographs and backups stored off-site or in the cloud; a claims protocol naming who notifies the insurer and what evidence to preserve.</p><h3>Insurance is not risk management</h3><p>Insurance transfers some financial consequences of risk. It does not prevent the event, protect your reputation, keep your customers or satisfy regulators. You still need fire safety, cybersecurity, quality control, contracts, compliance and continuity plans. Insurers increasingly reward good risk management with better terms — and penalise its absence.</p><h3>How you’ll know it’s working</h3><p>Good insurance management shows up before and after a claim:</p><p>· <strong>Sums insured</strong> are reviewed every year against current replacement values.</p><p>· <strong>Exclusions and conditions</strong> are known to the people who must comply with them.</p><p>· <strong>Claims</strong> are paid in full or close to it, within reasonable time.</p><p>· <strong>Gaps</strong> — business interruption, cyber, product liability, directors’ liability — are consciously accepted or covered.</p><p>· <strong>Documentation</strong> — stock records, asset registers, maintenance logs — is ready before any loss.</p><h3>Where this is heading</h3><p>Climate events, cyber attacks and supply-chain shocks are making losses larger and more frequent. Insurers are responding with higher premiums, stricter conditions and data-driven pricing that uses sensors, satellite images and claims history. <strong>Businesses with good risk controls and clean data will get better terms; those without will find cover more expensive — or unavailable.</strong></p><h3>What to do on Monday, by role</h3><p>· <strong>Owner:</strong> read the exclusions section of your largest policy yourself.</p><p>· <strong>CFO:</strong> compare sums insured with current asset and stock values.</p><p>· <strong>Operations head:</strong> confirm that conditions — fire systems, security, maintenance — are actually met.</p><p>· <strong>Broker:</strong> ask for a written gap analysis of risks you are not covered for.</p><h3>If you run a small business</h3><p>Consider an illustrative garment manufacturer whose factory suffers a fire. Its policy covers ₹5 crore of stock, but actual stock on the day is ₹8 crore after a large seasonal order. Under the average clause, the insurer pays only part of the loss — about five-eighths of the assessed amount.</p><p>Business interruption is not covered, so salaries and loan instalments continue during three months of closure without income. <strong>The owner had insurance. He did not have enough of the right insurance.</strong> After rebuilding, he reviews cover quarterly during peak seasons and adds business-interruption cover.</p><h3>“But…” — the best counterargument</h3><p><strong>The objection:</strong> “Insurance is a cost that rarely pays out. Insurers look for reasons to reject claims. We would rather keep the premium and take the risk ourselves.”</p><p><strong>The answer:</strong> Self-insuring small, predictable losses can be sensible — higher deductibles often reduce premiums. But catastrophic losses — a fire, a major liability claim, a months-long shutdown — can end a business that could have survived with the right cover. And most rejected claims stem from preventable problems: underinsurance, non-disclosure and breached conditions. <strong>The answer to a badly bought policy is a well-bought policy, not no policy.</strong></p><h3>Questions people ask me</h3><p><strong>“How do I know my sum insured is right?”</strong></p><p>Use reinstatement or replacement values for buildings and machinery, current stock values (including peak-season levels), and a realistic business interruption period. Review annually or after any major change.</p><p><strong>“Should I buy cyber insurance?”</strong></p><p>If your business depends on systems, holds customer data or makes digital payments, it is worth evaluating — but read exclusions carefully and improve basic security first, because insurers increasingly require it.</p><p><strong>“Who should handle a claim?”</strong></p><p>Name one person in advance, keep insurer contact details and policy numbers accessible, and notify promptly.</p><p><strong>Where things stand (October 2026):</strong> In India, the Insurance Regulatory and Development Authority of India (IRDAI) has issued customer-protection rules on claim settlement timelines and policy disclosures, and premiums for business insurance are generally deductible as business expenses, subject to tax rules. GST input tax credit on some insurance services is restricted. Confirm current positions with your broker and tax adviser.</p><h3>Mistakes I see most often</h3><p>· Renewing without reviewing.</p><p>· Choosing the cheapest premium regardless of cover.</p><p>· Under-declaring values to save premium.</p><p>· Assuming property cover includes business interruption or cyber.</p><p>· Delaying claim notification while “assessing the damage.”</p><h3>Try this in 30 minutes</h3><p>Take your main property policy and answer:</p><p>1. What is the sum insured for stock, and what was your highest stock level last year?</p><p>2. Does the policy include business interruption? For how many months?</p><p>3. List three exclusions that could matter to your business.</p><p>4. List three conditions you must follow — and check whether you do.</p><p>5. Where are your records stored if the premises burn down?</p><p>Every gap you find is a conversation with your broker before renewal — not after a loss.</p><h3>The short version</h3><p>· Many businesses discover gaps only when a claim is reduced or rejected — Mondelez’s NotPetya dispute shows how exclusions can decide outcomes.</p><p>· Common causes: underinsurance and average clauses, non-disclosure, exclusions, breached conditions, late notice and poor records.</p><p>· Overlooked covers include business interruption, liability, professional indemnity, D&amp;O, cyber, key-person, trade credit and transit.</p><p>· Run the CLAIM Check before every renewal: Cover, Limits, Assumptions, Instructions, Make records claim-ready.</p><p>· Insurance supports risk management; it never replaces it.</p><p><strong>The premium is the price of the promise. The wording is the promise itself. Read it.</strong></p><h3>Before you go</h3><p>Once a year, before renewal, sit down with your broker and run the CLAIM check. It’s a dull hour that may decide whether your business survives its worst week.</p><h3>Further reading</h3><p>· <em>Mondelez International, Inc. v. Zurich American Insurance Co.</em>, Circuit Court of Cook County, Illinois (filed 2018; settled 2022), as reported in business and legal press.</p><p>· Swiss Re Institute, <em>sigma</em> reports on natural catastrophes and man-made disasters (2011 edition, Thailand floods). swissre.com</p><p>· Lloyd’s of London — history. lloyds.com</p><p>· Insurance Regulatory and Development Authority of India — regulations on protection of policyholders’ interests. irdai.gov.in</p><p>◆ ◆ ◆</p><p><em>A version of this article first appeared on ComplianceCalendar.in.</em></p><p><em>A note on numbers for readers outside India: 1 lakh = 100,000 and 1 crore = 10 million (₹ = Indian rupees).</em></p><p><em>I’m Aashok Kumar Sharma, founder of Aashok F&amp;C Advisory — a finance, accounting and compliance practice working with founders, SMEs and finance teams in India and abroad. I write about the numbers, rules and decisions that quietly decide whether businesses last. Follow me here on Medium for the rest of this 100-part series.</em></p><p><em>This article is general information, not legal, tax or investment advice. Rules change; check current provisions or speak to a qualified professional before acting. Examples labelled illustrative are composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6c1af8c21edd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Cost of a Cheap Accountant Is Never on the Invoice]]></title>
            <link>https://medium.com/@aashokfcadvisory/the-cost-of-a-cheap-accountant-is-never-on-the-invoice-066ea1f2c2b1?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/066ea1f2c2b1</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[accounting]]></category>
            <category><![CDATA[finance]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Tue, 17 Nov 2026 12:26:01 GMT</pubDate>
            <atom:updated>2026-11-17T12:26:01.477Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 29 OF 100 | MEDIUM EDITION | ACCOUNTING, SMALL BUSINESS, ENTREPRENEURSHIP, TAX, FINANCE</strong></p><p>The Cost of a Cheap Accountant Is Never on the Invoice</p><p><em>Arthur Andersen fell within months. Most bad accounting fails more quietly — through missed deductions, notices and decisions made on wrong numbers.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>I’ll declare my interest up front: I’m an accountant. So take this with whatever salt you like. But I’ve also spent years cleaning up after arrangements that looked cheap and turned out to be very expensive.</p><p>The true cost of poor accounting rarely appears on the fee invoice. It shows up later, as penalties, lost credits, rejected loans and decisions made on numbers nobody checked — in any country.</p><p>In 2001, Arthur Andersen was one of the five largest accounting firms in the world, with tens of thousands of employees across dozens of countries.</p><p>Within months of the collapse of its client Enron, the firm was finished. Its conviction in 2002 for obstruction of justice was later overturned by the US Supreme Court in 2005 — but by then, clients had left and the firm had effectively ceased practising.</p><p>The Andersen story is usually told as a lesson for auditors. It is also a lesson for clients: <strong>an adviser who becomes too eager to please the person paying the fee is not protecting that person. They are endangering them.</strong></p><h3>The “small fee” that wasn’t</h3><p>A business owner proudly mentions that their accounting and compliance work costs very little. “Just a small fee every month. They file everything.”</p><p>Then a notice arrives.</p><p>Input tax credit was claimed on invoices the supplier never reported. TDS was deducted but filed with errors. Expenses were booked without proper documents. Returns were filed, but nobody checked whether they matched the books.</p><p>The fee was low. <strong>The cost was not.</strong></p><h3>A lesson from history: why professions exist</h3><p>In 1854, a group of accountants in Edinburgh received a royal charter as the Society of Accountants in Edinburgh — often cited as the first professional body of accountants. Professional institutes followed around the world, including the Institute of Chartered Accountants of India, established by an Act of Parliament in 1949.</p><p>Professions were created for a reason: <strong>clients usually cannot judge the quality of specialist work themselves</strong>, so society needed standards, ethics and accountability to protect them. That protection works best when clients understand what they are actually buying.</p><h3>What you are really paying for</h3><p>Many owners see accounting as data entry and form filing — find someone who can do it cheaply and the job is done.</p><p>But good accounting and advisory work is about:</p><p>· <strong>Accuracy</strong> — transactions recorded correctly, with evidence.</p><p>· <strong>Judgement</strong> — knowing what the law requires and how it applies to your facts.</p><p>· <strong>Anticipation</strong> — spotting problems before they become notices.</p><p>· <strong>Advice</strong> — structuring decisions to be compliant and efficient.</p><p>· <strong>Insight</strong> — turning numbers into information you can act on.</p><p>The fee pays for all of these — or for only the first, done poorly.</p><p><strong>You do not pay an accountant for the hours they spend. You pay for the mistakes you never make.</strong></p><h3>The hidden costs of poor accounting</h3><p>· <strong>Penalties and interest</strong> — late filings, wrong returns and mismatches add up, often exceeding years of professional fees.</p><p>· <strong>Lost tax benefits</strong> — missed deductions, unclaimed credits, wrong regime choices, poor structuring.</p><p>· <strong>Notices and disputes</strong> — time, money and management attention, sometimes for years.</p><p>· <strong>Bad decisions</strong> — inaccurate or late reports mean pricing, hiring and investment by guesswork.</p><p>· <strong>Weaker credit and funding</strong> — unreliable statements reduce borrowing capacity and investor confidence.</p><p>· <strong>Fraud opportunities</strong> — weak reconciliations let errors and fraud go undetected.</p><h3>Warning signs your current arrangement is not enough</h3><p>· You receive financial statements only once a year.</p><p>· Nobody explains the numbers in plain language.</p><p>· Returns are filed, but reconciliations are never shared.</p><p>· You learn about deadlines when penalties arrive.</p><p>· Questions are answered with “it’s always done this way.”</p><p>· Your accountant agrees with every suggestion — including risky ones.</p><p>· Notices are handled late or without a clear strategy.</p><p>· There is no one to talk to about structure, planning or growth.</p><h3>The uncomfortable part</h3><p>Some owners choose cheap advisers deliberately — because they want someone who will “adjust” things without asking difficult questions.</p><p>That arrangement feels convenient until the day the adjustments must be defended.</p><p><strong>An adviser who never says no is not protecting you. They are protecting their fee.</strong></p><h3>The SCOPE Test for choosing an adviser</h3><h3>S — Specialisation</h3><p>A proprietorship, a growing private company, a funded startup, an exporter and a group with international transactions need different expertise. Ask about relevant experience in your industry and situation.</p><h3>C — Controls and process</h3><p>How do they reconcile your accounts? Track deadlines? Match GST and TDS data with your books? Who reviews the work before it is filed?</p><h3>O — Openness</h3><p>Do they tell you about legal changes, deadlines and risks before you ask? Do they explain in plain language? Will they tell you “no” — and show you the lawful way to reach your goal?</p><h3>P — Professional standing</h3><p>Verify qualifications and membership of the relevant professional body. Check for independence where the law requires it (for example, your statutory auditor should not also be the person keeping your books in situations where rules prohibit it).</p><h3>E — Engagement letter</h3><p>A written engagement letter should state exactly what is included — bookkeeping, filings, reconciliations, advisory, notices — what is not, the fees and the timelines.</p><h3>Choosing the right level of support</h3><p>As businesses grow, they often move through stages:</p><p>· <strong>Very small</strong> — Typical support: Bookkeeping plus compliance filings by a qualified firm</p><p>· <strong>Growing</strong> — Typical support: Monthly management reporting, reconciliations, tax planning</p><p>· <strong>Scaling</strong> — Typical support: A part-time or virtual CFO plus specialist tax and audit advisers</p><p>· <strong>Established</strong> — Typical support: In-house finance team, supported by external specialists</p><h3>How you’ll know it’s working</h3><p>A good accountant shows up in what does not happen:</p><p>· <strong>No surprise notices</strong> or penalties from tax and regulatory authorities.</p><p>· <strong>Timely, accurate monthly reports</strong> you can use for decisions.</p><p>· <strong>Proactive advice</strong> — warnings and opportunities raised before you ask.</p><p>· <strong>Clean audits</strong> with few adjustments.</p><p>· <strong>Clear documentation</strong> of what was filed, when and on what basis.</p><h3>Where this is heading</h3><p>AI and automation will make basic bookkeeping and return filing cheaper. The value of an accountant will shift further towards judgement, advice, controls, planning and handling tax authorities’ data-driven scrutiny. <strong>The cheapest accountant of the future will be software. The most valuable will be the professional who tells you what the numbers mean — and what to do next.</strong></p><h3>What to do on Monday, by role</h3><p>· <strong>Owner:</strong> apply the SCOPE Test to your current accountant or firm.</p><p>· <strong>Finance team:</strong> list every compliance task and who is responsible for it.</p><p>· <strong>Accountant:</strong> prepare a list of risks and improvement ideas for the business.</p><p>· <strong>Both:</strong> agree a written engagement letter that defines scope, timelines and responsibilities.</p><h3>If you run a small business</h3><p>Consider an illustrative owner of a ₹10 crore business who pays an accountant ₹15,000 a month — the lowest quote he found. Returns are filed, but nobody reviews input tax credit mismatches, TDS defaults or the treatment of a large related-party loan.</p><p>Two years later, notices arrive: credit disallowed, interest on TDS defaults and questions about the loan. The total cost exceeds ₹30 lakh. <strong>A more experienced adviser at twice the fee would have cost about ₹3.6 lakh more over those two years.</strong> The cheap option was the expensive one.</p><h3>“But…” — the best counterargument</h3><p><strong>The objection:</strong> “All accountants file the same forms. A return is a return. Why pay three times as much for the same output?”</p><p><strong>The answer:</strong> The form is the same; what goes into it is not. Two returns can look identical and carry very different risks: one reconciled with books, supplier data and bank statements; the other filled in from incomplete records. The difference appears only when a notice, an audit or a loan application tests it. <strong>You are not buying a form. You are buying confidence that it will hold.