Learning From Mistakes

In the early years of Marico, I made a decision that shaped everything that followed. I gave young people serious responsibility. Not after years of proving themselves. Right at the start. Real roles. Real ownership. Real consequences. Some of them made mistakes. That was expected. But they learned fast. They asked for feedback. They iterated. They took initiative. Many of them went on to become some of the best leaders in the organisation. My view is simple. If someone makes an error, but they take ownership, improve, and do not repeat it, they have earned the right to grow. If someone avoids responsibility or hides from mistakes, they are not ready. Empowerment is not about handing out titles. It is about giving people the room to fail safely and the support to recover intelligently. You cannot build a high-performance organisation if every decision needs your approval. The real test of leadership is not in how well you control outcomes. It is in how confidently you let others lead. #leadership #hiring #talent #growth #future #team #success

I don’t understand the shame associated with making mistakes. No one is perfect. No one delivered high-quality work on their first attempt. High-quality work IS the outcome of mistakes reflected upon. Use the RISE framework to turn any mistake into progress: R – Recognize Identify the mistake clearly without defensiveness. I – Investigate Ask why it happened. Was it a skill gap, a misunderstanding, or a process flaw? S – Synthesize Turn the insight into a principle or takeaway. E – Execute Differently Apply the learning immediately in your next project or task. Making mistakes is not the mistake. Not learning from them is. Quote from the book: It Always Seems Impossible Until It's Done. PS: If you found this helpful, I share more insights to help you build a growth mindset. Follow along if you’re on the same journey.

When Style Disrupts Safety: A Lesson in Product Design Today, while driving behind the sleek Mahindra BE.6E, a futuristic and stylish EV, I experienced something unexpected. The car in front of me braked suddenly. I had maintained a safe distance, so I stopped comfortably. But something felt off. Why did the braking catch me off guard? Then I realized: The brake lights were too subtle. The taillights are designed as a thin rectangular LED strip, stunning to look at, no doubt. But the brake lights occupy only a tiny section on the top edge of that strip. Visually stylish, but functionally weak. In real traffic conditions, where immediacy and clarity are critical, this design doesn’t help other drivers react intuitively. This reminded me of a fundamental product design principle: Aesthetics must never come at the cost of usability. A good product delights not just by how it looks but by how well it works. Whether we’re designing: • A mobile app • A banking interface • Or a car’s tail lights …it’s our job as product managers and designers to make sure the experience is not just elegant, but intuitive, accessible, and safe. Lesson for us in Product Management: Design for the user’s reality, not just the brand’s imagination. Functionality and clarity should never be hidden behind a glossy UI, whether it’s a screen… or an LED strip on a car. #ProductManagement #UXDesign #Usability #AutomotiveDesign #DesignThinking #BuildWithEmpathy

Most of us get at least 1 opportunity that changes our career trajectory; I am not referring to the biggest/ largest role we do but the role that prepares us for these bigger, larger roles. It is the role that sets us up for the future.   For me, it was moving from an HR Head role for a mid-size company in India where I managed a large team to an individual contributor role where I would work in a specialized area but at a global scale within the same company. I was advised by many not to pursue it but I was open. What was the attraction – it would allow me to work with HR & Business Leaders of over 50 countries and it was a new role; so I could shape it.   I got experiences I could have never imagined for myself. Conducting a goal setting workshop for the leadership team in Japan; piloting a leadership program for Western Europe and conducting a performance management training for the Bangladesh team. I worked with colleagues from so many different cultures and backgrounds. I changed – both personally and professionally. I was humbled with everything I didn't know. I learned how to adapt, I became less judgmental and a lot more open minded.   The role also gave me an opportunity to work with very senior leaders and it was a booster immersion in how they think; the questions they ask and how they make decisions. Most importantly, I learned to operate without authority. And that helped me do my next role better.   Here are 4 of my biggest learnings:   1. Don't judge roles by size/ scale; think about the potential to impact. A larger role (textbook definition) doesn’t always result in higher impact - the context is very important. 2. Careers are about skill stacking. And instead of mastering one skill, we need to take a step back from our core strength and build a winning combination of skill sets that are unique to us and make us more effective. 3. Experience is different from experiences. Too often we define learning very narrowly in the professional context. Joining a well settled team and making your place in it; managing demanding peers; building a team or keeping it intact in challenging times; navigating a large & complex organization; working for an inspiring leader; managing conflicting priorities of different stakeholders. These are all experiences. 4. Be open, take risks - it is a key quality in managing our careers successfully. And we all need some discomfort to help us reach our full potential. #experiences #careers

