<?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 IFRS-LABO on Medium]]></title>
        <description><![CDATA[Stories by IFRS-LABO on Medium]]></description>
        <link>https://medium.com/@contact_34117?source=rss-7023d46147b2------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*ldVnv_pA-jAwGqP8oVlEDA.png</url>
            <title>Stories by IFRS-LABO on Medium</title>
            <link>https://medium.com/@contact_34117?source=rss-7023d46147b2------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 10:19:43 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@contact_34117/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[Beyond Vibe Coding — Vol. 1]]></title>
            <link>https://medium.com/@contact_34117/beyond-vibe-coding-vol-1-f0dfb39f821f?source=rss-7023d46147b2------2</link>
            <guid isPermaLink="false">https://medium.com/p/f0dfb39f821f</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[ifr]]></category>
            <category><![CDATA[cpa]]></category>
            <category><![CDATA[vibe-coding]]></category>
            <category><![CDATA[accounting]]></category>
            <dc:creator><![CDATA[IFRS-LABO]]></dc:creator>
            <pubDate>Thu, 03 Sep 2026 05:20:40 GMT</pubDate>
            <atom:updated>2026-09-03T05:20:40.734Z</atom:updated>
            <content:encoded><![CDATA[<h3>AI Made Coding Cheaper. Specification Became the Moat.</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DUfztcTUAR-54wJoU-h0Lw.png" /></figure><p>I built IFRS-OneQ, a multilingual IFRS education app, on a Surface Go2</p><p>A 2020 model. 8GB RAM. An SD card for extra storage. Total AI spend to launch: $20 a month. One tool. Cursor. Everything else — Gemini, GPT, Claude — was free.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_ZI-RakUxrnL6qsq6zC2Tg.jpeg" /></figure><p>While the internet was busy debating M3 Max vs. M4, optimal VRAM, and which local LLM benchmark ran fastest — I shipped an app to the Google Play Store.</p><p>On that same machine, I also wrote a newsletter that reaches subscribers in 34 countries, and built a company from scratch.</p><p>The bottleneck was never the machine. It was never the model. It was never the budget.</p><p>AI made coding cheaper. It did not make product building cheap.</p><h3>What I Actually Did</h3><p>Let me be precise: I did not “vibe code.”</p><p>I wrote user stories. I defined acceptance criteria. I ran product reviews. I treated my AI tools like a development team — and myself like the product owner.</p><p>But here’s the part people usually skip over: I didn’t use one AI. I used a team of AIs. Each with a different role, matched to what that model actually does well.</p><p><strong>Gemini</strong> handled research. Deep knowledge retrieval, domain verification, technical context. When I needed to confirm how a standard worked or cross-check an edge case, Gemini was the right tool — broad knowledge, reliably thorough. Free tier. Sufficient.</p><p><strong>GPT</strong> handled the surface layer. Marketing images, SEO copy, launch assets. Creative output where polish matters more than depth. Free tier. Sufficient.</p><p><strong>Claude</strong> acted as the technical reviewer — someone who understands the code, but whose job is to evaluate, not to build. Neutral, no agenda, no cheerleading. Free tier. Sufficient.</p><p><strong>Cursor</strong> handled implementation. $20 a month. The only tool I paid for.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*E-vInbxbyZPdF2wvItekVA.png" /></figure><p>The workflow was closer to a sprint cycle than a prompt session. Before Cursor wrote a single line of code, it had to explain the plan — what it was going to build, how, and why. Claude reviewed that plan. It usually caught one of three things: hidden complexity, vague acceptance criteria, or scope drift. Then I made the call: approve, revise, or reject.</p><p>I didn’t design this workflow deliberately. It just felt natural — because separating builder from reviewer is Product Management. I’ve been doing it with humans for 20 years. When AI tools became capable enough, I applied the same discipline. The instinct to assign roles, to not let the builder also be the reviewer, to require a plan before execution — that came from two decades of product and audit work, not from reading about AI workflows.</p><p>That is not prompting.</p><p>That is product management — applied to an AI team.</p><p>But I should be honest about something: it wasn’t clean. The machine crashed. A lot. And how I worked through those crashes taught me something about AI development that no amount of spec writing could have.</p><p>More on that in Vol. 2. For now — the philosophy.</p><h3>The Wrong Competition</h3><p>Most of the AI development conversation is still focused on the wrong layer.