</strong></p><h3>Questions people ask me</h3><p><strong>“How often should my accountant meet me?”</strong></p><p>At least quarterly for a small business, and monthly if you are growing fast or facing cash pressure — even for 30 minutes.</p><p><strong>“Should I give my accountant all my portal passwords?”</strong></p><p>Give access through proper authorisations where portals allow, keep a record of who has access, and change credentials when relationships end. Never let any single person control everything without oversight.</p><p><strong>“Is it bad to change accountants?”</strong></p><p>Not if there is a good reason — but frequent changes lose continuity. When changing, insist on a proper handover of records, working papers and pending matters.</p><p><strong>Where things stand (October 2026):</strong> In India, the Income-tax Act, 2025 took effect on 1 April 2026, the labour codes in November 2025 and the DPDP Rules in November 2025, while GST and TDS systems increasingly auto-match data. Advisers who are not actively updating their knowledge and processes for these changes may expose clients to avoidable mistakes. Ask yours how they have adapted.</p><h3>How to be a good client</h3><p>· Share complete, timely information.</p><p>· Keep business and personal transactions separate.</p><p>· Ask questions and read the reports.</p><p>· Respond to queries quickly.</p><p>· Never ask your adviser to do something unlawful.</p><p>· Pay fairly for quality work.</p><h3>Mistakes I see most often</h3><p>· Choosing advisers only on the lowest quote.</p><p>· Handing over every credential without oversight or access logs.</p><p>· Never reviewing what has been filed in your name.</p><p>· Switching advisers every year.</p><p>· Ignoring advice because it is inconvenient.</p><h3>Try this in 30 minutes</h3><p>Ask your current accountant five questions this month:</p><p>1. Do our books reconcile with our GST, TDS and income-tax filings?</p><p>2. What compliance risks do you see in our business?</p><p>3. Are there deductions or incentives we may be missing?</p><p>4. What deadlines and legal changes are coming in the next six months?</p><p>5. What would you improve in our financial controls?</p><p>Judge the answers on clarity and specifics. <strong>Vague answers are your answer.</strong></p><h3>The short version</h3><p>· Accounting fees pay for accuracy, judgement, anticipation, advice and insight — not just data entry.</p><p>· Poor accounting creates hidden costs: penalties, lost benefits, notices, bad decisions, weaker credit and fraud risk.</p><p>· Warning signs include annual-only reporting, no reconciliations, surprise penalties and advisers who agree with everything — the Andersen lesson in miniature.</p><p>· Choose advisers with the SCOPE Test: Specialisation, Controls, Openness, Professional standing, Engagement letter.</p><p>· Be a good client: share information, keep finances separate and act on advice.</p><p><strong>Good advice is an investment. Bad advice is a liability that has not been invoiced yet.</strong></p><h3>Before you go</h3><p>Judge your accountant by what goes wrong when they’re not there, not by what they charge. And be a good client too: give them clean records and timely information. The relationship works best when both sides take it seriously.</p><h3>Further reading</h3><p>· <em>Arthur Andersen LLP v. United States</em>, 544 U.S. 696 (2005).</p><p>· Bethany McLean and Peter Elkind, <em>The Smartest Guys in the Room: The Amazing Rise and Scandalous Fall of Enron</em> (Portfolio, 2003).</p><p>· Institute of Chartered Accountants of Scotland — history (royal charter of 1854). icas.com</p><p>· Chartered Accountants Act, 1949 (India); Institute of Chartered Accountants of India. icai.org</p><p>◆ ◆ ◆</p><p><em>A version of this article first appeared on ComplianceCalendar.in.</em></p><p><em>A note on numbers for readers outside India: 1 lakh = 100,000 and 1 crore = 10 million (₹ = Indian rupees).</em></p><p><em>I’m Aashok Kumar Sharma, founder of Aashok F&amp;C Advisory — a finance, accounting and compliance practice working with founders, SMEs and finance teams in India and abroad. I write about the numbers, rules and decisions that quietly decide whether businesses last. Follow me here on Medium for the rest of this 100-part series.</em></p><p><em>This article is general information, not legal, tax or investment advice. Rules change; check current provisions or speak to a qualified professional before acting. Examples labelled illustrative are composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=066ea1f2c2b1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[New Software Won’t Fix a Business That Doesn’t Understand Its Own Numbers]]></title>
            <link>https://medium.com/@aashokfcadvisory/new-software-wont-fix-a-business-that-doesn-t-understand-its-own-numbers-2f74a0bc567f?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/2f74a0bc567f</guid>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[accounting]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Mon, 16 Nov 2026 12:26:01 GMT</pubDate>
            <atom:updated>2026-11-16T12:26:01.578Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 28 OF 100 | MEDIUM EDITION | TECHNOLOGY, ERP, ACCOUNTING, SMALL BUSINESS, DIGITAL TRANSFORMATION</strong></p><p>New Software Won’t Fix a Business That Doesn’t Understand Its Own Numbers</p><p><em>Hershey’s 1999 ERP go-live left it unable to ship Halloween candy. The lesson for every system upgrade: process first, software second.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>Every few years a client tells me the new system will finally fix their numbers. Sometimes it does. More often it reproduces the old confusion faster and more expensively.</p><p>Whether you’re moving from spreadsheets to Tally, QuickBooks, Zoho, NetSuite or SAP, the same rule applies: technology amplifies whatever process sits underneath it.</p><p>In July 1999, the American chocolate maker Hershey switched on a new set of enterprise systems — order management, supply chain planning and customer relationship software — in one large go-live.</p><p>The timing could hardly have been worse: it was the run-up to Halloween, one of the company’s busiest seasons. Orders entered the new system but did not flow properly to warehouses and trucks. Widely reported accounts describe more than US$100 million of orders that could not be filled on time, and a sharp fall in quarterly profit.</p><p>Hershey’s software came from respected vendors. <strong>What failed was the way it was introduced into the business.</strong></p><h3>The impressive demo</h3><p>Real-time dashboards. Automated reconciliations. Inventory across locations. One-click financial statements.</p><p>The business signs, migrates its data and goes live.</p><p>Six months later, the team still maintains spreadsheets alongside the new system. Reports do not match. Stock figures are wrong. The finance team works longer hours than before. Managers have stopped trusting the dashboards.</p><p>The software works. <strong>The business did not change.</strong></p><h3>A lesson from history: the first business computer</h3><p>In 1951, J. Lyons &amp; Co. — a British company famous for its tea shops — ran LEO I, widely regarded as the world’s first computer used for routine business work, such as valuations, payroll and stock.</p><p>What is less remembered is why Lyons succeeded. For years before, it had a dedicated team studying and improving its clerical processes. When the computer arrived, <strong>it automated work that had already been simplified and understood.</strong></p><p>Seventy-five years later, the order still matters: process first, technology second.</p><h3>Why finance technology projects disappoint</h3><p>Industry surveys have long reported that a large share of ERP and major IT projects run over budget, over time, or fall short of expected benefits. The exact percentages vary; the pattern does not. The causes are rarely technical:</p><p>1. <strong>Automating broken processes.</strong> Informal approvals, unrecorded stock movements and late invoicing get recorded faster — not fixed.</p><p>2. <strong>Dirty data migrated as-is.</strong> Duplicate customers, inconsistent product codes, unreconciled opening balances and a chaotic chart of accounts go straight into the new system. Garbage in; expensive garbage out.</p><p>3. <strong>Over-customisation.</strong> Insisting the software must work exactly like old habits raises cost, delays go-live and complicates upgrades.</p><p>4. <strong>No internal owner.</strong> The project is left to the vendor. Nobody inside has the time and authority to decide.</p><p>5. <strong>Weak training.</strong> A short demo before go-live, then users must learn while doing their normal work.</p><p>6. <strong>Endless parallel systems.</strong> Spreadsheets kept “just in case” mean the new system never becomes the single source of truth.</p><p>7. <strong>Choosing for the wrong reasons.</strong> Because a competitor uses it, because it is fashionable, or because of the biggest discount.</p><p>8. <strong>Big-bang go-live at the worst time.</strong> Hershey’s lesson: never go live during peak season.</p><p><strong>Software amplifies discipline. It also amplifies the lack of it.</strong></p><h3>The uncomfortable part</h3><p>Many owners buy software hoping it will impose the discipline they never imposed themselves.</p><p>It will not.</p><p>A system can enforce an approval workflow — but only if someone designs the workflow and management respects it. It can produce profitability reports — but only if costs are recorded correctly and consistently.</p><h3>The Process-First Ladder</h3><p>Successful implementations climb a ladder in order. Skip a rung, and the project wobbles.</p><h3>Rung 1: Map the processes</h3><p>Document how work actually happens: order-to-cash, procure-to-pay, record-to-report, inventory, payroll. Note every handoff and exception.</p><h3>Rung 2: Fix them on paper</h3><p>Remove unnecessary approvals, duplicate entries and unclear ownership <strong>before</strong> configuring anything.</p><h3>Rung 3: Define requirements</h3><p>Business needs, statutory needs (tax, e-invoicing, statutory reports, audit trail) and growth needs for the next five years.</p><h3>Rung 4: Choose on fit, not features</h3><p>Evaluate options by fit with your industry and size, total cost of ownership, local support, integration, data security and the vendor’s track record with similar businesses. Ask to speak to three existing customers.</p><h3>Rung 5: Clean the data</h3><p>Reconcile opening balances. Merge duplicates. Standardise codes. Redesign the chart of accounts for management and statutory needs.</p><h3>Rung 6: Configure, don’t customise</h3><p>Adopt standard processes unless there is a strong business reason not to.</p><h3>Rung 7: Test with real scenarios</h3><p>Returns, credit notes, partial deliveries, tax adjustments, month-end close — using real data from past months.</p><h3>Rung 8: Train by role</h3><p>Practical exercises and written guides for each user group — not one general demo.</p><h3>Rung 9: Go live at a quiet time — and end the old system</h3><p>Choose a low-season date. Set a firm end date for parallel spreadsheets.</p><h3>Rung 10: Review and strengthen</h3><p>Monitor adoption, fix data quality quickly, review user access and segregation of duties, and measure benefits against the original business case.</p><h3>A worked example (illustrative)</h3><p>A trading company with ₹40 crore turnover moves from a spreadsheet-heavy setup to a mid-market ERP.</p><p><strong>Plan A (common):</strong> buy, import last year’s ledgers as they are, go live in March (year-end), train users in one afternoon. <strong>Result:</strong> stock mismatches, delayed year-end close, an unhappy auditor and staff who return to spreadsheets.</p><p><strong>Plan B (Process-First Ladder):</strong> two months mapping and simplifying purchase and sales processes; a new chart of accounts; cleaned item and customer masters; go-live on 1 July (quiet season); parallel run for one month only; role-based training. <strong>Result:</strong> a slower start, but a monthly close in ten working days within six months and reliable branch-level margins for the first time.</p><p>The software was identical. <strong>The sequence made the difference.</strong></p><h3>How you’ll know it’s working</h3><p>A successful finance-technology rollout shows up in outcomes, not features:</p><p>· <strong>Close time</strong> falls, and fewer manual adjustments are needed.</p><p>· <strong>Spreadsheets outside the system</strong> decrease.</p><p>· <strong>Data quality</strong> — duplicate customers, vendors and items — falls sharply.</p><p>· <strong>User adoption</strong> — staff use the system as designed, without workarounds.</p><p>· <strong>Audit trail</strong> — every entry is traceable, meeting legal requirements.</p><h3>Where this is heading</h3><p>AI features are being built into ERP and accounting software — from automatic invoice capture to predictive cash forecasting and natural-language reporting. They will deliver real value only to businesses with clean processes and data. <strong>The next generation of software will reward process discipline even more than the last.</strong></p><h3>What to do on Monday, by role</h3><p>· <strong>Owner:</strong> before approving any new software, ask which three business problems it will solve and how success will be measured.</p><p>· <strong>CFO:</strong> map the order-to-cash and procure-to-pay processes as they really work today.</p><p>· <strong>IT head:</strong> clean customer, vendor and item master data before migration.</p><p>· <strong>Users:</strong> list every spreadsheet you use outside the current system and why.</p><h3>“But…” — the best counterargument</h3><p><strong>The objection:</strong> “Our current processes work fine for us. We just need a better tool. Spending months on process mapping is consultant talk.”</p><p><strong>The answer:</strong> If processes truly work, mapping them takes days and proves it. If they do not, mapping reveals it before you pay to automate the problem. For most SMEs, Rungs 1 and 2 take a few weeks, not months. Compare that with the cost of a failed go-live — lost orders, delayed audits and staff who stop trusting the system. <strong>A few weeks of thinking is the cheapest insurance any software project can buy.</strong></p><h3>Questions people ask me</h3><p><strong>“Should a small business move from Tally to a bigger ERP?”</strong></p><p>Only when the business has outgrown the current system’s capability — multiple locations, complex inventory, manufacturing, or integration needs. Many small businesses get more value by using their existing software properly.</p><p><strong>“Cloud or on-premise?”</strong></p><p>Cloud systems reduce hardware and maintenance burdens and suit most SMEs, but check data security, backup, local support and exit terms.</p><p><strong>“How long should implementation take?”</strong></p><p>For a small business, a few weeks to a few months; for complex multi-site operations, much longer. Be wary of promises that ignore data cleaning and training.</p><p><strong>Where things stand (October 2026):</strong> In India, companies must use accounting software with an audit trail (edit log) for every transaction that cannot be disabled, and auditors report on compliance. GST e-invoicing and e-way bill integration apply above turnover thresholds, the DPDP Rules impose safeguards on personal data held in systems, and records must be retained for statutory periods. Build these requirements into Rung 3.</p><h3>What about AI?</h3><p>AI-enabled finance tools can read invoices, match transactions, suggest account codes, flag anomalies and draft reports. They can sharply reduce manual work.</p><p>But AI depends on structured, reliable data and clear processes even more than traditional software does. <strong>AI layered on chaotic data produces confident-sounding but unreliable answers.</strong></p><h3>Mistakes I see most often</h3><p>· Buying software before fixing processes.</p><p>· Migrating unreconciled balances.</p><p>· Letting the vendor own the project.</p><p>· Customising everything.</p><p>· Giving every user administrator access.</p><p>· Running parallel spreadsheets indefinitely.</p><p>· Going live in your busiest season.</p><h3>One-page checklist</h3><p>· Processes mapped before software selection</p><p>· Master data cleaned before migration</p><p>· Audit trail enabled and tested</p><h3>Try this in 30 minutes</h3><p>Before your next software decision, answer in writing:</p><p>1. What three problems will this solve — specifically?</p><p>2. Which processes must change first?</p><p>3. Who inside the business owns the project, and how many hours a week can they give it?</p><p>4. Is our data clean enough to migrate? Who will clean it?</p><p>5. How will we measure success after six months?</p><p>If you cannot answer all five, you are not ready to buy.</p><h3>The short version</h3><p>· Software projects fail mainly from automated broken processes, dirty data, over-customisation, no internal owner, weak training, endless parallel systems and badly timed go-lives — as Hershey’s 1999 experience showed.</p><p>· Climb the Process-First Ladder: map, fix, define, choose on fit, clean, configure, test, train, go live quietly, review.</p><p>· Build in compliance: audit trail requirements, e-invoicing, data protection and record retention.</p><p>· AI amplifies the value of good data and processes — and the damage of bad ones.</p><p>· A few weeks of process thinking is the cheapest insurance a software project can buy.</p><p><strong>Technology is an accelerator. Make sure you are pointed in the right direction before you press it.</strong></p><h3>Before you go</h3><p>Before your next software project, map the process, clean the data and agree the reports you need. Then choose the tool. In that order, a new system is an upgrade; in any other order, it’s an expensive experiment.</p><h3>Further reading</h3><p>· Contemporary reporting on Hershey Foods’ 1999 systems implementation (e.g., <em>The Wall Street Journal</em> and <em>Computerworld</em>, 1999).</p><p>· Georgina Ferry, <em>A Computer Called LEO: Lyons Tea Shops and the World’s First Office Computer</em> (Fourth Estate, 2003).</p><p>· Companies (Accounts) Rules, 2014 (India), as amended — audit trail requirement. mca.gov.in</p><p>· Thomas H. Davenport, <em>Mission Critical: Realizing the Promise of Enterprise Systems</em> (Harvard Business School Press, 2000).</p><p>◆ ◆ ◆</p><p><em>A version of this article first appeared on ComplianceCalendar.in.</em></p><p><em>A note on numbers for readers outside India: 1 lakh = 100,000 and 1 crore = 10 million (₹ = Indian rupees).</em></p><p><em>I’m Aashok Kumar Sharma, founder of Aashok F&amp;C Advisory — a finance, accounting and compliance practice working with founders, SMEs and finance teams in India and abroad. I write about the numbers, rules and decisions that quietly decide whether businesses last. Follow me here on Medium for the rest of this 100-part series.</em></p><p><em>This article is general information, not legal, tax or investment advice. Rules change; check current provisions or speak to a qualified professional before acting. Examples labelled illustrative are composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2f74a0bc567f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Banks Say No: What Lenders See in Your Accounts That You Don’t]]></title>
