<?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 Krishnakant Singh on Medium]]></title>
        <description><![CDATA[Stories by Krishnakant Singh on Medium]]></description>
        <link>https://medium.com/@krishnakant-singh?source=rss-3054102b4f30------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*i4_ksPbdh3SaTR0jccDWRA.jpeg</url>
            <title>Stories by Krishnakant Singh on Medium</title>
            <link>https://medium.com/@krishnakant-singh?source=rss-3054102b4f30------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 00:33:48 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@krishnakant-singh/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[When a Good Customer Experience Becomes a Bad Business Decision?]]></title>
            <link>https://medium.com/@krishnakant-singh/when-a-good-customer-experience-becomes-a-bad-business-decision-f67a1243029f?source=rss-3054102b4f30------2</link>
            <guid isPermaLink="false">https://medium.com/p/f67a1243029f</guid>
            <category><![CDATA[business-decisions]]></category>
            <category><![CDATA[customer-support]]></category>
            <category><![CDATA[cx-strategy]]></category>
            <category><![CDATA[customer-experience]]></category>
            <dc:creator><![CDATA[Krishnakant Singh]]></dc:creator>
            <pubDate>Thu, 12 Nov 2026 12:40:25 GMT</pubDate>
            <atom:updated>2026-11-12T12:42:21.261Z</atom:updated>
            <content:encoded><![CDATA[<p>We often talk about going the extra mile for customers. Waiving a fee, offering a refund, making an exception to a policy or arranging something that wasn’t originally promised.</p><p>These gestures can turn an unhappy customer into a happy one. And honestly, there is nothing wrong with that.</p><p>But have you ever wondered what happens when a company starts doing this regularly?</p><p>Let’s say a customer requests a refund even though the transaction doesn’t qualify under the company’s refund policy. The support agent, wanting to avoid a poor experience, makes an exception and processes the refund.</p><p>The customer is happy. The issue is resolved. Everyone moves on.</p><p>But what if the same situation happens 500 times a day?</p><p>What if the company starts spending money on refunds that it was never supposed to offer? What if customers begin expecting these exceptions? And what if the same resources could have been used to solve a much bigger problem affecting thousands of customers?</p><p>This is where things get interesting.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pqiBvjtnwsvTaA3geL-l4w.png" /><figcaption>(AI Generated Image)</figcaption></figure><h3>When customer satisfaction becomes the only goal</h3><p>I think one of the mistakes companies make is treating customer satisfaction as the ultimate measure of a good customer experience.</p><p>It isn’t always that simple.</p><p>Sometimes, the easiest way to make a customer happy is not necessarily the right decision for the business.</p><p>Consider expedited deliveries. A customer wants an order delivered earlier than promised. The company arranges it to ensure a good experience. But if fulfilling that request means disrupting other deliveries or adding significant operational costs, was it really the best decision?</p><p>Or take manual interventions. A customer has an unusual request, and the support team goes out of its way to resolve it. That’s great. But if hundreds of similar requests require manual intervention every day, perhaps the problem isn’t with the support team. Perhaps the process itself needs to change.</p><p>The point isn’t that companies should stop going the extra mile. It’s that we need to look beyond the immediate interaction.</p><h3>One happy customer versus a thousand affected customers</h3><p>There’s another side to this that we often overlook.</p><p>A decision that makes one customer happy can sometimes make things worse for several others.</p><p>When operational resources are limited, prioritising one request means something else might get delayed. When companies keep making exceptions, they might unintentionally create inconsistencies in the experience they offer to different customers.</p><p>And customers notice these inconsistencies.</p><p>This is why I believe customer experience cannot be evaluated only by looking at individual interactions. We also need to understand the impact of those decisions when they are repeated at scale.</p><h3>So, should we stop making exceptions?</h3><p>Not at all.</p><p>There are situations where making an exception is absolutely the right thing to do. Policies are important, but they shouldn’t prevent companies from recognising genuine customer problems.</p><p>However, there’s a difference between making an exception and building an entire process around exceptions.</p><p>If the same problem keeps coming back, perhaps we should stop treating it as an individual customer issue and start looking at why it’s happening in the first place. <a href="https://medium.com/@krishnakant-singh/why-great-customer-experience-begins-before-customers-need-support-33e827efa24c">Issue prevention</a> plays a crucial role here.</p><p>Sometimes, the right answer is a refund. Sometimes, it’s fixing a process. And sometimes, it’s simply being transparent with the customer about what the company can and cannot do.</p><p>I don’t think a company becomes customer-centric by saying yes to everything.</p><p>It becomes customer-centric when it understands what its customers need, makes sensible decisions and delivers a consistent experience without making the business unsustainable.</p><p>After all, the objective isn’t to make one customer happy at any cost. It’s to build a business that can keep making its customers happy, tomorrow and beyond.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f67a1243029f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[I Went on Medium to Read. I Ended Up Playing a Game.]]></title>