</p><p>Which model. Which prompt structure. Which machine. How much VRAM. Whether to run local or in the cloud.</p><p>These are real questions. They are just not the first question.</p><p>The first question is: <em>what are you building, and do you understand the problem deeply enough to specify it well?</em></p><p>That is the real bottleneck. Not coding. Not prompting. Specification.</p><p>Because “what to build” is not an information problem. It is a judgment problem. And judgment only becomes useful when it can be turned into specification.</p><p>That is where most AI-first product conversations quietly break down.</p><h3>AI Can Retrieve Patterns. It Cannot Retrieve Pain.</h3><p>AI has access to public information. Public information is mostly success.</p><p>Launches. Feature pages. Case studies. Best practices. Product teardown threads. The polished version of what worked.</p><p>What it rarely sees is failure.</p><p>The features that benchmarked well but solved nothing. The products that looked right but felt wrong. The workflows users abandoned without ever filing a ticket. The assumptions that seemed obvious until they hit production.</p><p>That knowledge rarely gets published. It lives in operators’ heads. In edge cases. In scars. In the quiet reasons experienced people don’t trust certain workflows, even when the UI looks correct.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eTPiWKBq33czsWeGK3ydDw.png" /></figure><p>Which means AI can retrieve patterns. It cannot retrieve pain.</p><p>And product specification is mostly the work of translating pain into structure.</p><p>Building from public information alone produces a product that looks like every other successful product — because that’s all the model has ever seen. Competent execution of someone else’s already-validated idea.</p><p>That is not a product. That is a benchmark.</p><p>I’ve spent two decades in accounting, audit, and financial product management. I know exactly where IFRS practitioners lose confidence. I know which concepts the textbooks explain poorly. I know the moment a junior accountant hits a wall — because I’ve been that person, and I’ve watched hundreds of others hit the same wall.</p><p>No prompt retrieves that. No benchmark captures it.</p><p>That is the moat.</p><p>Not writing code. Not generating UI. Not prompting faster.</p><p>Knowing where reality breaks — and being able to write that down clearly enough for a machine to build against it.</p><h3>Why Domain Expertise Matters More Now, Not Less</h3><p>This is the shift that gets missed in the excitement around AI development tools.</p><p>AI does not eliminate expertise. It raises the market value of expertise that can be specified.</p><p>For most of software history, domain experts needed engineers to realize their ideas. That translation layer was expensive, slow, and lossy. Context leaked. Edge cases disappeared. Ambiguity got implemented as false certainty.</p><p>AI compresses that gap.</p><p>But compression only helps if the person holding the problem can specify it. The advantage doesn’t go to the person with the best prompt. It goes to the person who understands the problem well enough to define what “done” actually means.</p><p>The accountant who can turn accounting judgment into acceptance criteria. The logistics manager who can turn operational edge cases into product rules.</p><p>These people don’t need to win the model benchmark competition. They don’t need the M4. They need to bring what AI cannot replace: years of exposure to a specific problem, and the ability to write down what they know.</p><p>The more AI handles implementation — the more valuable deep domain knowledge becomes.</p><p>Not less. More.</p><h3>One More Thing Before Vol. 2</h3><p>I said the bottleneck was never the machine.</p><p>That is true at the level of philosophy. At the level of practice, the machine pushed back — hard. And working through that taught me something I didn’t expect: that the habits I’d built over 20 years in audit and product management weren’t just useful for writing specifications. They were useful for debugging. For questioning AI outputs. For knowing when to push back and when to trust.</p><p>Vol. 2 is about the crashes. What caused them, how I worked through them, and why a weak machine might be the best teacher you can have.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f0dfb39f821f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Beyond Vibe Coding — Vol. 2 Disciplining Vibe Coding: An Auditor’s Approach]]></title>
            <link>https://medium.com/@contact_34117/beyond-vibe-coding-vol-2-disciplining-vibe-coding-an-auditors-approach-093cac8f8d67?source=rss-7023d46147b2------2</link>