            <link>https://medium.com/@aashokfcadvisory/why-banks-say-no-what-lenders-see-in-your-accounts-that-you-dont-49a3a7541dc3?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/49a3a7541dc3</guid>
            <category><![CDATA[business]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[accounting]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[finance]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Sun, 15 Nov 2026 12:26:01 GMT</pubDate>
            <atom:updated>2026-11-15T12:26:01.401Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 27 OF 100 | MEDIUM EDITION | SMALL BUSINESS, BANKING, FINANCE, ENTREPRENEURSHIP, LENDING</strong></p><p>Why Banks Say No: What Lenders See in Your Accounts That You Don’t</p><p><em>Viable small businesses get declined every day. Usually the problem isn’t the business — it’s the story the numbers tell.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>I’ve sat on both sides of the table — preparing loan files for owners and seeing how lenders read them. The frustration is real: a business that’s clearly working gets turned down, and nobody explains why.</p><p>The reason is usually that lenders don’t see what the owner sees. They see records. If your books, bank statements and tax filings tell three different stories, no amount of explaining fixes it. That’s true in India, the UK and almost everywhere else.</p><p>In February 2018, Punjab National Bank disclosed a fraud involving unauthorised letters of undertaking issued in favour of companies linked to the jeweller Nirav Modi. The amount involved ran into more than ₹13,000 crore.</p><p>The case shook Indian banking. In the years that followed, lenders tightened controls, verification and documentation across the system.</p><p>If your loan application today feels more demanding than it would have a decade ago, part of the reason is that <strong>banks have learned, painfully, to trust documents only when documents agree with each other.</strong></p><h3>The frustration every owner knows</h3><p>“The business is doing well. Why did the bank reject us?”</p><p>Sales are growing. Customers are loyal. The owner works hard and has never defaulted. Yet the loan is declined — or approved for a fraction of the request, at a higher rate, with more collateral.</p><p>From the owner’s side, it feels unfair. From the lender’s side, the decision usually comes down to one sentence:</p><p><strong>“The documents do not support the story.”</strong></p><h3>A lesson from history: lend on evidence</h3><p>In 1873, the British journalist and economist Walter Bagehot wrote <em>Lombard Street</em>, a book still read by central bankers. One of its lasting ideas is that sound lending depends on <strong>good security and reliable information</strong> — a lender should lend freely only against what it can trust.</p><p>A century and a half later, the instruments have changed. The principle has not. <strong>Lenders lend to evidence, not enthusiasm.</strong></p><h3>How lenders actually think</h3><p>A lender’s first question is not “Is this business growing?” It is: <strong>“Will we get our money back, on time, even if things go wrong?”</strong></p><p>Bankers have long organised that question around the <strong>five Cs of credit</strong>:</p><p>· <strong>Character</strong> — the track record and integrity of the borrower and promoters.</p><p>· <strong>Capacity</strong> — the ability to repay from cash flow.</p><p>· <strong>Capital</strong> — how much the owners have invested and kept in the business.</p><p>· <strong>Collateral</strong> — security available if repayment fails.</p><p>· <strong>Conditions</strong> — the industry, economy and purpose of the loan.</p><p>Every document you submit is evidence for or against each C.</p><h3>The Triangle of Truth</h3><p>Here is the single most important idea in this article. Lenders now test your business against what I call <strong>the Triangle of Truth</strong> — three independent records that must tell the same story:</p><p>1. <strong>Your financial statements</strong> — turnover, profit, assets, liabilities.</p><p>2. <strong>Your bank statements</strong> — actual money in and out.</p><p>3. <strong>Your tax data</strong> — GST returns, income-tax returns, TDS records.</p><p>If the three corners agree, your story is believable. If they disagree — turnover in the accounts is far above GST turnover, or bank credits are far below declared sales, or business receipts flow through personal accounts — the lender sees inconsistency, not explanation.</p><p><strong>A loan application is not judged by how good your story sounds, but by how well your three records agree.</strong></p><h3>The ten most common reasons viable businesses are rejected</h3><p>1. <strong>Declared income does not support the loan.</strong> Owners who minimise reported profit to save tax discover the same figures limit their borrowing.</p><p>2. <strong>Bank statements do not match financial statements.</strong> Especially when business receipts flow through personal accounts.</p><p>3. <strong>GST returns tell a different story.</strong> Lenders increasingly analyse GST data directly.</p><p>4. <strong>Weak repayment capacity.</strong> A low debt service coverage ratio (DSCR) or thin interest cover.</p><p>5. <strong>Overdue statutory dues.</strong> Unpaid taxes and provident fund suggest cash stress — and create claims that may rank alongside or ahead of lenders.</p><p>6. <strong>Poor credit history.</strong> Late payments on loans, cards or guarantees given for others affect business and promoter credit reports.</p><p>7. <strong>Over-leverage.</strong> Existing borrowing already high relative to equity and cash flow.</p><p>8. <strong>Thin promoter contribution.</strong> Owners with little money at risk, or capital withdrawn as fast as it is introduced.</p><p>9. <strong>Late or weak financial statements.</strong> Unaudited, delayed or poorly prepared accounts suggest weak management.</p><p>10. <strong>Unclear purpose.</strong> A request without a credible use and repayment plan.</p><h3>The uncomfortable part</h3><p>Many owners prepare for a loan the week before applying. Lenders evaluate behaviour over years.</p><p><strong>Creditworthiness is not created in an application. It is built in every filing, every payment and every set of accounts long before you ask for money.</strong></p><h3>How to prepare for better credit</h3><h3>12–24 months before you need funding</h3><p>· Keep business and personal finances completely separate.</p><p>· Route all business receipts and payments through business accounts.</p><p>· File GST, TDS and income-tax returns on time — and make sure they reconcile with your books.</p><p>· Pay statutory dues and existing instalments on time.</p><p>· Check business and personal credit reports; correct errors.</p><p>· Retain profits to build net worth.</p><h3>3–6 months before applying</h3><p>· Prepare audited or professionally reviewed statements.</p><p>· Prepare projections linked to historical performance.</p><p>· Write a clear loan proposal: purpose, amount, repayment plan, security offered.</p><p>· For working capital, prepare receivables, inventory and payables data in the format lenders use (in India, often CMA data for larger facilities).</p><h3>When applying</h3><p>· Present one consistent story across the Triangle of Truth.</p><p>· Explain unusual items proactively — before the lender finds them.</p><p>· Explore government-supported MSME schemes and credit guarantees where eligible.</p><p>· Compare banks, NBFCs and digital lenders on terms, flexibility and covenants — not just rate.</p><h3>What lenders appreciate (but rarely say)</h3><p>· Owners who understand their own numbers.</p><p>· Monthly management information available quickly.</p><p>· Honest disclosure of problems with a credible plan.</p><p>· Diversified customers and stable margins.</p><p>· A history of keeping small promises.</p><h3>How you’ll know it’s working</h3><p>Bank-ready financials show up in faster, cheaper credit:</p><p>· <strong>Loan approvals</strong> take less time, with fewer queries.</p><p>· <strong>Interest rates and collateral requirements</strong> improve.</p><p>· <strong>Covenant compliance</strong> is comfortable, with headroom.</p><p>· <strong>Stock and receivable statements</strong> match the accounts.</p><p>· <strong>Banker meetings</strong> discuss growth rather than explanations.</p><h3>Where this is heading</h3><p>Banks and NBFCs increasingly use account-aggregator data, GST returns, bureau scores and AI models to decide credit. Many decisions will be partly automated. <strong>The story you tell your banker will matter less than the data your systems tell them — automatically.</strong> Clean, consistent and timely financial data will become the single most important factor in borrowing.</p><h3>What to do on Monday, by role</h3><p>· <strong>Owner:</strong> ask your banker what would improve your credit rating or limit.</p><p>· <strong>CFO:</strong> reconcile the stock and receivable statements submitted to the bank with your books.</p><p>· <strong>Accountant:</strong> ensure audited accounts are finalised early, not at the last date.</p><p>· <strong>Sales team:</strong> reduce overdue receivables — they are the first thing lenders discount.</p><h3>If you run a small business</h3><p>Consider an illustrative ₹25 crore trading business repeatedly refused a higher working-capital limit. A review shows that its stock statements to the bank do not match its books, profits fluctuate because of year-end adjustments and director loans move in and out without documentation.</p><p>The owner fixes all three in one year. <strong>At the next renewal, the bank increases the limit by 40% at a lower rate.</strong> Nothing about the business changed except what the bank could trust.</p><h3>“But…” — the best counterargument</h3><p><strong>The objection:</strong> “Banks are biased against small businesses. They want collateral for everything and ignore our real potential. The system is the problem, not our documents.”</p><p><strong>The answer:</strong> There is truth here. Small businesses do face a credit gap worldwide, and collateral requirements can be harsh. That is why governments created credit-guarantee schemes and why cash-flow-based lending using digital data is growing. But those very schemes still depend on consistent, verifiable records. <strong>The fastest way to beat the system’s bias is to give it nothing to doubt.</strong></p><h3>Questions people ask me</h3><p><strong>“Will more lenders help if one says no?”</strong></p><p>Applying to many lenders at once can generate multiple credit enquiries and signal distress. Understand why the first lender declined, fix the cause, then apply strategically.</p><p><strong>“Can I show higher profit just for the loan?”</strong></p><p>No. Inflating figures for a lender is misrepresentation, and the Triangle of Truth will expose it. Build real profit and declare it honestly.</p><p><strong>“What DSCR do banks look for?”</strong></p><p>It varies by lender and loan type. Many lenders look for a DSCR comfortably above 1 — often around 1.25 or more for term loans — but check with your bank.</p><p><strong>Where things stand (October 2026):</strong> Indian MSMEs can access collateral-free loans backed by the Credit Guarantee Fund Trust for Micro and Small Enterprises (CGTMSE), and lenders increasingly use GST data, Account Aggregator-based bank data and the RBI’s Unified Lending Interface for faster, cash-flow-based decisions. Credit bureaus provide MSME ranks and commercial reports. Eligibility and terms change — check current conditions.</p><h3>Mistakes I see most often</h3><p>· Inflating stock or receivables in statements to banks.</p><p>· Projections disconnected from historical reality.</p><p>· Diverting loan money to purposes other than those stated.</p><p>· Ignoring covenants after approval.</p><p>· Treating the bank as an adversary instead of a partner.</p><h3>One-page checklist</h3><p>· Stock statements match the books</p><p>· Audited accounts ready early</p><p>· Covenant headroom checked every quarter</p><h3>Try this in 30 minutes: test your triangle</h3><p>For last financial year, write three numbers side by side:</p><p>1. Turnover in your financial statements.</p><p>2. Turnover in your GST returns.</p><p>3. Total business credits in your bank statements (excluding loans and capital).</p><p>If they differ by more than a few percent, write down why. <strong>Any difference you cannot explain in one sentence is a difference a lender will question.</strong></p><h3>The short version</h3><p>· Lenders ask one question: will we be repaid on time, even if things go wrong?</p><p>· The five Cs — character, capacity, capital, collateral, conditions — still frame every decision.</p><p>· Pass the Triangle of Truth: financial statements, bank statements and tax data must tell one story.</p><p>· Common rejections: under-declared income, mismatched records, weak DSCR, overdue dues, poor credit history, over-leverage, thin promoter capital, late accounts and unclear purpose.</p><p>· Creditworthiness is built over years — start 12–24 months before you need money.</p><p><strong>Banks lend to evidence, not enthusiasm.</strong></p><h3>Before you go</h3><p>Bankability isn’t built in the week you apply. It’s built month by month, by keeping your books, bank and tax records telling the same story. Start that habit two years before you need the loan.</p><h3>Further reading</h3><p>· Punjab National Bank, stock exchange disclosures (February 2018), and subsequent public reporting on the fraud.</p><p>· Walter Bagehot, <em>Lombard Street: A Description of the Money Market</em> (1873), public domain.</p><p>· Reserve Bank of India — Master Directions on lending to MSMEs; Unified Lending Interface. rbi.org.in</p><p>· Credit Guarantee Fund Trust for Micro and Small Enterprises. cgtmse.in</p><p>· World Bank, <em>SME Finance</em> resources on the global small-business credit gap. worldbank.org</p><p>◆ ◆ ◆</p><p><em>A version of this article first appeared on ComplianceCalendar.in.</em></p><p><em>A note on numbers for readers outside India: 1 lakh = 100,000 and 1 crore = 10 million (₹ = Indian rupees).</em></p><p><em>I’m Aashok Kumar Sharma, founder of Aashok F&amp;C Advisory — a finance, accounting and compliance practice working with founders, SMEs and finance teams in India and abroad. I write about the numbers, rules and decisions that quietly decide whether businesses last. Follow me here on Medium for the rest of this 100-part series.</em></p><p><em>This article is general information, not legal, tax or investment advice. Rules change; check current provisions or speak to a qualified professional before acting. Examples labelled illustrative are composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=49a3a7541dc3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When Your Supplier Slips, You Pay: Protecting Your GST Input Tax Credit]]></title>
            <link>https://medium.com/@aashokfcadvisory/when-your-supplier-slips-you-pay-protecting-your-gst-input-tax-credit-3d0a18efc45e?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/3d0a18efc45e</guid>
            <category><![CDATA[business]]></category>
            <category><![CDATA[accounting]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Sat, 14 Nov 2026 13:55:59 GMT</pubDate>