I open sourced Sniffly (https://lnkd.in/geEk3HgN), a tool that analyzes Claude Code logs to help me understand my usage patterns and errors. Key learnings from spending so much time looking at the logs. 1. The biggest type of errors Claude Code made is Content Not Found (20 - 30%). It tries to find files or functions that don't exist. So I restructured my code base for discoverability, and the average number of steps Claude Code needs for each instruction went from 8 to 7 steps. 2. Traditional metrics of engineering hours/days don’t work for AI. Two metrics I use to evaluate the complexity of a project: - how many instructions I need to give AI - how often I have to interrupt it because it goes into the wrong direction Across my projects, the interruption rate is about 1 in 4 instructions. This means I still need to actively monitor the agent. 3. While most of the time, Claude Code can only go up to 10 steps before I need to interrupt it, it can occasionally go close to 100 steps. Just a year ago, people told me it was hard to get an agent to go above 5 steps! Claude Code’s favorite tools are, unsurprisingly, search tools (grep, ls, glob), which make up ⅓ of tool calls.

Some of my “worst” investments aren’t the ones that went to zero––they’re the investments where the real cost was saying no. Airbnb, Stripe, and Square: examples that were each a lesson in overestimating risks, underestimating potential, or not doubling down on conviction. These weren’t easy “no” decisions; they were shaped by what I knew at the time, but they revealed something critical about how we weigh opportunity and risk. For Airbnb, I passed on doubling down because I had already led an earlier round of investment and convinced myself I didn’t need to lead the next. Although I had more conviction as time went on, I didn’t press the advantage. For Stripe and Square, my deep knowledge of payments—gained from my time at PayPal—ironically became a blind spot. I knew how challenging the payments space was and let that knowledge overshadow the potential these companies offered. Reflecting on these––and many other misses––I find it’s important to continually challenge the narratives we build for ourselves as founders and investors. Sometimes, those narratives are grounded in logic, and other times they’re too heavily weighing past experiences that don’t necessarily predict future outcomes. No investor can win every bet; But the best outcomes come from a willingness to evolve, to revisit our decision-making frameworks, and to apply what we learn to the next opportunity.

You have done this a hundred times. You try to plug in a USB. It doesn't fit. You flip it. Still wrong. You flip it back. Finally it goes in. Annoying. But notice something. You can only plug it in the right way. The wrong way is blocked. That is not an accident. It is a 60-year-old idea from a Toyota factory, and it has a name: poka-yoke. In plain English, mistake-proofing. It came from a Japanese engineer named Shigeo Shingo in the 1960s. He noticed workers on one line kept forgetting to put a tiny spring under a switch button. Every manager's first instinct is the same: tell people to be more careful. Retrain them. Maybe write them up. Shingo did the opposite. He changed the process. Instead of grabbing each spring directly, workers first placed the springs in a small dish. Then they took each one from the dish into the product. Now if a spring was left in the dish, anyone could see it instantly. Defects dropped to zero. Overnight. No extra training. No blame. He first called the idea "baka-yoke," which means fool-proofing. Then he dropped that name. Calling workers fools, he said, was the wrong way to think. Everyone makes mistakes. A good system catches them. Here is the useful part. Shingo said you can mistake-proof almost anything in one of three ways. Worth keeping somewhere: 1️⃣ Make it physically impossible to do wrong. Shape it so only the right way fits. (The SIM card. The USB. The diesel nozzle too wide to enter a petrol car.) 2️⃣ Make the right count obvious. Set it up so a missing or extra piece shows at a glance. (Shingo's dish of springs.) 3️⃣ Lock the steps in order. Build it so a step cannot be skipped or reversed. (The ATM that returns your card before it gives you the cash.) Notice what all three have in common. Not one of them depends on people trying harder. They depend on better design. So before you retrain someone for an honest mistake, ask one question: ⁉️ Could I redesign this so the mistake simply cannot happen? That single question is one of the quietest superpowers in Lean and Six Sigma. Now I would love your help with something. I have only ever seen a small slice of this. What is the smartest piece of mistake-proofing you have come across, at work or in everyday life? Add it in the comments. I want to learn the ones I have missed, and I have a feeling this thread could turn into a brilliant list for all of us. Follow Rahul Iyer for Lean, Six Sigma, Project Management & AI Insights.