            <guid isPermaLink="false">https://medium.com/p/093cac8f8d67</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[auditor]]></category>
            <category><![CDATA[vibe-coding]]></category>
            <category><![CDATA[cpa]]></category>
            <dc:creator><![CDATA[IFRS-LABO]]></dc:creator>
            <pubDate>Thu, 03 Sep 2026 05:19:40 GMT</pubDate>
            <atom:updated>2026-09-03T05:19:40.014Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ogwX7eoqCVdBEfv4yXIFiA.png" /></figure><p>In <a href="https://ifrs-labo.com/">Vol. 1</a>, I mentioned that I built IFRS-OneQ on a Surface Go2. 8GB RAM. $20/month. One paid tool.</p><p>What I didn’t mention: it crashed. A lot.</p><p>This is the part I left out of Vol. 1. But it turned out to be the most important part. If you have started building something with AI, you will likely hit the same walls. Here is what I learned from each one.</p><h3>First Crash: Storage</h3><p>I started the way most people do. I had a clear image of what I wanted to build, so I just started — project files, AI context, data, everything on the C drive. Pure vibe.</p><p>I had a feeling it wouldn’t last. Within a few days, it froze.</p><p>After some investigation — asking both AI tools and checking the system state — the answer was straightforward: storage shortage. The Surface Go2 was not designed to carry that much locally.</p><p>The fix was obvious. Move everything to the SD card.</p><p>Simple tasks started running again. Then it froze again.</p><h3>Second Crash: Memory — and a Deeper Lesson</h3><p>The SD card solved the storage problem. It did not solve the processing load.</p><p>Even with more space, the AI kept stalling on certain tasks. So I started tracking what the AI was reading, and why it was freezing.</p><p>The problem was not the code.</p><p><strong>The problem was what I was handing the AI.</strong></p><p>I was feeding the AI hundreds of IFRS practice questions — large, dense, highly structured exam content. But the AI did not need to read the content of those questions to write application logic. It needed the schema, the rules, the behavior. Not the dataset itself.</p><p>When you hand an AI a full dataset, it tries to understand and optimize for the content. Processing load spikes. And then something worse happens.</p><p><strong>The AI starts making judgment calls that should be yours.</strong></p><p>When I handed over everything — including decisions about what mattered, what to emphasize, how to structure the content — the AI began researching complex patterns, weighing priorities, and sometimes making the wrong call. Then it would try to fix its own mistakes, consuming even more processing.</p><p>The results were telling. Sometimes the AI would quietly “summarize” carefully constructed question data. A dataset of 100 questions would silently become 95. Features I had built would be mysteriously disabled. Multilingual support — which was non-negotiable for IFRS-OneQ — kept getting broken every time I added something new.</p><p>This was the deeper lesson.</p><p>The moment you hand the AI not just a task, but the responsibility to decide what matters — it conflates structural understanding with domain judgment. And it either freezes, or produces something that ignores your intent entirely.</p><p>A higher-spec machine might have powered through the freeze. But the result would have been a product built on someone else’s priorities — or no one’s.</p><p>The constraint forced me to be precise about what the AI actually needed.</p><p><strong>It is not about what you feed the AI. It is about what you don’t.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Jk2ijIiQDNbsEpfXGh4sHQ.png" /></figure><p>Not just data. Judgment. Responsibility. Authority over what the product becomes.</p><p>I removed the question content entirely from the AI’s context. The freezing largely stopped.</p><h3>Third Crash: Old Memory</h3><p>One more problem emerged as development progressed.</p><p>Performance started degrading again — not because the AI was reading too much new information, but because it was <strong>carrying too much old information</strong>.</p><p>Old outputs. Old logic. Old relationships between parts of the codebase that no longer reflected where the code actually stood.</p><p>The tool was not overloaded with new input. It was overloaded with <strong>stale context</strong>.</p><p>Think of it like a phone that slows down when the cache fills up. An AI carrying old relationships starts taking longer to make new decisions.</p><p>The fix was to reset Cursor’s Codebase Index — deleting it and resyncing from scratch. The Index is how Cursor embeds and understands your codebase. Over time, as code changes accumulate, those embeddings carry stale relationships: associations from earlier logic that no longer reflects reality.