            <atom:updated>2026-11-14T13:55:59.523Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 31 | TAXATION — INDIA &amp; GLOBAL</strong></p><p><em>You paid the supplier in full, GST included, and you hold the invoice. You can still lose the credit.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>In 2017, the Delhi High Court heard <em>On Quest Merchandising India Pvt Ltd v. Government of NCT of Delhi</em>. The question at its heart was straightforward and painful: should a buyer who had genuinely bought goods and paid tax to the seller forfeit its tax credit because the seller never handed that tax over to the government?</p><p>Under the Delhi VAT law of the time, the court held that the provision, as applied to genuine buyers acting in good faith, had to be read down — such purchasers shouldn’t be penalised for a seller’s default.</p><p>That ruling came under the old VAT regime. The same question now lies at the centre of GST — and thousands of Indian companies confront it again each month.</p><h3>“I’ve paid. Why am I paying twice?”</h3><p>Ask GST-registered owners what annoys them most, and many will say:</p><p>“I paid my supplier the full amount, GST and all. I’ve got the tax invoice. So why is my input tax credit being refused?”</p><p>It feels profoundly unfair.</p><p>Yet in India’s GST design, input tax credit (ITC) isn’t just a right that springs from holding an invoice. <strong>It’s a conditional credit — and several of the conditions depend on what your supplier does.</strong></p><p><strong>Your credit is only as dependable as your least reliable supplier.</strong></p><h3>What history teaches: a chain is only as strong as its weakest link</h3><p>Value-added taxes were developed in Europe in the mid-twentieth century — France brought in a broad VAT in 1954 — around an elegant idea: every business in a chain pays tax only on the value it adds, by deducting the tax it paid on inputs. Today over 170 countries operate a VAT or GST.</p><p>That self-checking chain of invoices is the design’s great strength — and its weakness. When one link breaks, every later link comes under suspicion. <strong>Tax authorities everywhere use the chain to catch fraud. Honest firms have to use the very same chain to protect themselves.</strong></p><h3>How GST credit is meant to work</h3><p>To stop fake invoices and credit fraud, the law ties a buyer’s credit to the supplier’s reporting and payment. In broad terms, a registered buyer can claim ITC only when, among other conditions:</p><ul><li>It holds a valid tax invoice or another prescribed document.</li><li>It has actually received the goods or services.</li><li><strong>The supplier has reported the invoice in its outward-supply statement</strong>, and the invoice reveals up in the buyer’s auto-generated ITC statement (usually checked as GSTR-2B).</li><li><strong>The tax charged has truly been paid to the government</strong> by the supplier.</li><li>The buyer has filed its own return.</li><li>The credit is claimed within the time limit.</li><li>The credit isn’t caught by specific restrictions (for instance, certain motor vehicles, food and beverages, personal consumption and works contracts for immovable property, subject to exceptions).</li></ul><p>And one condition many firms miss: if the buyer doesn’t pay the supplier the value of the supply plus tax within <strong>180 days</strong> of the invoice date, the credit generally has to be reversed, with interest — though it can be reclaimed once payment is made.</p><h3>Where firms lose credit</h3><p>1. <strong>The supplier never reported the invoice</strong> — or reported it under the wrong GSTIN.</p><p>2. <strong>The supplier reported it but didn’t pay</strong> the tax collected.</p><p>3. <strong>Wrong GSTIN, wrong place of supply or wrong tax type</strong> — IGST charged where CGST/SGST applied, or the other way round.</p><p>4. <strong>Claims made too late</strong> — credits not taken within the statutory window can be lost for good.</p><p>5. <strong>Blocked credits claimed in error</strong> — bringing reversal, interest and possibly penalties.</p><p>6. <strong>Fake or non-existent suppliers</strong> — invoices from entities that never supply anything, sometimes part of organised fraud rings. Authorities have repeatedly run special drives against bogus registrations.</p><p>7. <strong>Suppliers left unpaid beyond 180 days</strong> — credit claimed but never reversed.</p><h3>Quick recap</h3><p><strong>The short version:</strong> GST credit is earned not when an invoice arrives, but once the whole chain — supplier reporting, supplier payment, your payment and your timing — has done its job.</p><h3>The hard reality</h3><p>Plenty of businesses pick suppliers on price and nothing else.</p><p>The cheapest supplier is sometimes cheapest precisely because it isn’t fully compliant — skipping returns, not paying tax or trading informally.</p><p><strong>A supplier who saves you 3% on price but costs you 18% in refused credit isn’t cheap at all.</strong></p><h3>The Credit Chain Guard</h3><p>Guard your credit with five protections.</p><h3>Guard 1: Reconcile monthly — before you file</h3><p>Compare your purchase register with the auto-generated ITC statement:</p><ul><li>In your books but missing from the statement → chase the supplier before claiming.</li><li>In the statement but missing from your books → check whether the goods arrived, or whether someone is misusing your GSTIN.</li><li>Differences in value, rate or tax type → get the supplier to correct them.</li></ul><h3>Guard 2: Check suppliers before buying</h3><ul><li>Verify GSTIN status, legal name and registration type before onboarding.</li><li>Look at filing history where you can.</li><li>Be wary of new suppliers quoting unusually low prices, especially in sectors notorious for fake invoicing.</li><li>Keep proof that supply was genuine: purchase orders, delivery challans, e-way bills, transport records and payment through banking channels.</li></ul><h3>Guard 3: Write compliance into contracts</h3><ul><li>Require suppliers to report invoices correctly and on time and to pay over the tax they collect.</li><li>Include a right to recover any credit lost through their non-compliance.</li><li>For repeat offenders, consider withholding the GST portion until invoices appear correctly.</li></ul><h3>Guard 4: Watch the 180-day clock</h3><p>Track unpaid supplier balances by age and reverse credit when the rules require it.</p><h3>Guard 5: Score suppliers on compliance</h3><p>Rate suppliers on GST compliance as well as on quality, price and delivery, and lean less on the weak ones.</p><h3>Case in point</h3><p>A manufacturer buys ₹10 lakh of materials plus ₹1.8 lakh GST from a new supplier offering prices 4% lower.</p><ul><li><strong>Price saving:</strong> roughly ₹40,000.</li><li><strong>Three months later:</strong> the supplier hasn’t filed returns, and the invoices never appear in the manufacturer’s ITC statement.</li><li><strong>Exposure:</strong> ₹1.8 lakh of credit at risk, interest if it was already claimed, management time and possibly a notice.</li></ul><p>The “cheaper” supplier cost more than four times what it saved. <strong>A supplier’s true price includes how reliable its tax compliance is.</strong></p><h3>When credit is refused anyway</h3><p>Even with every precaution, disputes happen. Courts have sometimes protected genuine buyers able to prove real transactions and payment — as the Delhi High Court did in its 2017 VAT ruling — but results depend on the facts, the statute and developing case law.</p><p>What strengthens your hand:</p><ul><li>Full documentation of genuine receipt and payment.</li><li>Evidence that you exercised reasonable diligence on the supplier.</li><li>Prompt, professional replies to notices.</li></ul><h3>Signs of progress</h3><p>ITC discipline reveals up in your reconciliations:</p><ul><li><strong>Monthly matching</strong> — purchase records agree with supplier-reported data before returns go in.</li><li><strong>Unmatched credit</strong> shrinks as a share of total credit.</li><li><strong>Supplier compliance scores</strong> improve, and non-filers are spotted early.</li><li><strong>Payments are tied</strong> to compliance for high-risk vendors.</li><li><strong>Notices about credit mismatches</strong> decline year by year.</li></ul><h3>The decade ahead</h3><p>GST is moving towards fully automated invoice-level matching, with e-invoicing thresholds falling over time and analytics flagging risky supply chains. <strong>Your credit will depend more and more on the compliance of everyone in your supply chain, not only your own.</strong> Choosing reliable suppliers will become as much a tax decision as a commercial one.</p><h3>Next steps for each role</h3><ul><li><strong>CFO:</strong> reconcile last quarter’s credit against supplier-reported data.</li><li><strong>Purchasing team:</strong> make GST compliance history part of supplier selection.</li><li><strong>Accounts payable:</strong> hold back part of payment for suppliers with repeated mismatches.</li><li><strong>Owner:</strong> ask how much credit is at risk right now.</li></ul><h3>Frequently asked questions</h3><p><strong>“Can I take credit if the invoice reveals in GSTR-2A but not GSTR-2B?”</strong></p><p>Credit is generally linked to the auto-generated statement for the relevant period. Track timing differences and claim when the invoice appears, within the time limits.</p><p><strong>“What if my supplier’s registration is cancelled later on?”</strong></p><p>Keep documents proving the supply and payment were genuine, and get professional advice swiftly if notices arrive.</p><p><strong>“Do small suppliers truly matter?”</strong></p><p>Yes. Fraud networks frequently use small, newly registered entities. Diligence should be proportionate — but never zero.</p><p><strong>Where things stand (October 2026):</strong> GST credit conditions under section 16 of the CGST Act (including the supplier-reporting condition), the 180-day payment rule, time limits for claiming credit and blocked credits under section 17(5) continue to apply. GST rates were rationalised from 22 September 2025, and return systems increasingly lock auto-populated data. Confirm current rules on the GST portal or with your adviser.</p><h3>The toughest pushback — and a reply</h3><p><strong>The pushback:</strong> “Collecting tax from my supplier is the government’s job. Why should honest buyers have to police their suppliers?”</p><p><strong>The reply:</strong> In principle the objection carries real weight — courts and policymakers have argued over it for many years. In real life, the system currently puts part of the risk on buyers, and waiting for the law to change won’t protect this month’s credit. Supplier diligence pays off beyond tax, too: compliant suppliers tend to be more dependable on quality, paperwork and continuity. <strong>You can disagree with the rule and still shield yourself from it.</strong></p><h3>Common mistakes to avoid</h3><ul><li>Claiming credit on invoices missing from your ITC statement without a clear basis.</li><li>Ignoring system alerts about mismatches.</li><li>Paying suppliers in cash or through third parties.</li><li>Leaving ITC reconciliation until year-end.</li><li>Taking on suppliers without basic checks.</li></ul><h3>Quick checklist</h3><p>☐ Monthly credit reconciliation before filing</p><p>☐ Supplier compliance reviewed each quarter</p><h3>Try this today (30 minutes)</h3><p>1. What share of last month’s purchase ITC matched your auto-generated statement?</p><p>2. Name the five suppliers with the largest mismatches.</p><p>3. List supplier invoices unpaid for over 150 days — the 180-day deadline is near.</p><p>4. Check whether your team has been trained on blocked credits.</p><p>Send each of those five suppliers a courteous compliance request this week.</p><h3>Key takeaways</h3><ul><li>GST input tax credit is conditional — and several conditions hinge on your supplier’s reporting and tax payment.</li><li>Credit is lost through unreported invoices, unpaid supplier tax, wrong GSTIN or tax type, late claims, blocked credits, fake suppliers and bills left unpaid beyond 180 days.</li><li>Apply the Credit Chain Guard: monthly reconciliation, supplier checks, compliance clauses, a 180-day tracker and compliance scoring.</li><li>Cheap, non-compliant suppliers frequently turn out to be the dearest.</li><li>Courts have sometimes shielded genuine buyers, but documents and diligence are your strongest defence.</li></ul><p><strong>Input tax credit isn’t money in your account until the entire chain has done its job.</strong></p><p><em>Coming next in this series: Why the GST Notice Came — and How to Answer It Calmly</em></p><h3>Sources and references</h3><ul><li><em>On Quest Merchandising India Pvt Ltd v. Government of NCT of Delhi</em>, Delhi High Court (2017).</li><li>Central Goods and Services Tax Act, 2017, sections 16 and 17. cbic-gst.gov.in</li><li>OECD, <em>Consumption Tax Trends</em> (latest edition) — on VAT/GST systems worldwide. oecd.org</li><li>GST Network — taxpayer advisories on GSTR-2B and ITC. gst.gov.in</li></ul><p><strong>About the author</strong></p><p>Aashok Kumar Sharma founded Aashok F&amp;C Advisory, a Guwahati-based practice offering finance, accounting, compliance and advisory support to SMEs, MSMEs, founders, chartered accountants and overseas clients. Over more than 16 years he has worked across fintech, technology, manufacturing and professional services — including financial reporting, reconciliations, audit readiness and internal controls for a US-based fintech.</p><p>Found this helpful? Keep it for reference and pass it on to a founder, finance team or business community that could use it. — Aashok F&amp;C Advisory</p><p><em>Disclaimer: This article is intended for general information and education only. It is not legal, tax, investment or accounting advice for any specific situation. Laws, rates and thresholds change; verify current provisions with official sources or a qualified professional before acting. Cases cited are drawn from public regulatory and court records; scenarios described as examples are illustrative composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3d0a18efc45e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Tax Notices Aren’t Random: What Triggers Scrutiny and How to Respond Calmly]]></title>
            <link>https://medium.com/@aashokfcadvisory/tax-notices-arent-random-what-triggers-scrutiny-and-how-to-respond-calmly-51124e68e2b4?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/51124e68e2b4</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[accounting]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[finance]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Sat, 14 Nov 2026 13:55:44 GMT</pubDate>
            <atom:updated>2026-11-14T13:55:44.615Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 32 OF 100 | MEDIUM EDITION | TAX, GST, INDIA, SMALL BUSINESS, COMPLIANCE</strong></p><p><em>Tax systems now run on data matching. A notice usually means two numbers disagree — and the answer is a reconciliation, not panic.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>The first reaction to a tax notice is almost always fear. I’ve watched clients freeze for weeks, which is usually the worst thing to do.</p><p>This piece uses India’s GST as the example, but tax authorities everywhere — HMRC, the IRS, the ATO — increasingly select cases by matching data. Understanding why you got a notice is half of answering it well.</p><p>In 1086, William the Conqueror ordered a survey of almost every landholding in England — who held it, how much it was worth, how many ploughs, mills and animals it had. The result, the Domesday Book, was used to understand and assess the kingdom’s resources and dues.</p><p>Nine and a half centuries later, India’s GST system does something similar every month — not by sending clerks on horseback, but by comparing millions of invoices, returns, e-invoices, e-way bills and payments automatically.</p><p><strong>When the numbers disagree, a notice is not bad luck. It is the system doing what it was built to do.</strong></p><h3>“Why us? We pay our taxes.”</h3><p>The email arrives. A notice on the GST portal. Panic follows.</p><p>In most cases, a notice is not the result of someone deciding to target a business. It is the result of data being compared — and differences being found.</p><p>GST generates enormous volumes of structured information. Authorities use analytics to compare it across returns, periods and businesses. <strong>When your numbers disagree with each other, or with your suppliers’ and customers’ numbers, questions follow.</strong></p><h3>The most common triggers</h3><p>1. <strong>Outward supply mismatches.</strong> Sales in the invoice-level return (GSTR-1) do not match tax paid in the summary return (GSTR-3B).</p><p>2. <strong>Excess input tax credit.</strong> Credit claimed exceeds what appears in the auto-generated ITC statement based on suppliers’ filings — one of the most frequent triggers.</p><p>3. <strong>E-way bills versus returns.</strong> Goods moved under e-way bills but not reflected in returns, or values that differ significantly.</p><p>4. <strong>E-invoices versus returns.</strong> For businesses required to e-invoice, differences between registered e-invoices and reported supplies.</p><p>5. <strong>Annual return and reconciliation differences.</strong> Monthly returns, annual return and audited financial statements that disagree.</p><p>6. <strong>Turnover mismatches across tax systems.</strong> GST turnover that differs significantly from income-tax turnover.</p><p>7. <strong>Suppliers later flagged as non-genuine.</strong> Buyers who claimed credit on their invoices may receive notices.</p><p>8. <strong>Refund claims.</strong> Examined closely for eligibility and documentation.</p><p>9. <strong>Classification or rate disputes.</strong> Especially after rate changes — such as the major rationalisation effective September 2025.</p><p>10. <strong>Non-filing or late filing.