            <link>https://medium.com/@krishnakant-singh/i-went-on-medium-to-read-i-ended-up-playing-a-game-994312b5463c?source=rss-3054102b4f30------2</link>
            <guid isPermaLink="false">https://medium.com/p/994312b5463c</guid>
            <category><![CDATA[games]]></category>
            <category><![CDATA[good-words]]></category>
            <category><![CDATA[new-features]]></category>
            <category><![CDATA[try-new-things]]></category>
            <category><![CDATA[game-medium]]></category>
            <dc:creator><![CDATA[Krishnakant Singh]]></dc:creator>
            <pubDate>Fri, 16 Oct 2026 17:09:15 GMT</pubDate>
            <atom:updated>2026-10-16T17:12:29.119Z</atom:updated>
            <content:encoded><![CDATA[<p>I was on Medium recently when I stumbled upon something I wasn’t expecting: a word game.</p><p>It’s called <strong>Good Words</strong>, and the idea is pretty simple.</p><p>You get a 5×5 grid of letters and three minutes to find as many words as you can. You connect neighbouring letters, including diagonally, but you can’t use the same tile twice in one word. Each word needs to be at least four letters long, and longer words earn more points.</p><p>That’s it. No complicated rules. Just you, a grid of letters, and a timer that’s probably moving faster than you’d like.</p><p>I went in out of curiosity. But what caught my attention wasn’t just the game. It was the fact that I was finding it on Medium.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JWM4dT-MlJyCF7dLoYbCDQ.png" /><figcaption>Good Words</figcaption></figure><h3>Why I found this interesting</h3><p>I’ve mostly thought of Medium as a place to read articles, discover new perspectives, and occasionally publish something of my own.</p><p>A word game feels like a slightly unexpected addition to that experience.</p><p>And I don’t necessarily mean that as a criticism.</p><p>Sometimes, the most interesting product additions are the ones that make you pause and think, “Oh, I didn’t expect that here.”</p><p>Good Words is a small example of that. You don’t need to commit to reading a long article or even have a particular topic in mind. You can just open the game and play a quick round.</p><p>There’s also an interesting detail in how it works: the board changes every three and a half minutes, and the instructions say that whoever is there plays the round with you.</p><p>So, while you’re playing on your own screen, you’re sharing the board with other players. It adds a social element to what would otherwise be a solitary word puzzle.</p><h3>The product question I’m left with</h3><p>Here’s what I’m curious about.</p><p>When does an extra feature make a platform more enjoyable, and when does it simply become another distraction?</p><p>There’s no universal answer. It depends on what people come to the platform for and whether the new feature complements that experience.</p><p>For some people, Good Words might be a fun little break between articles. For others, it might be something they try once and never open again.</p><p>And that’s okay. Not every feature needs to become everyone’s favourite.</p><p>The more interesting question is whether people find enough value in it to return.</p><p>A feature being easy to try is one thing. Becoming part of someone’s routine is another.</p><p>I don’t know whether <strong>Good Words </strong>will become a regular part of my Medium experience. But I like that it gave me something unexpected to explore.</p><p>And, admittedly, I did spend longer looking at that grid than I originally intended.</p><p><strong>Have you tried Good Words yet?</strong> What do you think of Medium adding games to the platform? Share your thoughts in the comments!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=994312b5463c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Your Customer Support Team Knows What’s Broken. Is Your Product Team Listening?]]></title>
            <link>https://medium.com/@krishnakant-singh/your-customer-support-team-knows-whats-broken-is-your-product-team-listening-82542ab176e1?source=rss-3054102b4f30------2</link>
            <guid isPermaLink="false">https://medium.com/p/82542ab176e1</guid>
            <category><![CDATA[tickets]]></category>
            <category><![CDATA[customer-engagement]]></category>
            <category><![CDATA[customer-experience]]></category>
            <category><![CDATA[product-management]]></category>
            <category><![CDATA[cx-strategy]]></category>
            <dc:creator><![CDATA[Krishnakant Singh]]></dc:creator>
            <pubDate>Mon, 28 Sep 2026 18:23:02 GMT</pubDate>
            <atom:updated>2026-09-28T18:23:02.401Z</atom:updated>