</p><p>After the reset, performance noticeably improved.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XqPuu-L7cw03VkDCklV8mg.png" /></figure><p>To be clear: I don’t know if this applies to other AI tools — the tools keep evolving and I can’t guarantee it reproduces today. But the principle felt clear:</p><p><strong>Don’t let the AI remember too much. Old context is noise. Reset the Index.</strong></p><h3>Debugging Like an Auditor</h3><p>There is one more habit worth naming, because it shaped how I worked through every one of these crashes.</p><p>When something broke, I did not paste the error into the AI and do what it said.</p><p>I asked why.</p><p>If the explanation didn’t make sense — if I couldn’t follow the reasoning, or something felt off — I pushed back. <em>Why is that the cause? Why does that fix it? What assumptions are you making?</em></p><p>This instinct came from my time at an audit firm. I had a manager with a particular habit. I would give an explanation. He would say: “I see. And why?” I would answer again. He would say: “Okay. And why?”</p><p>It never stopped at the first answer. The point was not to find the answer. The point was to reach the truth beneath the answer. In audit, surface-level understanding is not enough. You are accountable for the reasoning, not just the conclusion.</p><p>I applied the same discipline to AI.</p><p>For error diagnosis, logs went in as raw text — lightweight, unambiguous, precise. For visual errors, Gemini handled image input well and was more generous with volume at the time, though that may have changed.</p><p>But the core habit was consistent: receive the answer, then ask why.</p><p>Most AI responses are correct at the surface. The ones that aren’t tend to sound just as confident as the ones that are. The only way to tell the difference is to keep asking until you hit something you can verify — or until the reasoning breaks down.</p><p>That is not skepticism for its own sake.</p><p><strong>It is audit thinking applied to a development workflow. Don’t accept the answer. Understand it.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JIN-XO2Yvr0NoXbn9QgPDw.png" /></figure><h3>What the Crashes Actually Taught Me</h3><p>All three crashes pointed to the same discipline:</p><p>Be deliberate about what information the AI holds at any given moment — and equally deliberate about what authority you do not hand over.</p><p>A high-spec machine might have let me be lazy. I could have fed the AI everything and let processing power compensate for poor information design. It probably would have produced something functional.</p><p>But functional is not the same as intentional. The product that comes out of undisciplined AI input reflects no one’s priorities clearly — not yours, not your users’. It becomes something that roughly resembles what you wanted, built on a foundation no one thought carefully about.</p><p>The Surface Go2 did not allow that.</p><h3>What That Machine Built</h3><p>Here is what that Surface Go2 produced — while I was working full-time.</p><p><a href="https://ifrs-labo.com/posts/ifrs-oneq"><strong>IFRS-OneQ</strong> </a>— Multilingual IFRS education app, <em>now live on </em><a href="https://play.google.com/store/apps/details?id=com.ifrslabo.ifrsoneq"><em>Google Play</em></a><em> and </em><a href="https://ifrs-oneq.ifrs-labo.com/"><em>Web</em></a><em>.</em></p><p><a href="https://www.linkedin.com/build-relation/newsletter-follow?entityUrn=7442121907753132032"><strong>IFRS Deep Dive</strong></a> — a newsletter with subscribers in 30+ countries across 6 continents.</p><p><a href="https://ifrs-labo.com/"><strong>IFRS-LABO</strong></a> — the company behind both, built from scratch.</p><p>One Surface Go2. One SD card. $20 a month.</p><p>The bottleneck was never the machine.</p><p><em>Vol. 3 moves from design discipline to economic discipline. Building got cheaper. Maintaining did not. The question nobody asks before they build — and why it has shaped every major shift in enterprise software history.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=093cac8f8d67" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[IFRS 18 Early Adoption Database: What the First Real-World Cases Are Telling US]]></title>
            <link>https://medium.com/@contact_34117/ifrs-18-early-adoption-database-what-the-first-real-world-cases-are-telling-us-fcc33254eb55?source=rss-7023d46147b2------2</link>
            <guid isPermaLink="false">https://medium.com/p/fcc33254eb55</guid>
            <category><![CDATA[ifrs18]]></category>
            <category><![CDATA[pmm]]></category>
            <category><![CDATA[case]]></category>
            <category><![CDATA[early-adoption]]></category>