</strong> Late fees, interest and, in some cases, cancellation of registration.</p><h3>A case from the record: the fake-registration drives</h3><p>In recent years, Indian GST authorities have run nationwide special drives to detect fake and non-existent registrations used to generate invoices without real supplies. Such drives have identified large numbers of suspect registrations — and buyers who had claimed credit from them received notices, even when they insisted they had acted in good faith.</p><p>The lesson for honest businesses is uncomfortable: <strong>your exposure depends partly on data you do not control.</strong> That is exactly why your own reconciliation matters so much.</p><h3>Types of communication you may receive</h3><p>Without listing every form, GST communications broadly fall into:</p><ul><li><strong>System intimations of mismatches</strong> — often asking you to pay or explain.</li><li><strong>Scrutiny of returns</strong> — an officer asks you to explain discrepancies.</li><li><strong>Departmental audit.</strong></li><li><strong>Show cause notices</strong> proposing tax, interest and penalties.</li><li><strong>Registration-related notices</strong> — for example, proposed cancellation.</li></ul><p>For periods from FY 2024–25 onward, demand proceedings fall under a unified provision (section 74A of the CGST Act), with different penalty consequences depending on whether fraud or suppression is alleged. Time limits and penalties depend on the provision and period involved.</p><p><strong>A notice is not a judgement. It is a question. Answer it well, and many end without a demand.</strong></p><h3>The CALM Response</h3><p>When a notice arrives, follow <strong>CALM</strong>:</p><h4>C — Check the notice and the clock</h4><ul><li>Note the period, the amount, the issue and the provision cited.</li><li>Confirm it is validly issued (for example, with a document identification number where required).</li><li>Write down the reply deadline. Missing it can lead to orders passed without your side being heard.</li></ul><h4>A — Analyse and reconcile</h4><p>Pull your books, returns, e-invoices, e-way bills and supplier statements for the period. Find the exact source of the difference.</p><h4>L — Label the difference</h4><ul><li><strong>Genuine error by you</strong> — consider paying with interest promptly; this may reduce penalties.</li><li><strong>Timing difference</strong> — show where the transaction was reported in another period.</li><li><strong>Supplier’s error</strong> — gather evidence and work with the supplier to correct it.</li><li><strong>Disagreement on law</strong> — prepare a reasoned legal response.</li></ul><h4>M — Make a written, evidence-based reply</h4><p>A clear, organised reply with reconciliations, documents and legal reasoning is far stronger than a general denial. Attend any hearing prepared to walk through your reconciliation. If an order goes against you, appeals are available within prescribed time limits, usually requiring a pre-deposit.</p><h3>A worked example (illustrative)</h3><p>A distributor receives an intimation that its ITC for one month exceeds its auto-generated statement by ₹6.4 lakh.</p><p><strong>Analysis shows:</strong></p><ul><li>₹3.1 lakh — invoices reported by suppliers in the following month (timing).</li><li>₹2.0 lakh — one supplier reported under the wrong GSTIN (supplier error; corrected in next return).</li><li>₹1.3 lakh — credit on a staff welfare event, which falls under blocked credits (genuine error).</li></ul><p><strong>Reply:</strong> a reconciliation table, supplier correspondence and evidence of the next month’s reporting for ₹5.1 lakh; voluntary reversal with interest for ₹1.3 lakh. <strong>Outcome:</strong> a manageable payment instead of a demand on the full amount plus penalty.</p><h3>Prevention is cheaper than defence</h3><ul><li><strong>Monthly three-way reconciliation:</strong> books, invoice-level returns and summary returns.</li><li><strong>Monthly ITC reconciliation</strong> with auto-generated statements.</li><li><strong>E-way bill and e-invoice reconciliation</strong> with sales.</li><li><strong>Annual reconciliation</strong> of GST turnover with financial statements and income-tax returns.</li><li><strong>Classification review</strong> whenever rates change or new products launch.</li><li><strong>Supplier compliance checks</strong> for major vendors.</li><li><strong>Respond to intimations promptly</strong> — before they become formal notices.</li></ul><h3>The uncomfortable part</h3><p>Many businesses treat GST as a monthly <strong>filing</strong> exercise rather than a monthly <strong>reconciliation</strong> exercise. Returns are filed to meet deadlines. Differences are “to be adjusted later.” Later becomes the annual return. Then a notice.</p><p><strong>In a data-driven system, every unreconciled difference is a future question.</strong></p><h3>How you’ll know it’s working</h3><p>A calm, prepared response system shows up in outcomes:</p><ul><li><strong>Response deadlines</strong> are always met.</li><li><strong>Notices are resolved</strong> at the first stage more often.</li><li><strong>Demands are reduced</strong> because replies are supported with documents.</li><li><strong>Repeat notices</strong> on the same issue fall, because root causes are fixed.</li><li><strong>Records are ready</strong> — reconciliations and documents can be produced within days.</li></ul><h3>Where this is heading</h3><p>Tax notices are increasingly generated by data analytics comparing GST returns, e-invoices, e-way bills, income-tax data and bank information. <strong>The trigger for scrutiny will more often be a data mismatch than an officer’s choice.</strong> Businesses that reconcile their own data before filing will receive fewer notices — and answer the rest faster.</p><h3>What to do on Monday, by role</h3><ul><li><strong>Finance head:</strong> create a notice register with dates, issues, owners and status.</li><li><strong>Accountant:</strong> reconcile GST returns with books and e-invoices for the last financial year.</li><li><strong>Owner:</strong> ensure that no notice is ignored because it arrived by email or on a portal.</li><li><strong>Adviser:</strong> review responses before submission for significant notices.</li></ul><h3>If you run a small business</h3><p>Consider an illustrative ₹8 crore trader who receives a notice about a mismatch between sales in GST returns and e-way bills. His first instinct is to call a contact in the department.</p><p>Instead, his accountant prepares a reconciliation showing that the difference arises from cancelled e-way bills and a sales return. The reply, supported by documents, is filed within a week. <strong>The notice is closed without demand</strong> — and the trader starts reconciling every month to avoid the next one.</p><h3>“But…” — the best counterargument</h3><p><strong>The objection:</strong> “Many notices are mechanical and wrong. The system flags timing differences as tax evasion. Why should we spend so much effort on what is the department’s mistake?”</p><p><strong>The answer:</strong> Many intimations are indeed mechanical, and some are wrong. That is precisely why monthly reconciliation pays: a business that already has its reconciliation ready can answer a wrong notice in hours, with evidence. A business that must reconstruct the period from scratch spends weeks — and may still lose. <strong>Preparation does not mean accepting every notice. It means being able to defeat the wrong ones quickly.</strong></p><h3>Questions people ask me</h3><p><strong>“Can I ignore a system intimation if I believe it is wrong?”</strong></p><p>No. Reply with your explanation and evidence. Silence is often treated as acceptance.</p><p><strong>“Should I pay disputed amounts to avoid penalties?”</strong></p><p>Sometimes paying genuine errors early reduces penalties. For disputed amounts, take advice — payment may affect your position.</p><p><strong>“Who should handle notices?”</strong></p><p>A designated person who monitors the portal regularly, supported by your tax adviser.</p><p><strong>Where things stand (October 2026):</strong> Key provisions include section 61 (scrutiny of returns), section 73/74 for earlier periods and section 74A for periods from FY 2024–25, plus appeal provisions requiring pre-deposit. GST rates were rationalised from 22 September 2025, and return systems increasingly lock auto-populated values. Confirm current procedures on the GST portal and with your adviser.</p><h3>Mistakes I see most often</h3><ul><li>Filing returns without reconciling them.</li><li>Using GSTR-3B as a balancing figure.</li><li>Replying without evidence.</li><li>Missing reply deadlines.</li><li>Paying demands without checking whether they are correct.</li><li>Letting the same mismatch repeat every month.</li></ul><h3>Try this in 30 minutes</h3><p>For the last three months:</p><p>1. Compare GSTR-1 sales with GSTR-3B sales. Any difference?</p><p>2. Compare GSTR-3B ITC with your auto-generated ITC statement. Any difference?</p><p>3. Compare e-way bill values with reported supplies.</p><p>4. Check: who logs into the GST portal to look for notices, and how often?</p><p>Write one sentence explaining every difference. Any difference you cannot explain is your next notice — unless you fix it first.</p><h3>The short version</h3><ul><li>GST notices usually come from data mismatches, not random selection — the modern equivalent of a Domesday survey, run every month.</li><li>Common triggers: GSTR-1 vs GSTR-3B differences, excess ITC, e-way bill and e-invoice gaps, annual return differences, fake suppliers, refunds, classification and late filing.</li><li>Respond with CALM: Check the notice and deadline, Analyse and reconcile, Label the difference, Make a written evidence-based reply.</li><li>Prevent notices with monthly three-way and ITC reconciliations and annual reconciliation with financial statements.</li><li>Preparation lets you correct genuine errors cheaply and defeat wrong notices quickly.</li></ul><p><strong>Every unreconciled difference today is a potential notice tomorrow.</strong></p><h3>Before you go</h3><p>Notices are part of modern tax life. Keep monthly reconciliations, respond on time, answer only what’s asked and support every point with documents. Calm and complete beats clever and late.</p><h3>Further reading</h3><ul><li>The National Archives (UK) — Domesday Book (1086). nationalarchives.gov.uk</li><li>Central Goods and Services Tax Act, 2017, sections 61, 73, 74, 74A and 107. cbic-gst.gov.in</li><li>Central Board of Indirect Taxes and Customs — press releases on special drives against fake registrations. cbic.gov.in</li><li>GST Network — advisories on return filing and auto-populated data. gst.gov.in</li></ul><p>◆ ◆ ◆</p><p><em>A version of this article first appeared on ComplianceCalendar.in.</em></p><p><em>A note on numbers for readers outside India: 1 lakh = 100,000 and 1 crore = 10 million (₹ = Indian rupees).</em></p><p><em>I’m Aashok Kumar Sharma, founder of Aashok F&amp;C Advisory — a finance, accounting and compliance practice working with founders, SMEs and finance teams in India and abroad. I write about the numbers, rules and decisions that quietly decide whether businesses last. Follow me here on Medium for the rest of this 100-part series.</em></p><p><em>This article is general information, not legal, tax or investment advice. Rules change; check current provisions or speak to a qualified professional before acting. Examples labelled illustrative are composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=51124e68e2b4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Free Trade Agreements: The Duty Savings Most Exporters Never Claim]]></title>
            <link>https://medium.com/@aashokfcadvisory/free-trade-agreements-the-duty-savings-most-exporters-never-claim-87bfffdfc0b4?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/87bfffdfc0b4</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[accounting]]></category>
            <category><![CDATA[business]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Sat, 14 Nov 2026 13:55:30 GMT</pubDate>
            <atom:updated>2026-11-14T13:55:30.929Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 68 OF 100 | MEDIUM EDITION | INTERNATIONAL TRADE, EXPORT, INDIA, TAX, SUPPLY CHAIN</strong></p><p><em>India signed more trade agreements in four years than in the previous decade. Many exporters still pay full duty.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>Governments announce trade agreements with great fanfare. Then many exporters quietly keep paying the full duty, because claiming the preference looked too complicated.</p><p>The examples here are Indian agreements, but rules of origin work similarly in every FTA, from the EU’s to the USMCA. The savings are real; so is the paperwork.</p><p>Between May 2022 and October 2026, India brought into force or concluded more trade agreements than in the previous decade: with the <strong>UAE</strong> (in force May 2022), <strong>Australia</strong> (December 2022), the <strong>EFTA</strong> countries — Switzerland, Norway, Iceland and Liechtenstein (October 2025), <strong>Oman</strong> (June 2026), the <strong>United Kingdom</strong> (July 2026) and <strong>New Zealand</strong> (signed April 2026), along with the <strong>European Union</strong> agreement concluded in January 2026.</p><p>Each agreement removes or reduces tariffs on thousands of products. Each was negotiated over years.</p><p>Yet studies of Indian businesses have repeatedly found that only a <strong>minority of eligible exports</strong> actually claim FTA preferences — estimates have often been around a quarter.</p><p><strong>The duty saving exists on paper. Most exporters never collect it.</strong></p><h3>Why savings are left on the table</h3><p>Ask exporters why they do not use FTAs, and the answers are familiar:</p><ul><li>“We didn’t know our product was covered.”</li><li>“The rules of origin are too complex.”</li><li>“Getting the certificate takes too long.”</li><li>“Our buyer never asked.”</li><li>“We tried once and customs rejected it.”</li></ul><p>Each reason is understandable. Together, they mean many exporters compete with one hand tied — while competitors from the same or partner countries pay lower duties.</p><h3>A lesson from history: the treaty that started a network</h3><p>In 1860, Britain and France signed the <strong>Cobden-Chevalier Treaty</strong>, cutting tariffs on each other’s goods. It included a <strong>most-favoured-nation clause</strong>: any better terms later given to a third country would also apply to the partner. Within a few years, a web of similar treaties spread across Europe, and trade grew rapidly.</p><p>The treaty was negotiated by politicians. <strong>Its benefits were collected by merchants who learned the new rules first.</strong> The same is true today: agreements open the door, but only prepared businesses walk through it.</p><h3>How FTAs work</h3><p>An FTA lowers or removes customs duties on goods that <strong>originate</strong> in the partner countries. To claim the lower duty, the importer must show that the goods meet the <strong>rules of origin</strong> in that agreement. Common tests include:</p><ul><li><strong>Wholly obtained</strong> — for example, agricultural or mineral products produced entirely in one country.</li><li><strong>Change in tariff classification</strong> — the product’s HS code changes at a specified level because of processing in the exporting country.</li><li><strong>Regional value content</strong> — a minimum percentage of value is added in the exporting country or region.</li><li><strong>Specific processes</strong> — certain manufacturing operations must take place locally.</li></ul><p>Minimal operations such as packing, labelling or simple assembly usually do not confer origin.</p><p><strong>Proof</strong> comes through a certificate of origin or, under some agreements, a self-declaration by an approved exporter — supported by records.</p><p><strong>An FTA is not a discount. It is a right you must prove.</strong></p><h3>India’s verification rules</h3><p>India’s <strong>Customs (Administration of Rules of Origin under Trade Agreements) Rules, 2020</strong> — known as <strong>CAROTAR</strong> — require <strong>importers</strong> claiming FTA benefits in India to have origin information and to provide it when asked. Customs can verify claims and deny benefits if information is inadequate.</p><p>For exporters, partner countries apply similar checks. Wrong claims can lead to denied preferences, duty demands with interest and penalties, and damage to the buyer relationship.</p><h3>The uncomfortable part</h3><p>Many businesses treat rules of origin as a customs-agent problem. They send whatever documents the agent asks for and hope the claim works.</p><p>When a partner country’s customs authority audits origin two years later, there are no cost records, no supplier declarations and no explanation of how the rule was met. The buyer receives a duty demand — and blames the exporter.</p><p><strong>An FTA claim you cannot prove is not a saving. It is a delayed liability for your customer.</strong></p><h3>The FTA Claim Ladder</h3><p>Climb four steps for every product and market:</p><h4>Step 1 — Classify</h4><p>Confirm the correct <strong>HS code</strong> for your product, in agreement with your buyer’s import classification. Origin rules are linked to codes.</p><h4>Step 2 — Qualify</h4><p>Check the <strong>product-specific rule</strong> in the agreement. Use your bill of materials, supplier origins and cost data to test whether the product qualifies.