            <content:encoded><![CDATA[<p>In my 5+ years of working in customer experience, I have worked across different businesses and problem statements, from fintech to quick commerce.</p><p>The products were different. The customers were different. Even the problems I was solving were different.</p><p>But there was one thing I kept seeing.</p><p><strong>Customers were constantly telling us where they were struggling with the product.</strong></p><p>A part of my team’s work has been around working with Operations teams and building SOPs and processes for handling customer queries. So I have spent a lot of time looking at why customers contact us in the first place.</p><p>And over time, I started noticing that sometimes the question a customer asks isn’t really the problem.</p><h4>A Support Ticket Is More Than a Ticket</h4><p>I remember while working on the Digital Gold business at my organisation, we were getting quite a few customer queries around a very simple question:</p><p><strong>“How do I sell my gold?”</strong></p><p>The interesting part was that the option to sell Digital Gold already existed. There was a product flow for it.</p><p>So, at first, this could easily have been treated as a customer education problem. The Operations team could tell customers where to find the option, the SOP could have been updated, and the ticket could have been closed.</p><p>But over time, I’ve come to realise that a ticket is often just a symptom. It tells us that something has gone wrong, but not necessarily what went wrong underneath. To figure out what actually went wrong, I looked at the journeys of customers on the app before they had raised the tickets.</p><p>What I found surprised me.</p><p><strong>A large number of them weren’t even reaching the page where the option to sell their gold was available.</strong></p><p>The more I looked at the journey, the more I realised that this wasn’t really a question of whether customers knew how to sell their gold. They knew what they wanted to do; they just couldn’t find their way to it.</p><p>That made me look at the problem differently.</p><p>Instead of asking, “How do we help customers answer this query?”, the more important question became:</p><p><strong>“What is stopping customers from discovering this capability on their own?”</strong></p><p>The answer wasn’t another Support response. There was something in the product experience that was getting in the customer’s way.</p><p>This is why I have started looking at customer queries a little differently.</p><p>A Support ticket is usually treated as something that needs to be resolved. And of course, it does. The customer has a problem and someone needs to help them.</p><p>But when the same question keeps coming up, I think it is worth looking at what sits behind the question.</p><p><strong>Customers don’t usually tell us, “There is a discoverability issue in this journey.”</strong></p><p><strong>They say, “How do I sell my gold?”</strong></p><p>They tell us where they got stuck. It is up to us to understand why they got stuck there.</p><p>Sometimes the answer is obvious. Sometimes it isn’t.</p><p>And this is where I think Support can give Product teams something that is difficult to get from feedback surveys alone.</p><p>Not because Support has some special source of truth, but because it sits very close to the moments when customers get frustrated enough to ask for help.</p><h4>But There’s a Catch</h4><p>Of course, there is a catch.</p><p>Not every customer who faces a problem contacts Support. Some find a workaround. Some don’t bother. Some simply leave (<a href="https://medium.com/@krishnakant-singh/the-customer-who-never-complains-the-most-dangerous-customer-for-a-product-a6ba42702798">The Most Dangerous Customer for a Product</a>).</p><p>So Support data can never tell us the whole story.</p><p>For me, it is more like a starting point.</p><p>If I see the same issue coming up again and again, I want to understand what is happening in the product journey around it. I want to know whether the issue is with the product, the process, the communication or simply a one-off situation.</p><p>And this is also where I understand why Product teams can’t act on every customer problem.</p><p>There are business priorities. There are technical constraints. There are regulatory requirements. There are bigger product initiatives already in progress.</p><p><strong>A customer problem can be genuine and still not make it to the top of the roadmap.</strong></p><p>I don’t think that’s necessarily wrong.</p><p>What I do think is important is that we don’t lose the learning just because we aren’t fixing the problem immediately.</p><p>If the same problem keeps appearing, someone should be asking why.</p><p>Maybe the answer is a product change.</p><p>Maybe it is a process change.</p><p>Maybe it is simply that we haven’t explained something clearly enough.</p><p>And sometimes, after looking at the data, we might realise that the problem isn’t big enough to solve.</p><p>But at least then it becomes a conscious decision rather than a ticket that was closed and forgotten.</p><p>That, for me, is where the connection between Customer Experience and Product becomes interesting.</p><p>The Support team is dealing with the problem after it happens.</p><p>The Product team has the opportunity to change what happens before it does.</p><h4>Maybe the Goal Isn’t Fewer Tickets</h4><p>I don’t think the goal should be to make Support unnecessary. There will always be customers with unusual questions and situations that need human help.</p><p>But if customers keep contacting us about the same thing, I don’t want the answer to always be another SOP.</p><p>Sometimes, I want to go one step back and ask:</p><p><strong>Why did the customer need to contact us at all?</strong></p><p>Because that question can lead somewhere a ticket resolution never will.</p><p>And maybe that’s one of the most useful things a Support team can give a Product team:</p><p>not just a list of problems customers are reporting, but a better understanding of <strong>where the product itself is making customers ask for help.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=82542ab176e1" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI in Customer Experience]]></title>
            <link>https://medium.com/@krishnakant-singh/ai-in-customer-experience-16a2fafbac3e?source=rss-3054102b4f30------2</link>