            <category><![CDATA[disclosure]]></category>
            <dc:creator><![CDATA[IFRS-LABO]]></dc:creator>
            <pubDate>Tue, 18 Aug 2026 15:15:23 GMT</pubDate>
            <atom:updated>2026-08-18T15:15:23.150Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*waS7d6Ji4GEnBVffroVItg.png" /><figcaption>IFRS18 Early Adoption Database</figcaption></figure><h3>IFRS 18 is beginning to move from written requirements into real-world reporting practice.</h3><p>Companies have already begun to early adopt the new standard. And now we are entering an important new phase: <strong>IFRS 18 financial information subject to independent auditor review is beginning to emerge.</strong></p><p>Earlier examples included voluntary disclosures and preliminary financial information. These remain useful, but they are not necessarily comparable with financial information that has been subject to an auditor’s review.</p><p>At IFRS-LABO, we have been tracking IFRS 18 early adoption cases globally and analysing the underlying financial statements and corporate disclosures.</p><p>We have now organised the results into the <strong>IFRS 18 Early Adoption Database</strong>.</p><p><a href="https://ifrs-labo.com/ifrs18">IFRS 18 Early Adoption Database &amp; Implementation Resources | IFRS Labo</a></p><h3>IFRS 18 Early Adoption Database</h3><p>The database goes beyond a simple list of companies that have announced early adoption.</p><p>For each company, we provide basic information such as country, principal business and company size, together with key attributes of the IFRS 18 financial information itself, including:</p><ul><li>whether the reporting period is annual or interim;</li><li>whether the financial information is audited, reviewed, or not subject to external assurance;</li><li>management-defined performance measures (MPMs);</li><li>whether an MPM reconciliation is provided;</li><li>how foreign exchange differences are disaggregated;</li><li>function-by-nature expense disclosures; and</li><li>other notable implementation features.</li></ul><p>In addition to these structured data points, we review the underlying financial information and provide a short summary of features that we consider particularly relevant to IFRS 18 implementation.</p><p><strong>Explore the IFRS 18 Early Adoption Database → </strong><a href="https://ifrs-labo.com/ifrs18">https://ifrs-labo.com/ifrs18</a></p><p>The database also provides <strong>direct links to the original financial statements and source documents</strong> for each case, so readers who want to examine any of the disclosures discussed in this article in greater detail can access the primary sources through the database.</p><h3>Behind the database</h3><p>Building this database was not as simple as asking AI to identify IFRS 18 early adopters.</p><p>There were particular challenges in identifying new cases, locating and assessing the relevant primary-source documents, and analysing accounting practices for which very little real-world precedent yet exists. <strong>AI alone was not enough to build this database.</strong></p><p>We will discuss how we approached this challenge separately in a future article.</p><p>For now, let’s focus on what the actual IFRS 18 disclosures are telling us.</p><h3>First observation: auditor review makes a difference</h3><p>One of the strongest impressions from our research so far has been the difference between financial information <strong>with and without auditor involvement</strong>.</p><p>This distinction requires some care.</p><p>Unaudited financial information is not necessarily deficient. Earnings releases, voluntary disclosures and interim financial statements may serve different purposes and be subject to different disclosure requirements.</p><p>Nevertheless, when early adoption cases are used as benchmarks for IFRS 18 implementation, the nature of the underlying document matters.</p><p>In the cases we reviewed, financial information subject to auditor review generally provides a more complete picture of IFRS 18 implementation.</p><p>The difference is particularly noticeable in areas such as MPM reconciliations.</p><p>For this reason, the IFRS 18 Early Adoption Database separately identifies the <strong>assurance status</strong> of each case as <strong>Audit, Review or No external assurance</strong>.</p><p>Among the cases currently available, <strong>Kao, Kagome and Vodafone Qatar</strong> provide particularly useful examples because their IFRS 18 financial information has been subject to auditor review.</p><p>And once we look inside those disclosures, some interesting differences emerge.</p><h3>1. MPMs: real-world implementation is beginning to emerge</h3><p>Management-defined performance measures are naturally one of the most closely watched areas of IFRS 18.