</p><h4>Step 3 — Certify</h4><p>Obtain the <strong>certificate of origin</strong> through the official process — in India, mostly through the Directorate General of Foreign Trade’s online platform — or make a valid self-declaration where allowed.</p><h4>Step 4 — Document</h4><p>Keep an <strong>origin file</strong> for each product: classification reasoning, bill of materials, supplier declarations, cost workings and production records. Retain it for the period required by the agreement and laws.</p><p><strong>Climb all four steps, every time. Skipping one turns a saving into a risk.</strong></p><h3>A worked example (illustrative)</h3><p>Consider an illustrative Indian manufacturer of auto parts exporting ₹40 crore a year to the UAE and Australia. It has never claimed FTA preferences because “the buyer handles customs.”</p><p><strong>Using the FTA Claim Ladder:</strong></p><ul><li><strong>Classify:</strong> confirms HS codes with both buyers.</li><li><strong>Qualify:</strong> finds that 80% of its products meet the change-in-classification rule; others need a value-content calculation, which they pass.</li><li><strong>Certify:</strong> registers on the DGFT platform and obtains certificates of origin.</li><li><strong>Document:</strong> builds an origin file for each product.</li></ul><p><strong>Result:</strong> the buyers’ duties on its products fall — in many tariff lines to zero. One Australian buyer increases orders by 25% because the landed cost is now lower than a competing supplier’s. <strong>The products did not change. The paperwork did.</strong></p><h3>What leaders should do now</h3><ul><li><strong>Owners:</strong> ask your export team for a list of FTA markets and products where you are not claiming preferences — and why.</li><li><strong>CFOs:</strong> quantify the duty saving available to your buyers and use it in pricing and negotiations.</li><li><strong>Export and customs teams:</strong> build origin files and train on product-specific rules.</li><li><strong>Sales teams:</strong> tell buyers you can provide valid FTA origin proof — it is a selling point.</li></ul><h3>How you’ll know it’s working</h3><ul><li><strong>FTA utilisation</strong> — the share of eligible exports shipped with valid origin proof — rises towards 100%.</li><li><strong>Origin files</strong> exist for every qualifying product.</li><li><strong>Verification requests</strong> are answered quickly without denied claims.</li><li><strong>Buyers</strong> cite your FTA eligibility in sourcing decisions.</li></ul><h3>Where this is heading</h3><p>India’s network of trade agreements is expanding quickly, including with the UK, EFTA, Oman, New Zealand and the European Union. Rules of origin are becoming more detailed and more digitally verified. <strong>The exporters who master origin rules will turn each new agreement into new orders.</strong></p><h3>If you run a small business</h3><p>Consider an illustrative Indian spice exporter shipping to Australia without claiming preferences under the India–Australia agreement. Its buyer pays the normal duty.</p><p>After obtaining certificates of origin through the DGFT platform, the exporter’s products enter Australia duty-free in the relevant tariff lines. <strong>The buyer increases orders, and the exporter negotiates a slightly better price — sharing the duty saving.</strong></p><h3>“But…” — the best counterargument</h3><p><strong>The objection:</strong> “FTA paperwork costs time and money, and the duty saving goes to the buyer, not us. Why should we bother?”</p><p><strong>The answer:</strong> The duty saving lowers your buyer’s landed cost, which makes your product more competitive than suppliers from countries without an FTA. That can win orders, protect prices or justify a better margin. Once origin files are built, repeating the process is quick. <strong>You are not doing the buyer’s paperwork. You are building your price advantage.</strong></p><h3>Questions people ask me</h3><p><strong>“How do we know if our product is covered?”</strong></p><p>Check the agreement’s tariff schedules for your HS code and the partner’s duty rates, using official commerce ministry and partner customs resources.</p><p><strong>“Can we claim FTA benefits on imports into India?”</strong></p><p>Yes, if goods meet the origin rules. Under CAROTAR, importers must possess origin information and respond to verification requests.</p><p><strong>“Do trading companies qualify?”</strong></p><p>Origin depends on where goods were produced, not who sells them. Traders can supply origin documents for qualifying goods, but must have supporting information.</p><p><strong>“What if our buyer is in the EU?”</strong></p><p>The India–EU agreement was concluded in January 2026 and is expected to be signed in December 2026; it will only apply after ratification and entry into force. Prepare origin data now.</p><p><strong>Where things stand (October 2026):</strong> India’s agreements with the UAE, Australia, ASEAN, Japan, Korea, EFTA (from October 2025), Oman (from 1 June 2026) and the UK (from 15 July 2026) are in force; the India–New Zealand agreement is scheduled to enter into force on 20 October 2026; the India–EU agreement was concluded in January 2026 and awaits signature and ratification. CAROTAR 2020 governs origin verification on imports into India.</p><h3>Mistakes I see most often</h3><ul><li>Assuming the buyer or customs agent will handle FTA claims.</li><li>Claiming preferences without checking the product-specific rule.</li><li>Keeping no records of supplier origin and costs.</li><li>Using simple processing or transhipment to claim origin.</li><li>Ignoring FTAs when choosing markets.</li></ul><h3>One-page checklist</h3><ul><li>FTA eligibility checked for every product and market</li><li>Origin files kept for each product</li><li>Certificates of origin obtained before shipment</li><li>Buyers told about FTA eligibility</li></ul><h3>Try this in 30 minutes</h3><p>List your top five export products and markets. For each, find: <strong>is there an FTA? What is the normal duty? What is the FTA duty? Are we claiming it?</strong></p><p>Every “no” next to a duty difference is money your buyers are paying — and a reason they might choose someone else.</p><h3>The short version</h3><ul><li>India&#39;s FTA network has expanded quickly — UAE, Australia, EFTA, Oman, UK, New Zealand and the concluded EU agreement — but many exporters do not claim the benefits.</li><li>FTAs reduce duties only for goods meeting rules of origin, proved with certificates and records.</li><li>CAROTAR 2020 and partner-country audits make documentation essential.</li><li>Climb the FTA Claim Ladder: Classify, Qualify, Certify, Document.</li><li>FTA preferences can lower buyers&#39; landed costs and win orders.</li></ul><p><strong>Trade negotiators open the door. Your paperwork decides whether you walk through it.</strong></p><h3>Before you go</h3><p>Classify your products correctly, check whether they qualify under the origin rules, get the right certificate and keep the evidence. Unclaimed preferences are margin left at the border.</p><h3>Further reading</h3><ul><li>Ministry of Commerce and Industry, Government of India — trade agreements and tariff schedules. commerce.gov.in</li><li>Directorate General of Foreign Trade — Common Digital Platform for Certificates of Origin. trade.gov.in</li><li>Central Board of Indirect Taxes and Customs — Customs (Administration of Rules of Origin under Trade Agreements) Rules, 2020. cbic.gov.in</li><li>Asian Development Bank, research on FTA utilisation by Asian firms, including India. adb.org</li><li>Douglas A. Irwin, <em>Free Trade Under Fire</em> (Princeton University Press, 5th edition, 2020).</li></ul><p>◆ ◆ ◆</p><p><em>A version of this article first appeared on ComplianceCalendar.in.</em></p><p><em>A note on numbers for readers outside India: 1 lakh = 100,000 and 1 crore = 10 million (₹ = Indian rupees).</em></p><p><em>I&#39;m Aashok Kumar Sharma, founder of Aashok F&amp;C Advisory — a finance, accounting and compliance practice working with founders, SMEs and finance teams in India and abroad. I write about the numbers, rules and decisions that quietly decide whether businesses last. Follow me here on Medium for the rest of this 100-part series.</em></p><p><em>This article is general information, not legal, tax or investment advice. Rules change; check current provisions or speak to a qualified professional before acting. Examples labelled illustrative are composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=87bfffdfc0b4" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Demographics Are Destiny: Ageing Nations, Young Workforces and the Business Map of 2050]]></title>
            <link>https://medium.com/@aashokfcadvisory/demographics-are-destiny-ageing-nations-young-workforces-and-the-business-map-of-2050-c6c6cedaad04?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/c6c6cedaad04</guid>
            <category><![CDATA[business]]></category>
            <category><![CDATA[accounting]]></category>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Sat, 14 Nov 2026 13:55:15 GMT</pubDate>
            <atom:updated>2026-11-14T13:55:15.562Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 71 OF 100 | MEDIUM EDITION | ECONOMICS, DEMOGRAPHICS, BUSINESS STRATEGY, FUTURE OF WORK, INDIA</strong></p><p><em>South Korea’s fertility rate fell to 0.72. India is the world’s most populous country. Your long-range plan should know the difference.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>Most business forecasts look three to five years ahead. Demographics move more slowly — and more predictably — than almost anything else in economics.</p><p>Whether you’re planning in India, Europe or East Asia, the age structure of your markets and workforce will shape demand, wages and investment for decades.</p><p>In 2023, South Korea’s total fertility rate — the average number of children a woman is expected to have — fell to <strong>0.72</strong>, the lowest recorded for any country. A rate of about 2.1 is needed to keep a population stable without migration.</p><p>In the same year, the United Nations estimated that <strong>India had overtaken China</strong> as the world’s most populous country. China’s population had begun to shrink in 2022 for the first time in six decades. Japan’s population had been falling for more than a decade.</p><p>Meanwhile, the median age in much of sub-Saharan Africa was under 20.</p><p><strong>These numbers do not make headlines every day. But over the next 25 years, they will shape labour costs, consumer markets, savings, healthcare, migration and growth more than almost any policy decision.</strong></p><h3>Why demographics are different</h3><p>Most business forecasts are uncertain. Demographics are not entirely certain, but they are far more predictable:</p><ul><li>The workers of 2045 are already children today.</li><li>The retirees of 2050 are already in their 30s and 40s.</li><li>Fertility trends change slowly and are hard for governments to reverse.</li></ul><p>That makes demographics one of the few genuine long-term signals available to business leaders. <strong>Ignoring them is like planning a journey without looking at the map.</strong></p><h3>A lesson from history: the Black Death and wages</h3><p>Between 1347 and 1351, the Black Death killed perhaps a <strong>third or more</strong> of Europe’s population. Fields went unworked; workers were scarce.</p><p>Surviving labourers demanded higher wages. In England, the government passed the <strong>Statute of Labourers (1351)</strong> to cap wages at pre-plague levels. It largely failed. Over the following decades, wages rose, landlords lost power and the old feudal system weakened.</p><p><strong>When the number of workers falls, the price of labour rises — and laws struggle to hold it back.</strong> Ageing societies today face a slower version of the same force.</p><h3>The emerging map of 2050</h3><h4>Ageing and shrinking</h4><p>Japan, South Korea, China, much of Europe and parts of Eastern Europe face falling working-age populations and rising numbers of elderly people. The UN projects that by the 2030s, people aged 80 and over will outnumber infants in many countries.</p><h4>Still young and growing</h4><p>India, much of South and Southeast Asia and especially Africa will provide most of the growth in the world’s working-age population. Sub-Saharan Africa alone is projected to account for a large share of global population growth to 2050.</p><h4>In between</h4><p>The United States, parts of Latin America and others are ageing more slowly, supported by migration or higher fertility.</p><p><strong>The world is splitting into countries short of workers and countries with workers looking for jobs.</strong></p><p><strong>Ageing nations will have capital and need workers. Young nations will have workers and need capital.</strong></p><h3>What it means for business</h3><p>1. <strong>Labour scarcity in ageing economies</strong> — higher wages, more automation and robotics, more migration and offshoring of services.</p><p>2. <strong>The “silver economy”</strong> — rising demand for healthcare, elder care, financial planning, accessible products, leisure and housing for older people.</p><p>3. <strong>Consumer growth in young economies</strong> — education, first homes, first cars, entertainment and digital services.</p><p>4. <strong>Savings and capital flows</strong> — ageing savers seek returns abroad; young economies need investment.</p><p>5. <strong>Pension and fiscal pressure</strong> — higher taxes or reduced benefits in ageing countries.</p><p>6. <strong>Talent competition</strong> — global competition for skilled workers in health, technology and engineering.</p><h3>India’s window</h3><p>India’s working-age population will remain large for decades. Economists call this a potential <strong>demographic dividend</strong> — growth that comes when more people are of working age.</p><p>But a dividend is not automatic. It needs:</p><ul><li>Education and skills that match jobs.</li><li>Jobs — especially in manufacturing and services — for millions entering the workforce each year.</li><li>Higher participation of women in the workforce.</li><li>Health and infrastructure.</li></ul><p><strong>Without jobs, a young population becomes a demographic burden, not a dividend.</strong> India itself is ageing in its southern states, where fertility is already below replacement.</p><h3>The uncomfortable part</h3><p>Many businesses still plan as if the next 25 years will look like the last 25: abundant young workers in their home market, growing customer numbers everywhere, and stable pension systems.</p><p>In ageing markets, that assumption is already false. In young markets, the opportunity is often wasted because businesses do not invest in skills.</p><p><strong>Demographics do not ask for permission. They simply arrive.</strong></p><h3>The Four Demographic Questions</h3><p>Ask four questions about every major market where you hire, sell or invest:</p><h4>1. Workers — Will we find the people we need?</h4><p>How will the working-age population in this location change by 2035 and 2050? What will that do to wages and availability of skills?</p><h4>2. Customers — Who will our customers be?</h4><p>Will our core customers grow or shrink in number? Will their needs change with age?</p><h4>3. Savers and spenders — Where will money come from?</h4><p>Will customers be saving for retirement, spending on families or drawing down savings? How does that affect demand for our products and the cost of capital?</p><h4>4. Carers and dependants — Who will look after whom?</h4><p>How will rising care responsibilities affect employees and customers — and what services will be needed?</p><h3>A worked example (illustrative)</h3><p>Consider an illustrative Indian IT and engineering services company with clients in Japan and Germany.</p><p><strong>The Four Questions reveal:</strong></p><ul><li><strong>Workers:</strong> clients face severe shortages of engineers and care staff; India and Southeast Asia have growing talent pools.</li><li><strong>Customers:</strong> clients need automation, remote services and products for older users.</li><li><strong>Savers:</strong> client-country pension funds seek long-term investments in growing economies.</li><li><strong>Carers:</strong> client companies’ employees increasingly balance work and elder care.</li></ul><p><strong>The strategy:</strong></p><ul><li>Expands Japanese- and German-language engineering teams in India.</li><li>Builds offerings in industrial automation, remote monitoring and healthcare technology.</li><li>Creates training academies with Indian universities to grow skilled talent.</li></ul><p><strong>Within five years</strong>, demographic change in client countries becomes its biggest growth driver. <strong>The company did not predict the future. It read the demographic map that was already printed.</strong></p><h3>What leaders should do now</h3><ul><li><strong>CEOs and boards:</strong> include 10- and 25-year demographic trends in strategy reviews.</li><li><strong>HR leaders:</strong> plan for talent shortages in some markets and skills investment in others; design policies for older workers and carers.</li><li><strong>Product teams:</strong> design for ageing customers — accessibility, health, simplicity.</li><li><strong>Investors and CFOs:</strong> consider how demographics affect demand, wages and cost of capital in each market.</li></ul><h3>How you’ll know it’s working</h3><ul><li><strong>Long-term plans</strong> include demographic assumptions for each market.</li><li><strong>Workforce plans</strong> address future shortages and skills.</li><li><strong>Products and services</strong> for older customers grow as a share of revenue.