            <guid isPermaLink="false">https://medium.com/p/16a2fafbac3e</guid>
            <category><![CDATA[ai-in-customer-experience]]></category>
            <category><![CDATA[customer-experience]]></category>
            <category><![CDATA[cx-strategy]]></category>
            <category><![CDATA[genai]]></category>
            <dc:creator><![CDATA[Krishnakant Singh]]></dc:creator>
            <pubDate>Fri, 18 Sep 2026 19:43:26 GMT</pubDate>
            <atom:updated>2026-09-18T19:47:11.783Z</atom:updated>
            <content:encoded><![CDATA[<p>Remember a time when you called a customer support number and, after perhaps a short wait, you actually got connected to a real human?</p><p>Sounds like a dream today, right? But this used to be reality.</p><p>Today, try calling customer support and there is a good chance that the first “person” you speak to won’t be a person at all. It might be an AI voice bot that understands what you’re saying, responds almost instantly, and tries to resolve your problem without involving a human agent.</p><p>And I am not saying that is necessarily a good or a bad thing.</p><p>Having worked on implementing AI voice bots myself, I understand why businesses are moving in this direction. The economics are compelling. AI can operate 24/7, handle large volumes simultaneously, reduce repetitive work for agents, and potentially bring down the cost of every customer interaction.</p><p>But there is another side to this story. <strong>Is an AI-powered customer experience actually better for the customer?</strong></p><p>That is the question I want to explore.</p><h4>No, Customers don’t hate AI. They hate bad experiences</h4><p>In my experience, customers don’t actually hate AI, but they hate bad experience. Recent research by Gartner suggests that customers are not necessarily opposed to AI.</p><p><a href="https://www.gartner.com/en/newsroom/press-releases/2026-08-04-gartner-survey-finds-87-percent-of-customers-say-companies-using-genai-for-customer-service-must-provide-access-to-a-human-agent0">A 2026 Gartner survey</a> of 3,566 customers found that <strong>50% felt their interactions were easier when companies used GenAI</strong>. More interestingly, 58% of customers who use GenAI said they had used it to <em>complete a task</em>, rather than simply obtain information.</p><p>This is an important distinction.</p><p>Customers don’t necessarily want to talk to a human. They want their problem solved.</p><p>But there is a catch.</p><p>The same Gartner research found that <strong>87% of customers consider access to a human essential when companies use GenAI for customer service.</strong></p><p><strong>So the question is not “ AI or human”? It is, “when should AI handle the customer, and when should a human?”</strong></p><h4>Not every conversation needs a human</h4><p>Consider a simple question: <em>“Where is my order?”</em></p><p>AI can answer that in seconds.</p><p>Now consider: <em>“My order was supposed to arrive yesterday. It still hasn’t been delivered, and I have already contacted support twice.”</em></p><p>That is a very different problem.</p><p>The customer doesn’t just need information. They need context, ownership and a meaningful resolution.</p><p>This is where the design of AI-powered CX becomes important.I see three levels of AI in customer experience:</p><p><strong><em>1. AI that answers</em></strong></p><p>“Your order is delayed. The new delivery date is tomorrow.”</p><p>Useful.</p><p><strong><em>2. AI that assists</em></strong></p><p>“Your order has missed the promised delivery time. Would you like me to arrange a refund or escalate the issue?”</p><p>Better.</p><p><strong><em>3. AI that acts</em></strong></p><p>“Your order has breached the promised delivery time. I’ve initiated the refund and raised escalated internally to make sure it doesn’t happen again. You don’t need to contact support again.”</p><p>Now that’s adds real value for customer.</p><p>The real promise of AI in CX isn’t that customers can talk to machines. <strong>It is that customers may no longer need to talk to anyone.</strong></p><h4>The future isn’t AI vs. Human</h4><p>I don’t think the future of customer experience is about choosing between AI and humans.</p><p>It is about knowing <strong>where each one adds the most value.</strong></p><p><strong>AI for scale, speed and repetitive interactions.</strong></p><p><strong>Humans for empathy, judgement and exceptions.</strong></p><p>And most importantly:</p><p><strong>AI should know when to hand over to a human.</strong></p><p>Knowing what AI <em>shouldn’t</em> do is just as important as knowing what it can do. I explored this idea in one of my earlier blogs, <a href="https://medium.com/@krishnakant-singh/when-should-ai-say-i-dont-know-rethinking-ai-in-customer-support-f3bfb4b0b7dc"><strong>“When Should AI Say “I Don’t Know” : Rethinking AI in Customer Support”</strong></a>, where I wrote about why saying <em>no</em> can sometimes be better CX than giving a confident but incorrect answer.</p><p>Because in customer experience, a wrong answer delivered confidently can be far worse than an honest “I can’t help with this, but I can connect you to someone who can.”</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=16a2fafbac3e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What are 4Rs of Customer Experience?]]></title>
            <link>https://medium.com/@krishnakant-singh/what-are-4rs-of-customer-experience-b2d411565052?source=rss-3054102b4f30------2</link>
            <guid isPermaLink="false">https://medium.com/p/b2d411565052</guid>
            <category><![CDATA[customer-experience]]></category>