</p><p>Most of the early adopters identified in our research use measures that are treated as MPMs.</p><p>What differs considerably is the level of supporting disclosure.</p><p>Some unaudited early adoption disclosures identify an MPM without providing the same level of reconciliation information seen in reviewed financial information.</p><p>By contrast, the reviewed cases provide much more useful evidence of how the new requirements may actually operate in practice.</p><p>Among them, <strong>Vodafone Qatar and Kao broadly illustrate the type of MPM reconciliation many practitioners might have expected from reading IFRS 18.</strong></p><p>Kagome is different.</p><h3>Kagome: an MPM reconciliation starting from gross profit</h3><p>Kagome provides one of the most distinctive IFRS 18 MPM examples we have identified so far.</p><p>Kagome uses a performance measure structured broadly as follows:</p><p><strong>Gross profit<br>– Selling, general and administrative expenses<br>= Core operating income</strong></p><p>The resulting MPM reconciliation therefore effectively starts from <strong>gross profit</strong>.</p><p>At first sight, this may look unusual.</p><p>When practitioners think about an IFRS 18 MPM reconciliation, a natural model is to start with an IFRS-defined subtotal such as operating profit and reconcile that subtotal to an adjusted management measure.</p><p>Kagome demonstrates that this need not necessarily be the only possible structure.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/1*egtY5pbNEcnxaYwasuyuQg.png" /><figcaption>MPM Adjustments</figcaption></figure><p>Based on our reading of IFRS 18, <strong>this type of approach appears possible under the requirements of the standard.</strong></p><p>And that could have important practical implications.</p><h3>Could existing intermediate profit measures survive IFRS 18?</h3><p>Many companies have traditionally managed and communicated performance using several levels of profit.</p><p>A company might, for example, focus internally on gross profit, an operating-type measure and other intermediate profit measures.</p><p>For such companies, one practical question arising from IFRS 18 implementation is whether existing management measures need to be abandoned or substantially redesigned.</p><p>Kagome’s example suggests another possibility.</p><p><strong>An existing intermediate profit measure may, depending on the circumstances, be retained as an MPM, with the reconciliation designed around that measure.</strong></p><p>If this approach proves applicable more broadly, it could provide useful flexibility for companies seeking to preserve management measures already embedded in internal reporting and external communication.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XO60SWQ7-CFtsPVFo0ytIg.png" /></figure><p>In that sense, Kagome’s disclosure is more than simply an unusual MPM example.</p><p>It may provide an early real-world indication that IFRS 18 allows greater flexibility in preserving existing performance measures than might initially be assumed.</p><h3>But how are the tax and NCI effects calculated?</h3><p>That flexibility immediately creates another practical question.</p><p>IFRS 18 requires an MPM reconciliation to disclose the income tax effect and the effect on non-controlling interests of the reconciling items.</p><p>Once a broad expense category such as SG&amp;A becomes part of the reconciliation, those calculations could become complex.</p><p>Consider a multinational group.</p><p>SG&amp;A may arise across numerous subsidiaries in different jurisdictions, subject to different tax rates and, in some cases, different NCI ownership percentages.</p><p>One possible approach would be to identify the relevant SG&amp;A by group company, apply the applicable statutory tax rates and calculate the corresponding NCI effects based on ownership percentages.</p><p>But how granular does the calculation need to become?</p><p>For example, does the calculation reflect more detailed tax-effect considerations, including recoverability, at the individual adjustment level?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*R3AYrDz2-dcSR_a-bLfL0g.png" /></figure><p>The published disclosure does not allow us to determine Kagome’s entire internal calculation process.</p><p>Some form of practical or simplified calculation methodology may therefore be being used.</p><p>And this is where the Kagome example becomes particularly interesting.</p><h3>This MPM appears in reviewed financial information</h3><p>Kagome’s financial information containing this MPM structure and reconciliation was subject to an independent auditor’s review.</p><p>This should not be overstated.