</li><li><strong>Recruitment</strong> draws on young talent pools in growing regions.</li></ul><h3>Where this is heading</h3><p>Labour shortages in ageing economies will deepen, increasing demand for automation, migration and offshore services. India’s large working-age population will create opportunities in services, manufacturing and global capability centres — if skills and jobs keep pace. <strong>Demographics will quietly decide which markets grow and which shrink.</strong></p><h3>If you run a small business</h3><p>Consider an illustrative home-healthcare startup in Kerala, where the population is ageing faster than the national average. It trains local nurses and caregivers and offers elder-care packages to families whose children work abroad.</p><p>Demand grows steadily as more families need support. <strong>The startup did not create the trend. It read the demographic map and built a business on it.</strong></p><h3>“But…” — the best counterargument</h3><p><strong>The objection:</strong> “Demographic change is too slow to matter for our business plans. Technology, AI and automation will change things faster.”</p><p><strong>The answer:</strong> Technology and demographics interact. Automation and AI are being adopted fastest where workers are scarcest — partly because of demographics. And slow change is exactly what makes demographics powerful: it builds silently until it dominates. <strong>A five-year plan can ignore demographics. A business that wants to exist in 2050 cannot.</strong></p><h3>Questions people ask me</h3><p><strong>“Is India’s demographic dividend guaranteed?”</strong></p><p>No. It depends on jobs, skills, health, women’s participation and infrastructure. Some southern states already have below-replacement fertility.</p><p><strong>“Will migration solve labour shortages in ageing countries?”</strong></p><p>It can help, but politics, integration and competition for migrants limit how much.</p><p><strong>“What is the silver economy?”</strong></p><p>The economic activity linked to the needs and spending of older people — healthcare, care services, financial products, housing, travel and accessible products.</p><p><strong>&quot;How should small businesses think about demographics?&quot;</strong></p><p>Look at your local market: Is your customer base ageing or growing? Are young workers moving away? Plan products and hiring accordingly.</p><p><strong>Where things stand (October 2026):</strong> The UN&#39;s <em>World Population Prospects 2024</em> projects the world population to peak in the 2080s, with most growth to 2050 in Africa and South Asia. India is the world&#39;s most populous country, while China, Japan, South Korea and much of Europe face shrinking working-age populations. Use the latest UN and national data when planning long-term investments.</p><h3>Mistakes I see most often</h3><ul><li>Assuming a steady supply of young workers in every market.</li><li>Ignoring older customers as a growth segment.</li><li>Treating a demographic dividend as automatic.</li><li>Planning only for five years in long-life investments.</li><li>Overlooking employees&#39; growing care responsibilities.</li></ul><h3>One-page checklist</h3><ul><li>Demographic data reviewed for key markets</li></ul><h3>Try this in 30 minutes</h3><p>For your two most important markets, look up three numbers: <strong>median age today, projected working-age population change to 2050, and share of population over 65 in 2050.</strong></p><p>Then ask: what will these numbers mean for our customers, workers and costs? Write one action for each.</p><h3>The short version</h3><ul><li>South Korea&#39;s fertility rate of 0.72 (2023), China&#39;s shrinking population and India&#39;s overtaking of China signal a deep demographic shift.</li><li>The Black Death shows that when workers become scarce, wages rise and old systems change.</li><li>By 2050, ageing nations will be short of workers, while young nations will need jobs and capital.</li><li>Ask the Four Demographic Questions: workers, customers, savers and spenders, carers and dependants.</li><li>India&#39;s dividend is an opportunity, not a certainty — it needs skills and jobs.</li></ul><p><strong>The future customer and worker are already born. Plan for who they will be.</strong></p><h3>Before you go</h3><p>Put your key markets&#39; population trends into your next strategy review. Demographics won&#39;t decide your business on their own, but ignoring them is choosing to be surprised.</p><h3>Further reading</h3><ul><li>United Nations, Department of Economic and Social Affairs, <em>World Population Prospects 2024</em>. population.un.org</li><li>Statistics Korea, birth statistics for 2023 (February 2024). kostat.go.kr</li><li>National Bureau of Statistics of China, population data release (January 2023).</li><li>John Hatcher, <em>The Black Death: A Personal History</em> (Da Capo Press, 2008); Statute of Labourers, 1351.</li><li>Charles Goodhart and Manoj Pradhan, <em>The Great Demographic Reversal</em> (Palgrave Macmillan, 2020).</li></ul><p>◆ ◆ ◆</p><p><em>A version of this article first appeared on ComplianceCalendar.in.</em></p><p><em>I&#39;m Aashok Kumar Sharma, founder of Aashok F&amp;C Advisory — a finance, accounting and compliance practice working with founders, SMEs and finance teams in India and abroad. I write about the numbers, rules and decisions that quietly decide whether businesses last. Follow me here on Medium for the rest of this 100-part series.</em></p><p><em>This article is general information, not legal, tax or investment advice. Rules change; check current provisions or speak to a qualified professional before acting. Examples labelled illustrative are composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c6c6cedaad04" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Balance Sheet Is Your Real Report Card — Here’s How to Read It in Five Minutes]]></title>
            <link>https://medium.com/@aashokfcadvisory/the-balance-sheet-is-your-real-report-card-heres-how-to-read-it-in-five-minutes-2a3ef8f93bb9?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/2a3ef8f93bb9</guid>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[accounting]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Sat, 14 Nov 2026 12:26:01 GMT</pubDate>
            <atom:updated>2026-11-14T12:26:01.423Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 26 OF 100 | MEDIUM EDITION | ACCOUNTING, FINANCE, SMALL BUSINESS, INVESTING, ENTREPRENEURSHIP</strong></p><p>The Balance Sheet Is Your Real Report Card — Here’s How to Read It in Five Minutes</p><p><em>Lehman Brothers dressed up its balance sheet at every quarter-end. Five simple readings tell you what a profit figure can’t.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>Most owners I work with read the profit and loss account closely and skip the balance sheet entirely. It looks like a list of balances, and nobody explained what it’s saying.</p><p>That’s a pity, because it answers the questions that matter most: can we pay our bills, how much do we owe and what is really ours? It works the same way under Indian, UK or US accounting rules.</p><p>In March 2010, a court-appointed examiner published his report on the collapse of Lehman Brothers, the US investment bank whose failure in September 2008 helped trigger a global financial crisis.</p><p>Among his findings: at the end of several quarters, Lehman had used accounting transactions known internally as “Repo 105” to temporarily move tens of billions of dollars of assets off its balance sheet — making its borrowing look lower just as investors and rating agencies were looking.</p><p><strong>The profit figures had told one story. The balance sheet — the real one — told another.</strong></p><h3>The page most owners skip</h3><p>Many owners read one page of their financial statements: the profit and loss account. Did sales rise? Did profit rise? Good year.</p><p>The balance sheet is left to the accountant and the bank. That is a mistake.</p><p><strong>Profit measures performance over a period. The balance sheet measures position at a moment</strong> — what the business owns, what it owes, and what is left for its owners — and whether that picture is improving or deteriorating.</p><h3>A lesson from history: the investor’s discipline</h3><p>In 1934, in the depths of the Great Depression, Benjamin Graham and David Dodd published <em>Security Analysis</em>. Having watched investors lose fortunes by chasing reported earnings, they urged readers to examine what a company actually owned and owed — its assets, liabilities and true financial strength.</p><p>Graham’s later student, Warren Buffett, built one of the world’s great investment records on that discipline. <strong>The balance sheet is where an optimistic story meets arithmetic.</strong></p><h3>What the balance sheet shows</h3><p><strong>Assets = Liabilities + Equity</strong></p><p>· <strong>Assets</strong> — what the business controls: cash, receivables, inventory, property, equipment, investments, intangibles.</p><p>· <strong>Liabilities</strong> — what it owes: suppliers, lenders, taxes, employees, customer advances, provisions.</p><p>· <strong>Equity</strong> — what belongs to owners after all liabilities: capital contributed plus accumulated profits (or minus losses).</p><h3>The Five Health Readings</h3><p>Doctors read several measures to judge health. Owners should read five on the balance sheet.</p><h3>1. Liquidity — can we pay what is due soon?</h3><p>· <strong>Current ratio</strong> = current assets ÷ current liabilities.</p><p>· <strong>Quick ratio</strong> = (current assets − inventory) ÷ current liabilities.</p><p>A weak ratio suggests the business may struggle to pay bills without new borrowing. Context matters — some industries run safely on lower ratios.</p><h3>2. Leverage — how much is borrowed?</h3><p>· <strong>Debt-to-equity</strong> = total borrowings ÷ equity.</p><p>Higher leverage magnifies returns in good years and losses in bad ones. Rising leverage without rising profits is a warning.</p><h3>3. Lock-up — is money trapped?</h3><p>Watch three trends:</p><p>· <strong>Receivable days</strong> — how long customers take to pay.</p><p>· <strong>Inventory days</strong> — how long stock sits.</p><p>· <strong>Payable days</strong> — how long you take to pay suppliers.</p><p>If receivables and inventory grow faster than sales, cash is being locked up — even when profit looks fine.</p><h3>4. Returns — is the business worth the risk?</h3><p>· <strong>Return on equity (ROE)</strong> = net profit ÷ average equity.</p><p>· <strong>Return on capital employed (ROCE)</strong> = operating profit ÷ (equity + long-term debt).</p><p>A business can be profitable yet earn less than its owners could earn elsewhere with less risk.</p><h3>5. Net worth — is the foundation growing?</h3><p>If equity grows year after year through retained profits, the business is getting stronger. If it shrinks — through losses or withdrawals larger than profits — the business is getting weaker, whatever sales say.</p><p><strong>A growing business with a shrinking net worth is not growing. It is spending its foundation.</strong></p><h3>Red flags owners should look for</h3><p>· <strong>Receivables growing faster than revenue</strong> — customers may be struggling, or sales may be recorded too early.</p><p>· <strong>Old, unchanged receivables</strong> — amounts that never move may never be collected.</p><p>· <strong>Inventory rising without sales growth</strong> — stock may be slow-moving or overvalued.</p><p>· <strong>Loans and advances to related parties or directors</strong> — tying up funds and raising governance, legal and tax questions.</p><p>· <strong>Large “other current assets” or “suspense” balances</strong> — vague items deserve explanation.</p><p>· <strong>Statutory dues outstanding for long periods</strong> — interest, penalties and legal consequences.</p><p>· <strong>Short-term borrowing funding long-term assets</strong> — refinancing risk.</p><p>· <strong>Negative net worth</strong> — liabilities exceed assets; a serious signal for lenders and suppliers, and in some jurisdictions a trigger for directors’ legal duties.</p><p>· <strong>Capital work-in-progress that never finishes</strong> — projects stuck for years may represent lost money.</p><h3>Window dressing: the Lehman lesson for small businesses</h3><p>Window dressing means making the balance sheet look better on the reporting date than it really is: delaying supplier payments recorded, repaying an overdraft on the last day and redrawing it the next, booking sales early, or moving balances temporarily.</p><p>Small businesses sometimes do this before bank reviews. It may work once. But lenders compare balances across months, auditors test cut-off, and patterns become obvious. <strong>A balance sheet that looks good only on 31 March is not strong. It is dressed.</strong></p><h3>Contingent liabilities: the hidden page</h3><p>Disputed tax demands, guarantees given and pending litigation may not appear in the main balance sheet figures, but are disclosed in the notes. They can become real liabilities overnight. <strong>Every owner should know every one of them by heart.</strong></p><h3>The uncomfortable part</h3><p>Some owners deliberately keep profits low on paper to reduce tax, then wonder why banks will not lend and investors are uninterested.</p><p>Others show high profits on balance sheets full of receivables that will never be collected and stock that will never sell.</p><p><strong>Neither picture is honest. And dishonest balance sheets eventually present their bill.</strong></p><h3>How to use the balance sheet in practice</h3><p>1. <strong>Compare three years side by side.</strong> Trends reveal more than one year.</p><p>2. <strong>Read the notes.</strong> Borrowings, contingent liabilities, related-party transactions and accounting policies often contain the real story.</p><p>3. <strong>Track key ratios quarterly</strong>, not just annually.</p><p>4. <strong>Ask your accountant to explain every significant change</strong> in plain language.</p><p>5. <strong>Link it to the cash flow statement.</strong> Together they show where cash came from and where it went.</p><h3>How you’ll know it’s working</h3><p>A stronger balance sheet shows up in the Five Health Readings over time:</p><p>· <strong>Liquidity</strong> — current assets comfortably cover current liabilities.</p><p>· <strong>Leverage</strong> — debt falls relative to equity and cash flow.</p><p>· <strong>Working capital</strong> — receivables and inventory do not grow faster than sales.</p><p>· <strong>Asset quality</strong> — fewer doubtful debts, obsolete stock or idle assets.</p><p>· <strong>Equity growth</strong> — retained profits build the business’s net worth year after year.</p><h3>Where this is heading</h3><p>Lenders, investors and suppliers increasingly use automated tools to read balance sheets — from credit bureaus and GST data to AI-based financial analysis. <strong>A weak balance sheet will be spotted faster and priced higher. A strong one will open doors sooner.</strong> Owners who read their own balance sheet every quarter will see what outsiders see — before they act on it.</p><h3>What to do on Monday, by role</h3><p>· <strong>Owner:</strong> compare this year’s balance sheet with the last three years, line by line.</p><p>· <strong>CFO:</strong> list assets that may be overstated — old receivables, slow stock, unused equipment.</p><p>· <strong>Board:</strong> add the Five Health Readings to every quarterly review.</p><p>· <strong>Accountant:</strong> reconcile related-party and director loan balances.</p><h3>“But…” — the best counterargument</h3><p><strong>The objection:</strong> “I’m not an accountant. The balance sheet is technical. As long as the business makes profit and pays bills, I don’t need to understand ratios.”</p><p><strong>The answer:</strong> You do not need to be an accountant — five readings and three years of comparison are enough. And “paying bills” is exactly what the balance sheet predicts: liquidity and lock-up tell you whether you will be able to pay <strong>next</strong> quarter, not just this one. Lenders, investors and buyers will read your balance sheet whether you do or not. <strong>Better that you see the problem first.</strong></p><h3>Questions people ask me</h3><p><strong>“What is a good current ratio?”</strong></p><p>There is no universal number. Many analysts treat a ratio between about 1.2 and 2 as comfortable for typical trading businesses, but retailers with fast stock turnover can operate lower, and project businesses may need higher. Compare with your own history and your industry.</p><p><strong>“Is debt on the balance sheet always bad?”</strong></p><p>No. Debt that funds productive assets, with comfortable interest coverage and matched tenure, can increase returns. Debt that funds losses or withdrawals is the danger.</p><p><strong>“How often should I look at it?”</strong></p><p>Quarterly at minimum; monthly if cash is tight or the business is growing fast.</p><p><strong>Where things stand (October 2026):</strong> Indian companies present balance sheets in Schedule III format, which requires disclosures such as ageing of receivables and payables, key ratios with explanations of large changes, and details of loans to promoters and related parties. These disclosures make many of the red flags above visible to lenders and investors automatically.</p><h3>Mistakes I see most often</h3><p>· Ignoring the balance sheet because “the CA handles it.”</p><p>· Focusing only on revenue and profit growth.</p><p>· Leaving old, disputed balances uninvestigated.</p><p>· Treating loans to related parties as harmless.</p><p>· Window dressing before bank reviews.