🧵 Ever heard of a “Failure Résumé”? It might be the smartest career exercise you’re not doing. Here’s what it is—and why it can change the way you grow 👇 A failure résumé is exactly what it sounds like: Not a list of wins. Not your greatest hits. But your flops, screw-ups, and bad decisions. It’s uncomfortable—and incredibly useful. The idea comes from Tina Seelig at Stanford. She challenges her students to build a résumé of their failures. Then asks: “What can you learn from each one?” I made my own It wasn’t for the public. Just a long list of personal and professional misfires. Then I reviewed each one and asked: Was there a pattern? Was there a lesson? Turns out—yes. My biggest insights? Mistake #1: Starting projects based on untested assumptions. Assuming I “knew enough” instead of doing the homework. Mistake #2: Saying yes to things I wasn’t fully committed to. Half-hearted effort = half-baked results. Those 2 patterns showed up again and again. But here’s the upside: Once I spotted them, I could fix them. That’s the power of a failure résumé. It turns regret into direction. So try this: List your failures. Big, small, awkward, and ugly. Then ask: Where did I go wrong? What keeps showing up? There’s gold buried under the cringe. You don’t need to share it with anyone. Just be honest. Be curious. And if you don’t do it? Well… you might have to add that to your failure résumé too 😅

Long ago in Delhi, a British officer put up a sign: “WE PAY FOR EVERY DEAD COBRA.” Soon, people lined up with snakes. Then something unexpected happened, they started breeding cobras to earn more money. When the plan was stopped, the breeders released their snakes. The result? Even more cobras than before. In hospitality, the same principle applies. Imagine a hotel rewarding staff for every positive guest review. At first, it seems smart but soon, some team members focus only on getting reviews, encouraging guests to leave quick feedback rather than delivering genuinely thoughtful service. When the incentive is removed, the focus on authentic service has diminished, and overall guest satisfaction may drop. The lesson: short-term incentives can backfire. Building a culture of genuine service, long-term habits, and intrinsic motivation creates results that last without “breeding more cobras.”

A junior pinged me late night … “Hey, sorry for disturbing so late, but the hub-allocation service just stopped writing events. I’ve been staring at the logs for an hour and can’t see it.” I was two chapters deep in a book, but a production freeze at Flipkart waits for no one. Ten minutes later we were on a call, screens shared, coffee in hand. What we saw: 1. CPU was fine, DB healthy. 2. Message Queue consumer lagging — but only for one partition. The suspect commit: a “tiny” config change that slipped past review because “it’s just YAML.” What we did: 1. Replayed the partition in staging → reproduced the freeze in 30 seconds. 2. Flipped the feature flag off, deployed a hotfix. 3. Wrote a one-liner unit test that fails if the critical topic/partition mapping ever changes without a version bump. Total downtime: 23 minutes. Total learning: off the charts. Three takeaways I shared with the team the next morning: 1. Small changes aren’t small in distributed systems. A single-line config tweak can strand an entire message bus. 2. Cultivate “safe-to-ping” culture. The bravest thing that junior engineer did wasn’t debugging at 1 a.m.; it was sending that message before things spiraled. 3. Automate the guardrails. Post-mortems are great, but a failing test is louder than any Confluence page. AfterMath: That junior pushed the unit test themselves, opened the merge request, and led the retro. Next sprint, they volunteered to refactor our event-routing configs; because now they own the problem. These are the moments that turn capable engineers into future tech leads.