</p><p>A review conclusion is not an audit opinion, nor does it mean that the auditor has separately endorsed every individual accounting judgement or internal calculation methodology used by management.</p><p>Nevertheless, <strong>from an implementation perspective, the fact itself is significant.</strong></p><p>This is not merely a theoretical interpretation of IFRS 18.</p><p>An MPM structured in this way, together with the required reconciliation, has actually appeared in IFRS 18 financial information that has gone through the auditor review process.</p><p>For companies considering how to preserve existing intermediate profit measures under IFRS 18, that makes Kagome an important case to study.</p><p>The example suggests that there may be meaningful flexibility in designing MPMs.</p><p>The potential trade-off is the practical complexity created by the corresponding tax and NCI calculations.</p><p>And from a preparer’s perspective, one question remains particularly interesting:</p><p><strong>How is this reconciliation actually calculated in practice?</strong></p><h3>2. Foreign exchange differences: this looks like a practical implementation requirement</h3><p>Foreign exchange differences provide a different lesson.</p><p>Across the early adoption cases reviewed by IFRS-LABO, companies generally <strong>disaggregate foreign exchange differences where FX effects arise from transactions belonging to different IFRS 18 categories</strong>.</p><p>There are cases where companies indicate that further disaggregation is not necessary.</p><p>However, the cases we have seen appear to reflect the underlying business structure — situations where significant FX effects do not arise across multiple activities — rather than a general approach of leaving foreign exchange differences aggregated.</p><p>The emerging practical message is therefore relatively clear.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*8n6ZG1j5yi6PrFH5TH3xnQ.png" /></figure><p><strong>For many companies, FX disaggregation is likely to become a core IFRS 18 implementation requirement.</strong></p><p>And this may be more important than it first appears.</p><h3>This is also a systems issue</h3><p>The accounting principle itself is relatively straightforward.</p><p>Foreign exchange differences need to be classified consistently with the income and expenses arising from the underlying items.</p><p>In practice, this means that a company may need to distinguish FX effects associated with operating, investing and financing activities.</p><p>The real question is how that information is captured.</p><p>If foreign exchange gains and losses arising from different activities have historically been accumulated in the same accounts, attempting to separate them only when preparing the IFRS 18 financial statements may require the finance team to trace transactions and reconstruct the underlying sources of FX movements accumulated throughout the period.</p><p>For a company with significant transaction volumes, that exercise could become substantial.</p><p>The practical lesson from the early adoption cases is therefore:</p><p><strong>Do not treat FX disaggregation merely as a year-end financial statement presentation issue.</strong></p><p>Companies preparing for IFRS 18 should consider whether their chart of accounts, sub-accounts, transaction attributes or other system data can identify foreign exchange differences according to the underlying activity <strong>from the beginning of the reporting period</strong>.</p><p>Otherwise, a relatively simple-looking presentation requirement could create a significant manual workload at the reporting date.</p><h3>3. Function-by-nature disclosures: the important examples are still to come</h3><p>The third area we are tracking is the disclosure of expenses by nature for companies that present operating expenses by function.</p><p>This could become one of the more operationally demanding aspects of IFRS 18, particularly where the necessary nature-based information is not readily available within existing reporting systems.</p><p>Interestingly, the current interim disclosures from <strong>Kao and Kagome already provide a glimpse of how this might work in practice</strong>.</p><p>Both companies present expenses by function in the statement of profit or loss, including <strong>cost of sales and selling, general and administrative expenses (SG&amp;A)</strong>. They then provide additional nature-based information for SG&amp;A in the notes.</p><p>For example, the SG&amp;A disclosures include items such as employee-related costs, depreciation and amortisation, as well as other nature-based expense categories.</p><p>However, <strong>neither company provides an equivalent nature-based breakdown of cost of sales in its current interim financial information</strong>.