</p><h3>Try this in 30 minutes</h3><p>Put your last three balance sheets side by side. Calculate the Five Health Readings for each year:</p><p>· <strong>Current ratio</strong></p><p>· <strong>Debt-to-equity</strong></p><p>· <strong>Receivable + inventory days</strong></p><p>· <strong>Return on equity</strong></p><p>· <strong>Net worth</strong></p><p>Mark each direction as improving or worsening. Any reading worsening for two years in a row deserves a conversation this month.</p><h3>The short version</h3><p>· The balance sheet shows financial position — what you own, owe and keep — and the Lehman case shows how much it can reveal or hide.</p><p>· Read the Five Health Readings: liquidity, leverage, lock-up, returns and net worth.</p><p>· Watch for red flags: rising receivables and stock, related-party advances, vague balances, overdue statutory dues and mismatched funding.</p><p>· Window dressing does not create strength; it only delays its discovery.</p><p>· Compare three years, read the notes, track ratios quarterly and know every contingent liability.</p><p><strong>Your profit tells you how fast you are moving. Your balance sheet tells you whether the vehicle will survive the journey.</strong></p><h3>Before you go</h3><p>You don’t need an accounting degree to read a balance sheet — just five questions and the habit of asking them every quarter. Start with the next set of accounts that lands on your desk.</p><h3>Further reading</h3><p>· Anton R. Valukas, <em>Report of Examiner, In re Lehman Brothers Holdings Inc.</em> (US Bankruptcy Court, S.D.N.Y., March 2010).</p><p>· Benjamin Graham and David L. Dodd, <em>Security Analysis</em> (McGraw-Hill, 1934).</p><p>· Companies Act, 2013 (India), Schedule III (as amended in 2021). mca.gov.in</p><p>· Martin Fridson and Fernando Alvarez, <em>Financial Statement Analysis: A Practitioner’s Guide</em> (Wiley; multiple editions).</p><p>◆ ◆ ◆</p><p><em>A version of this article first appeared on ComplianceCalendar.in.</em></p><p><em>I’m Aashok Kumar Sharma, founder of Aashok F&amp;C Advisory — a finance, accounting and compliance practice working with founders, SMEs and finance teams in India and abroad. I write about the numbers, rules and decisions that quietly decide whether businesses last. Follow me here on Medium for the rest of this 100-part series.</em></p><p><em>This article is general information, not legal, tax or investment advice. Rules change; check current provisions or speak to a qualified professional before acting. Examples labelled illustrative are composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2a3ef8f93bb9" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Machine Learning in Plain Business Language: Where Prediction Creates Real Value]]></title>
            <link>https://medium.com/@aashokfcadvisory/machine-learning-in-plain-business-language-where-prediction-creates-real-value-f9aa52ed2309?source=rss-a9a1dc867724------2</link>
            <guid isPermaLink="false">https://medium.com/p/f9aa52ed2309</guid>
            <category><![CDATA[entrepreneurship]]></category>
            <category><![CDATA[business]]></category>
            <category><![CDATA[startup]]></category>
            <category><![CDATA[finance]]></category>
            <category><![CDATA[accounting]]></category>
            <dc:creator><![CDATA[Aashok F&C Advisory]]></dc:creator>
            <pubDate>Fri, 13 Nov 2026 13:32:25 GMT</pubDate>
            <atom:updated>2026-11-13T13:32:25.048Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>ARTICLE 80 OF 100 | MEDIUM EDITION | MACHINE LEARNING, ARTIFICIAL INTELLIGENCE, BUSINESS, DATA SCIENCE, TECHNOLOGY</strong></p><p><em>Netflix paid $1 million for a better algorithm — and never fully used it. Here’s how to tell useful prediction from expensive novelty.</em></p><p>By Aashok Kumar Sharma · Founder, Aashok F&amp;C Advisory</p><p>Machine learning gets explained either as magic or as maths. Neither helps a business owner decide whether to fund a project.</p><p>At its heart, it’s prediction: using past data to estimate something you don’t yet know. The useful question — in any business, anywhere — is whether that prediction will change a decision enough to be worth the cost.</p><p>In 2006, Netflix offered a prize of <strong>US$1 million</strong> to anyone who could improve its movie-recommendation predictions by 10%. Thousands of teams from around the world competed. In 2009, a combined team called BellKor’s Pragmatic Chaos won.</p><p>Then something surprising happened. Netflix later explained that it did not put the full winning solution into production. The extra accuracy did not justify the engineering effort — and by then the business was shifting from DVDs by mail to streaming, which changed what needed to be predicted.</p><p>Twelve years later, the US property company <strong>Zillow</strong> made the opposite mistake. Its “Zillow Offers” business used algorithms to estimate home values and buy houses directly. In November 2021, Zillow announced it would shut the business, with losses of more than half a billion dollars and about a quarter of its staff to be cut. Its chief executive said the company had been unable to forecast future home prices accurately enough.</p><p><strong>One company held back a better prediction because it did not change the business enough. The other bet the business on a prediction that was not good enough.</strong></p><h3>What machine learning actually is</h3><p>Strip away the jargon:</p><p><strong>Machine learning (ML) is software that learns patterns from past data to make predictions about new cases.</strong></p><ul><li>Which customers are likely to stop buying?</li><li>How much of each product will sell next month?</li><li>Is this transaction likely to be fraud?</li><li>Will this borrower repay?</li><li>When is this machine likely to fail?</li></ul><p>Instead of a person writing every rule, the system finds patterns in thousands or millions of examples.</p><h3>A lesson from history: Halley’s life table</h3><p>In 1693, the astronomer <strong>Edmond Halley</strong> — famous for his comet — studied records of births and deaths in the city of Breslau. He built one of the first reliable <strong>life tables</strong>, estimating the probability that a person of a given age would live another year.</p><p>He showed how the table could be used to price annuities — regular payments for life — which governments were selling at prices that ignored age.</p><p><strong>Halley turned data into prediction, and prediction into better financial decisions.</strong> Modern machine learning does the same, at far larger scale.</p><h3>Where prediction creates real value</h3><p>ML is most valuable when:</p><p>1. <strong>A decision is made often</strong> — thousands of times a day or month.</p><p>2. <strong>Good data exists</strong> — historical examples of the outcome you want to predict.</p><p>3. <strong>Better prediction changes the action</strong> — what you stock, approve, price or prioritise.</p><p>4. <strong>The value of improvement is measurable</strong> — fewer stockouts, lower fraud, higher retention.</p><p>5. <strong>Errors are tolerable or controllable</strong> — mistakes can be caught or reversed.</p><p>Common high-value uses:</p><ul><li><strong>Demand forecasting</strong> — inventory and production.</li><li><strong>Fraud detection</strong> — payments and claims.</li><li><strong>Credit risk</strong> — lending decisions.</li><li><strong>Churn prediction</strong> — retention offers for at-risk customers.</li><li><strong>Predictive maintenance</strong> — fixing machines before failure.</li><li><strong>Dynamic pricing</strong> — matching prices to demand.</li><li><strong>Lead scoring</strong> — prioritising sales effort.</li></ul><p><strong>A prediction that does not change a decision is just an interesting number.</strong></p><h3>Where ML fails</h3><ul><li><strong>Rare or unprecedented events</strong> — models learn from the past; they struggle when the world changes, as Zillow found.</li><li><strong>Poor data</strong> — incomplete, biased or inconsistent records produce poor predictions.</li><li><strong>Unclear decisions</strong> — nobody knows what will be done differently with the prediction.</li><li><strong>No feedback loop</strong> — the model is never checked against actual outcomes.</li><li><strong>Over-confidence</strong> — treating predictions as certainties.</li></ul><h3>The uncomfortable part</h3><p>Many companies start ML projects because competitors do, or because they have “a lot of data.” They hire data scientists, build models and present impressive accuracy figures.</p><p>Then nothing changes in operations. The sales team ignores the lead scores. The warehouse ignores the forecast. The model sits unused.</p><p><strong>The problem is rarely the algorithm. It is starting with the technology instead of the decision.</strong></p><h3>The Five-Question Prediction Test</h3><p>Before approving any ML project, answer five questions:</p><p>1. <strong>Decision:</strong> What specific decision will this prediction improve — and who makes it?</p><p>2. <strong>Data:</strong> Do we have enough reliable historical data on the outcome?</p><p>3. <strong>Payoff:</strong> What is a 10% better prediction worth in rupees per year?</p><p>4. <strong>Error cost:</strong> What happens when the prediction is wrong — and how will we catch it?</p><p>5. <strong>Action:</strong> How will the prediction reach the person or system that acts — and will they use it?</p><p><strong>If you cannot answer all five, you have a data science experiment, not a business project.</strong></p><h3>A worked example (illustrative)</h3><p>Consider an illustrative grocery and household-goods distributor with 8,000 product lines and ₹600 crore revenue. Stockouts are common for fast movers, while slow movers pile up.</p><p><strong>The Five-Question Test:</strong></p><ul><li><strong>Decision:</strong> weekly reorder quantities for each product and warehouse.</li><li><strong>Data:</strong> three years of daily sales, promotions, seasons and supplier lead times.</li><li><strong>Payoff:</strong> each 1% reduction in stockouts is estimated to add about ₹2 crore in sales; each day of excess inventory costs about ₹40 lakh a year in working capital.</li><li><strong>Error cost:</strong> wrong forecasts lead to stockouts or excess — controlled by buyer review for large orders.</li><li><strong>Action:</strong> forecasts feed directly into the purchase-order system, with buyers approving exceptions.</li></ul><p><strong>Results after one year:</strong> stockouts down 30%, inventory down eight days, and buyers spend more time negotiating with suppliers instead of calculating orders. <strong>The model was useful because the decision, data and action were defined first.</strong></p><h3>How to start</h3><p>1. <strong>Pick one high-frequency decision</strong> with clear value.</p><p>2. <strong>Check data quality</strong> before building models.</p><p>3. <strong>Start simple</strong> — sometimes basic statistics beat complex models.</p><p>4. <strong>Run a pilot</strong> against current practice and measure.</p><p>5. <strong>Put predictions into workflows</strong> — not just dashboards.</p><p>6. <strong>Monitor performance</strong> — models drift as conditions change.</p><p>7. <strong>Keep humans in charge</strong> of high-stakes and unusual decisions.</p><h3>How you’ll know it’s working</h3><ul><li><strong>Decision outcomes</strong> — fewer stockouts, lower fraud, better retention — improve against a baseline.</li><li><strong>Model accuracy</strong> is tracked monthly and does not drift unnoticed.</li><li><strong>Adoption</strong> — the people who make the decision actually use the prediction.</li><li><strong>Value delivered</strong> in rupees exceeds the cost of building and running the model.</li></ul><h3>Where this is heading</h3><p>Machine learning will be built into almost every business application, often without being labelled as such. The competitive edge will come less from building models and more from <strong>having clean data, clear decisions and the discipline to act on predictions</strong> — and to know when not to trust them.</p><h3>What to do on Monday, by role</h3><ul><li><strong>CEO:</strong> name the one high-frequency decision you most want to improve.</li><li><strong>CFO:</strong> estimate the rupee value of a 10% better prediction for that decision.</li><li><strong>Data or IT team:</strong> check the quality of the data behind it.</li><li><strong>Business owner of the decision:</strong> agree how predictions will reach the people who act.</li></ul><h3>“But…” — the best counterargument</h3><p><strong>The objection:</strong> “We’re not a tech company. We don’t have data scientists or enough data. Machine learning is for big companies.”</p><p><strong>The answer:</strong> Many ML capabilities are now built into accounting, CRM, inventory and payment software — you may already be paying for them. Mid-sized businesses often have years of sales, inventory and customer data. The key is not hiring a large team; it is <strong>choosing one valuable decision</strong> and using available tools or partners. <strong>Start with the decision, not the data science.</strong></p><h3>Questions people ask me</h3><p><strong>“How is machine learning different from AI?”</strong></p><p>Machine learning is a major part of AI focused on learning patterns from data. Generative AI, such as chatbots, is built using machine-learning techniques.</p><p><strong>“How much data do we need?”</strong></p><p>It depends on the problem. Some useful models need a few thousand examples; others need millions. Quality and relevance matter more than size.</p><p><strong>“How do we know a model works?”</strong></p><p>Test it on data it has not seen, compare it with current methods and track real-world results after launch.</p><p><strong>“Can ML models be biased?”</strong></p><p>Yes. They learn from historical data, which may contain bias. Test outcomes across groups, especially in lending, hiring and pricing.</p><p><strong>Where things stand (October 2026):</strong> Machine-learning features are now embedded in most major ERP, CRM, banking and e-commerce platforms. High-impact uses such as credit scoring and hiring face growing regulatory scrutiny under laws such as the EU AI Act and data-protection rules. Document models used in significant decisions.</p><h3>Mistakes I see most often</h3><ul><li>Starting with “we need AI” instead of a decision.</li><li>Measuring success by model accuracy instead of business results.</li><li>Ignoring data quality.</li><li>Deploying models without monitoring drift.</li><li>Trusting predictions blindly in unusual conditions.</li></ul><h3>One-page checklist</h3><ul><li>Decision defined before data science starts</li><li>Rupee value of better prediction estimated</li><li>Data quality checked</li><li>Predictions built into workflows</li><li>Model accuracy monitored monthly</li><li>Humans in charge of high-stakes cases</li></ul><h3>Try this in 30 minutes</h3><p>List three decisions your business makes hundreds or thousands of times a year — reorders, credit limits, discounts, follow-ups.</p><p>For each, ask: <strong>“If we could predict the outcome 10% better, what would it be worth?”</strong> The highest number is your first ML candidate.</p><h3>The short version</h3><ul><li>Machine learning is prediction from past data — valuable only when it improves a decision.</li><li>Netflix did not deploy its full prize-winning model; Zillow lost heavily relying on predictions that were not good enough.</li><li>ML works best for frequent, measurable decisions with good data and controllable errors.</li><li>Apply the Five-Question Prediction Test: Decision, Data, Payoff, Error cost, Action.</li><li>Start simple, embed predictions in workflows, monitor drift and keep humans in charge of high stakes.</li></ul><p><strong>Do not ask what machine learning can do. Ask which decision you want to make better.</strong></p><h3>Before you go</h3><p>Before you approve a machine-learning project, answer five questions about the decision, the data, the baseline, the cost of errors and who will act on the output. If the answers are fuzzy, the project will be too.</p><h3>Further reading</h3><ul><li>Netflix Technology Blog, “Netflix Recommendations: Beyond the 5 stars (Part 1)” (April 2012).</li><li>Zillow Group, shareholder letter and announcement on winding down Zillow Offers (2 November 2021). investors.zillowgroup.com</li><li>Edmond Halley, “An Estimate of the Degrees of the Mortality of Mankind,” <em>Philosophical Transactions of the Royal Society</em> 17 (1693).</li><li>Ajay Agrawal, Joshua Gans and Avi Goldfarb, <em>Prediction Machines: The Simple Economics of Artificial Intelligence</em> (Harvard Business Review Press, 2018; updated 2022).</li></ul><p>◆ ◆ ◆</p><p><em>A version of this article first appeared on ComplianceCalendar.in.</em></p><p><em>A note on numbers for readers outside India: 1 lakh = 100,000 and 1 crore = 10 million (₹ = Indian rupees).</em></p><p><em>I’m Aashok Kumar Sharma, founder of Aashok F&amp;C Advisory — a finance, accounting and compliance practice working with founders, SMEs and finance teams in India and abroad. I write about the numbers, rules and decisions that quietly decide whether businesses last. Follow me here on Medium for the rest of this 100-part series.</em></p><p><em>This article is general information, not legal, tax or investment advice. Rules change; check current provisions or speak to a qualified professional before acting. Examples labelled illustrative are composites.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f9aa52ed2309" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>