            <category><![CDATA[cx-strategy]]></category>
            <category><![CDATA[cx-vs-business]]></category>
            <category><![CDATA[resolutions]]></category>
            <category><![CDATA[retention]]></category>
            <dc:creator><![CDATA[Krishnakant Singh]]></dc:creator>
            <pubDate>Sun, 13 Sep 2026 18:32:58 GMT</pubDate>
            <atom:updated>2026-09-18T18:59:42.006Z</atom:updated>
            <content:encoded><![CDATA[<p>I’ve always found Customer Experience interesting because it sits between the customer and the business.</p><p>The business has to think about cost, revenue, processes, risk and regulations but the customer wants their problem solved.</p><p>Over time, working in Customer Experience, I’ve started looking at CX through four different lenses:</p><p><strong>Resolution. Retention. Revenue. Regulation.</strong></p><p>These are four areas that I’ve come to understand through my experience and the kind of CX problems I’ve worked on.</p><p>Together, they offer a simple way to think about Customer Experience.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*dUzYnrH6qpbk711h5R5grQ.png" /></figure><h4>1. Resolution</h4><p>Let’s start with the obvious one.</p><p><strong>Did we solve the customer’s problem? How much time we took to solve the problem?</strong></p><p>There is a huge difference between responding to a customer and actually resolving their problem. I’ve seen this closely in my work. For one of the business units I worked with, we handled post-delivery issues such as customers receiving the wrong, damaged or expired product.</p><p>We built a customer profiling system that helped identify eligible cases and route them towards the right resolution. For these cases, resolution time went from <strong>days to near-instantaneous.</strong></p><p>The real improvement wasn’t just faster support. It was changing the process so that the customer didn’t have to wait for the problem to move through multiple steps.</p><p>A response closes a conversation. <strong>A resolution closes a problem.</strong></p><h4>2. Retention</h4><p>Solving a problem doesn’t always mean retaining the customer.</p><p>Imagine a situation where a customer whose order was delayed may eventually get a refund. Technically, the issue is resolved. But if they had to contact support, explain the problem, wait and follow up multiple times, it can still damage trust.</p><p>This is something I wrote about earlier in <a href="https://medium.com/@krishnakant.alcheringa/why-great-customer-experience-begins-before-customers-need-support-33e827efa24c"><em>Why Great Customer Experience Begins Before Customers Need Support</em></a>, where I looked at CX through <strong>Issue Prevention, Ticket Prevention and Post-Ticket Experience</strong>.</p><p>All three can influence retention.</p><p>The fewer issues means fewer tickets customers need to raise. And when they actually need help, a good post-ticket experience can determine whether they walk away frustrated or feeling that the company took care of them.</p><p><strong>Resolution fixes the problem. Retention measures the relationship.</strong></p><h4>3. Revenue</h4><p>This is where CX conversations sometimes get interesting.</p><p>Customer Experience isn’t just about making customers happy. It has a direct impact on the business.</p><p>A confusing checkout can affect conversion.<br>A failed payment can mean a lost transaction.<br>A poor onboarding experience can increase drop-offs.<br>A difficult refund journey can generate more support contacts.</p><p>If thousands of customers are facing the same problem, maybe the answer isn’t to hire more people or built more AI agents to respond on tickets. <strong>We should fix the problem itself.</strong></p><p>A recurring customer problem is often a <strong>product opportunity hiding inside an operational problem. </strong>So instead of asking only, <em>“Did this improve CSAT?”</em>, we should also ask: <strong>“What did this experience do to the business?”</strong></p><h4>4. Regulation</h4><p>This one becomes particularly important in businesses where regulations are part of everyday product decisions.</p><p>Customers want convenience. Businesses want growth. But not everything can be designed purely around what is easiest for the customer.</p><p>Companies need to follow certain rules and regulations placed by regulatory bodies across different businesses. Same goes while providing a resolution to customer.</p><p>One simple example I’ve seen across almost all the businesses I’ve worked in is the <strong>maximum Turnaround Time (TAT) prescribed by regulatory bodies for resolving customer issues</strong> in their respective lines of business.</p><p>These TATs aren’t just internal targets. Breaching them can have serious consequences, including regulatory action.</p><p>The real CX challenge is: <strong>How do we make the compliant experience as simple as possible?</strong></p><p>There is a big difference between saying, <em>“We can’t do this because of policy,”</em> and explaining what the customer <strong>can</strong> do and why a particular step is required.</p><p>The outcome may be the same. <strong>But the experience isn’t.</strong></p><h3>But Biggest Problem: 4Rs often don’t come together</h3><p>And this is probably where CX gets complicated.</p><p>The fastest resolution may not be the safest one. The cheapest solution may hurt retention. A highly compliant process may introduce friction. A great customer experience may sometimes come at a cost to the business.</p><p>So CX isn’t really about maximising one metric.</p><p>It is about making trade-offs.</p><p><strong>Resolution:</strong> Did we solve the problem?