</p><p><a href="https://ifrs-labo.com/posts/ifrs18-function-by-nature-matrix">IFRS 18 Operating Expenses: Why the Function-by-Nature Matrix Is Harder Than It Looks</a></p><p>This is an interesting observation, but it is too early to draw a conclusion from it.</p><p>The additional nature-based expense disclosures introduced by IFRS 18 are principally relevant to annual financial statements and are not necessarily required in the same form in interim reporting. Accordingly, the current interim disclosures do not tell us whether Kao and Kagome intend to follow the same approach in their annual IFRS 18 financial statements.</p><p><strong>Will they continue to provide nature-based information primarily for SG&amp;A, or will additional information relating to cost of sales also appear at year-end?</strong></p><p>We do not yet know.</p><p>This is why the first annual IFRS 18 financial statements will be particularly important.</p><p>They should provide much better evidence of how companies interpret and implement the new expense disclosure requirements in practice.</p><p>And, as with FX disaggregation, this is not merely a disclosure question.</p><p>Producing the required information may involve mapping expenses across functions, expense natures, entities and reporting systems. Depending on a company’s existing accounting structure, this could become a significant data and systems exercise.</p><p>For now, therefore, the conclusion is simple:</p><p><strong>Kao and Kagome give us an early preview — but the full picture will only emerge with the annual financial statements.</strong></p><p>IFRS-LABO will continue to track these developments.</p><h3>What the early adoption cases are telling us</h3><p>The population of IFRS 18 early adopters is still small.</p><p>It is far too early to identify established market practice.</p><p>Nevertheless, comparing the actual disclosures already provides some useful implementation lessons.</p><p><strong>First, the nature of the financial information matters.</strong> Financial information subject to auditor review generally provides a more complete view of IFRS 18 implementation than preliminary or unaudited disclosures. When using early adopters as benchmarks, the assurance status of the underlying information is therefore an important attribute.</p><p><strong>Second, MPMs may offer more flexibility than expected.</strong> Kagome’s unusual gross-profit-based reconciliation suggests that existing intermediate profit measures may potentially be preserved within the IFRS 18 framework. But that flexibility can create practical questions, particularly around the calculation of tax and NCI effects.</p><p><strong>Third, some apparently presentational requirements are really data and systems issues.</strong> FX disaggregation is a good example. The early cases suggest that disaggregation will be necessary for many companies, and waiting until the reporting date to determine where FX differences originated could create a significant manual burden.</p><p>For function-by-nature disclosures, meanwhile, the evidence is not yet available. The first annual IFRS 18 financial statements should tell us considerably more.</p><p>These are exactly the types of issues that are difficult to identify by reading the standard alone.</p><p><strong>They become visible when actual financial statements are compared across companies.</strong></p><p>That is one of the reasons we built the IFRS 18 Early Adoption Database.</p><h3>More to come</h3><p>These are exactly the kinds of implementation issues that are difficult to see from the standard alone. They become visible when actual financial statements are compared across companies.</p><p>That is why we built the IFRS 18 Early Adoption Database.</p><p>We will continue updating the database as new cases emerge, with particular attention to the first annual IFRS 18 financial statements.</p><p><strong>Explore the IFRS 18 Early Adoption Database → </strong><a href="https://ifrs-labo.com/ifrs18"><strong>https://ifrs-labo.com/ifrs18</strong></a></p><p><a href="https://ifrs-labo.com/ifrs18">IFRS 18 Early Adoption Database &amp; Implementation Resources | IFRS Labo</a></p><h3>Disclaimer</h3><p>This article and the IFRS 18 Early Adoption Database are provided for informational purposes only and do not constitute professional advice.</p><p>The analysis is based on publicly available information as of the date of publication and reflects IFRS-LABO’s interpretation. Readers should refer to the original financial statements and applicable IFRS Accounting Standards when making accounting or reporting decisions.</p><p>© IFRS LABO LLC. All rights reserved.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=fcc33254eb55" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>