<br><strong>Retention:</strong> Did we preserve the relationship?<br><strong>Revenue:</strong> Did we create or protect business value?<br><strong>Regulation:</strong> Did we do it responsibly?</p><p>A good CX team has to think about all four.</p><h3>So, what is great CX?</h3><p>I don’t think great Customer Experience means customers never face problems.</p><p>To be honest, that’s unrealistic.</p><p>Orders will be delayed. Payments will fail. Products will have bugs. Things will go wrong.</p><p>What matters is what happens next.</p><p>For me, that is what these <strong>4Rs of Customer Experience</strong> help us think about.</p><p><strong>Resolution. Retention. Revenue. Regulation.</strong></p><p>Because ultimately, great CX isn’t just about making the customer happy.</p><p><strong>It is about making the experience work for both the customer and the business.</strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b2d411565052" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Customer Who Never Complains: The Most Dangerous Customer for a Product]]></title>
            <link>https://medium.com/@krishnakant-singh/the-customer-who-never-complains-the-most-dangerous-customer-for-a-product-a6ba42702798?source=rss-3054102b4f30------2</link>
            <guid isPermaLink="false">https://medium.com/p/a6ba42702798</guid>
            <category><![CDATA[product-strategy]]></category>
            <category><![CDATA[customer-experience]]></category>
            <category><![CDATA[customer-support]]></category>
            <category><![CDATA[user-experience]]></category>
            <category><![CDATA[product-management]]></category>
            <dc:creator><![CDATA[Krishnakant Singh]]></dc:creator>
            <pubDate>Wed, 26 Aug 2026 17:58:20 GMT</pubDate>
            <atom:updated>2026-09-18T19:00:07.590Z</atom:updated>
            <content:encoded><![CDATA[<p>There is one type of customer every customer support team knows well: The angry customers.</p><p>They raise a ticket, call support, leave a bad rating, or ask to speak to a supervisor. They may be difficult to handle, but at least they tell you there is a problem.</p><p>But what about the other kind of customers. <strong>Those who face an issue and say nothing.</strong></p><p>They don’t raise a ticket. They don’t leave a review. They simply change their behaviour. They stop using a feature. They find another way to complete the task. Or eventually, they move to another product.</p><p><strong>This customer can be much more dangerous. </strong>They are the one you might be losing silently.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GmtopzE9KduzV564d1AEkA.png" /><figcaption>The Customer You Might Be Losing Silently</figcaption></figure><h4>Not every problem becomes a ticket</h4><p>Customer support tickets are one of the easiest ways for a company to understand customer pain points. Customers tell what went wrong and we try to fix it.</p><p><strong>But here’s the important limitation, we only hear from customers who choose to complain.</strong></p><p>Imagine while making a UPI transaction, it fails once, but goes through in 2nd attempt. You will not raise a customer support ticket. But now imagine it happens a few more times. You may never contact customer support. You might simply start using another app.</p><p>From company’s perspective nothing happened, but customer’s trust in the product has slowly started to disappear.</p><h4>Customers adapt before they leave</h4><p>In my experience, customers rarely go from ‘Happy’ to ‘Gone’ overnight. Usually, there is a middle stage where they adapt. They stop using a particular feature. They purchase less frequently. Their activity goes down.</p><p>None of these actions necessarily creates a support ticket. This is why I think customer experience teams need to look beyond what customers say and start paying closer attention to what customers do.</p><h4>The danger of celebrating fewer tickets</h4><p>While ticket reduction usually a good thing, there is a trap. A company can become very good at reducing tickets without actually improving the customer experience by improving the chatbot, adding more FAQs or redirecting customers to self-service.</p><p><strong>The ticket volume goes down but did the customer’s problem go away?</strong></p><p>There is a big difference between solving a customer’s problem and reducing the number of times they tell you about it. The first improves the experience. The second only improves a metric.</p><h4>What should we look for?</h4><p>This is where product and CX teams can work together. Instead of only asking, <em>“How many customers complained?”</em>, we can also ask:</p><p><em>“What did customers do after they experienced the problem?”</em></p><p><em>“Did they retry?”</em></p><p><em>“Did they abandon the journey?”</em></p><p><em>“Did they stop using the feature?”</em></p><p><em>“Did they eventually stop using the product altogether?”</em></p><p>Good CX has always been about understanding customer behaviour, not just collecting customer feedback.</p><p>Silent doesn’t always mean dissatisfation or satisfaction. Customers change their behaviour for many reasons. But when behaviour changes unexpectedly, then that becomes a clue to deep dive, because the customer who gave you one star rating or DSAT is actually giving you a chance to fix the problem.</p><p>The customer who quietly stops coming back may not be giving you that chance.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a6ba42702798" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[When Should AI Say “I Don’t Know” : Rethinking AI in Customer Support]]></title>
            <link>https://medium.com/@krishnakant-singh/when-should-ai-say-i-dont-know-rethinking-ai-in-customer-support-f3bfb4b0b7dc?source=rss-3054102b4f30------2</link>
            <guid isPermaLink="false">https://medium.com/p/f3bfb4b0b7dc</guid>
            <category><![CDATA[ai-in-customer-support]]></category>
            <category><![CDATA[ai-in-customer-experience]]></category>
            <category><![CDATA[customer-experience]]></category>
            <category><![CDATA[ai-vs-humans]]></category>
            <category><![CDATA[ai-vs-contact-center]]></category>
            <dc:creator><![CDATA[Krishnakant Singh]]></dc:creator>
            <pubDate>Tue, 11 Aug 2026 09:45:35 GMT</pubDate>
            <atom:updated>2026-09-18T19:00:28.498Z</atom:updated>
            <content:encoded><![CDATA[<p>Over the last five years working in Customer Experience Strategy, I have been searching answer for one simple question on a daily basis: <strong>How do we make it easier for a customer to get their problem solved?</strong></p><p>The answer has changed quite a bit over the years.</p><p>Earlier, the focus was largely on building better contact centers— reducing response times, improving resolution rates and making sure customers had someone to talk to when things went wrong. Then came self-service, chatbots and automation. Today, AI is pushing that conversation much further with Conversational / Voice Bots.</p><p>We can now build systems that understand a customer’s problem, look at context and have a conversation.</p><p>That is exciting.</p><p>But it has also made me think about a slightly different question:</p><p><strong>What happens when AI reaches a problem it shouldn’t try to solve?</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pzOkC2En5gpaE71IihIJlA.png" /><figcaption>When should AI say “I Don’t Know?”</figcaption></figure><h3>The temptation to automate</h3><p>It is easy to see why companies want to automate customer support. AI can respond instantly, handle thousands of conversations at once and operate around the clock with cost much less than Human agents.</p><p>For simple and predictable customer problems, this makes perfect sense.</p><p>The problem starts when <strong>automation itself becomes the objective</strong>.</p><p>Imagine a company celebrating its AI handling 80% of customer conversations. But what if some of those customers didn’t actually get their problems resolved? They might ask the same question again, contact support through another channel, or simply give up — which is perhaps the worst outcome for any company.</p><p>From a metric point of view, the numbers may still look good, but from the customer’s perspective, they don’t.</p><p>That’s why I think we need to differentiate between <strong>handling a conversation and solving a problem</strong>.</p><h3>The cost of automation shouldn’t come at the cost of Customer Experience</h3><p>There is no denying that AI can significantly change the economics of customer support. Companies should absolutely look for cost efficiencies.</p><p>But I don’t think the cost of support should be optimized at the expense of customer experience.</p><p>One experience from my earlier work particularly shaped my thinking here.</p><p>While working in quick commerce, we often received queries about order status, especially after the promised delivery time had been breached. We could have automated these tickets by simply providing a new delivery TAT. Instead, we introduced live chat for cases where customers needed more than a standard or automated response — they needed someone to understand the problem and help resolve it.</p><p>The objective was simple: <strong>Make sure the customer actually got their problem solved, rather than just getting an answer.</strong></p><p>Our CSAT increased significantly after the change, and we also saw an improvement in customer retention during the same period.</p><p>It was a useful reminder that the cheapest support interaction isn’t necessarily the most valuable one.</p><p>A frustrated customer may not leave immediately, but repeated poor experiences can affect trust and, eventually, retention.</p><p>This is why I believe the question shouldn’t simply be:</p><p><em>“How much of customer support can we automate?”</em></p><p>It should be:</p><p><em>“Where can automation reduce cost without impacting the customer experience?”</em></p><h3>When should AI say “I don’t know”?</h3><p>I don’t think every difficult conversation needs to be transferred to a human. That would defeat most of the purpose of AI. But there should be a limit on when not to automate.</p><p>There is also a human element to support. A customer who has already tried multiple ways to resolve a problem may not benefit from being pushed through another automated flow.</p><p>This is where I think <strong>“I don’t know” can actually be a product capability.</strong></p><p>The important part, however, is what happens next.</p><h3>“I don’t know” shouldn’t mean “You’re on your own”</h3><p>A good AI handoff could look something like:</p><p><em>“I don’t have enough information to resolve this confidently. I’ll connect you with a specialist and share the details of our conversation so you don’t have to explain everything again.”</em></p><p>The AI hasn’t solved the problem, but it has still done something valuable: It has recognized its limitation and made sure that customer is not impacted further.</p><p>The customer shouldn’t have to start from scratch with a human agent and say, <em>“As I already explained to the chatbot…”</em></p><p>The context should move with them.</p><h3>The future is probably a combination</h3><p>I don’t see the future of customer support as humans versus AI.</p><p>AI will be better suited for predictable, well-understood problems. Humans will continue to add value where situations require judgment, empathy or accountability.</p><p>The interesting and important part is deciding where that boundary should sit and that boundary doesn’t have to be the same for every customer or every problem.</p><p>For example, a long-term customer facing an issue with Rs. 90,000 worth UPI transaction may warrant a different support journey from someone asking a routine question. But that also doesn’t mean loyal customers should always get a human. It means <strong>customer context and the potential impact of the interaction can be considered when designing the experience.</strong></p><p><strong>AI can handle the simple. Humans can handle the complex.</strong></p><p>Also, I think we should resist measuring progress purely by how many human interactions we eliminate.</p><p>The better question is whether we’re <strong>reducing customer effort while protecting customer trust</strong>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f3bfb4b0b7dc" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Why Great Customer Experience Begins Before Customers Need Support?]]></title>
            <link>https://medium.com/@krishnakant-singh/why-great-customer-experience-begins-before-customers-need-support-33e827efa24c?source=rss-3054102b4f30------2</link>
            <guid isPermaLink="false">https://medium.com/p/33e827efa24c</guid>
            <category><![CDATA[cx-strategy]]></category>
            <category><![CDATA[customer-support]]></category>
            <category><![CDATA[customer-experience]]></category>
            <dc:creator><![CDATA[Krishnakant Singh]]></dc:creator>
            <pubDate>Wed, 05 Aug 2026 19:04:53 GMT</pubDate>
            <atom:updated>2026-09-18T19:00:50.980Z</atom:updated>
            <content:encoded><![CDATA[<p>Ask someone to define the customer experience and I am sure they will start talking about support tickets. A very common misconception in the industry is that Customer Experience starts when customers create support tickets and common metrics which are being tracked are CSAT, Average Resolution Time or Response Time. Though they are key metrics to track a good customer support, they come very late in the customer experience journey.</p><p>Over the past five years working in Customer Experience Strategy, I’ve had the opportunity to solve customer problems across different stages of the customer journey. One observation has consistently stood out: great customer experience begins long before a customer reaches out to support.</p><p>Let’s understand Customer Experience holistically using the three fundamental pillars: <strong>Issue Prevention, Ticket Prevention, and Post-Ticket Experience.</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*62VlU2t1MNLAtOWOcODnLg.png" /><figcaption>Three Pillars Of Customer Experience</figcaption></figure><h4>1. Issue Prevention: The best support ticket is the one that never needs to exist</h4><p>Issue Prevention is the first category where a best customer experience can be delivered. The best customer experience is where customers do not have to think about creating tickets.</p><p>Imagine a quick-commerce app that promises grocery delivery in 10 minutes. On a rainy day, the platform already knows deliveries will take longer. Instead of accepting the order with a 10-minute promise and updating the ETA later, a better customer experience is to set the right expectation upfront. Informing customers of the delay before checkout enables an informed decision and prevents unnecessary anxiety after the order is placed.</p><p>This can be achieved using better product designs and clear communications.</p><h4>2. Ticket Prevention: Enable customers to resolve issues without contacting support</h4><p>Not every issue can be prevented, but, not every issue needs a support ticket interaction either.</p><p>Ticket Prevention is about empowering customers with right information, tools and automation to resolve their issues quickly and independently. This reduces customer anxiety and effort without needing customer support.</p><p>Imagine you’re transferring money to a friend via UPI, and the transaction gets stuck. Your immediate concern is, “What happened to my money?” If the app doesn’t provide clarity, uncertainty quickly turns into anxiety, often leading to a support ticket. A simple reassurance on the transaction page — “Your money is safe. We’re processing your transaction and will update you shortly.” — can address the customer’s concern instantly and eliminate the need to contact support.</p><p>Unlike Issue Prevention, where the issue never happens, here the issue exists — but the customer doesn’t need to raise a ticket.</p><h4>3. Post Ticket Experience: Deliver an empathetic, efficient, and trustworthy support experience</h4><p>Okay, now we are at the last leg.</p><p>This is the stage where customers feel the need to reach out to support. By this point, they’ve often experienced inconvenience, uncertainty, or frustration. The goal of this pillar is to restore trust through timely, accurate, and empathetic support interactions.</p><p>Core philosophy for Post Ticket Experience is, when prevention isn’t possible, recovery becomes the experience customers remember.</p><h4><strong>Summary</strong></h4><p>A great customer experience s not built on resolving support ticket faster but it is built on reducing the need of support. Organisations which excel at customer experience, invests across the pillars.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=33e827efa24c" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>