Deepak Gupta

The secret behind the rising profitability of Indian #startups can often be traced back to a single line item: the convenience fee. In FY 25: - Zomato : Profit -> Rs 527 Cr, Platform fee collected -> Rs 650 Cr - MakeMyTrip: Profit -> $95.3 M, Platform fee collected -> $110.5 M - IRCTC's convenience fee contributes to only 21% of revenue but 74% of profits For years, the belief was that “Indians don’t pay for services.” But that narrative is changing. Consumers today are willing to pay for the ease, speed, and added comfort they receive; so much so that even Amazon has started charging an extra convenience fee on #ecommerce orders. While these are growth engines but deciding on the convenience fee is extremely tricky. Take Zepto or Swiggy, where convenience fees are often the subject of heated rants. Yet, very few complain about IRCTC’s fee, which is significantly higher and has been around since its inception. So, what makes users tolerate a convenience fee in one case and resent it in another? An interesting lens through which to view this is the Delta 4 theory, popularised by Kunal Shah from CRED. It’s a framework for evaluating whether an industry or startup is likely to thrive, and it applies remarkably well to the psychology behind convenience fees. Here’s how it works: imagine you’re rating your experience of availing a service on a scale of 1 to 10. Take railway ticket booking, for instance. The traditional way, standing in long queues at a physical counter, would hardly get a 2 or 3. But booking the same ticket online via IRCTC? Most of us would rate that an 8 or 9. That gap, 9 minus 2 , is the delta IRCTC adds. And because it’s greater than 4, the #business sustains, and users rarely complain about paying a convenience fee. Now, apply the same lens to other industries: - Flight bookings: going all the way to the airport to book at the counter is a hassle, so paying ₹150–200 extra online doesn’t hurt. - Food delivery or groceries: Blinkit, or food delivery platforms, often face more resistance because the alternative (walking a few meters to a nearby restaurant or store) isn’t all that painful. So, the higher the Delta, the less likely people are to complain about the convenience fee. Convenience fees don’t operate in isolation. Product managers at Quick Commerce and food delivery #apps are constantly A/B testing factors like: - Average order value (AOV): Higher-value orders can absorb a small fee more easily. - Purchase frequency: High-frequency purchases tend to make the fee less noticeable over time. As these #tech startups mature, convenience fees are becoming the difference between #growth and profitability. They represent a new social contract: customers willing to pay for efficiency, convenience, and saved time. #kneeledge

810

59 Comments

Left Microsoft, built a 20,000 CR company, and donates 40 lakh every year. 1. During his 8-year stint at Microsoft, Samir Bodas created history by becoming the director for SME – Sales and Marketing, Microsoft's first foray into small businesses. He was responsible for anti-piracy, licensing, and demand generation. However, the dot-com boom burst in 2000. 👇🏻 2 He left Microsoft, wanting to start his life from scratch. In 2002, he joined Disha Technologies, an early-stage IT Services firm, as its CEO, focusing on software testing and quality. He led the brand from 1 CR in sales to an acquisition by the Indian IT company Aztecsoft for 54 CR in just two years. 🙌 3. Samir returned to Pune, India, and was seeking new ideas. He met Monish Darda and got inspired by him to start a startup together. They began searching for ideas and discovered that business contracts were ripe for disruption through the use of cloud and AI. Samir wanted to take action. 🤔 4. In 2009, he decided to start a company with dual headquarters in Bellevue (USA) and Pune to attract customers globally. He named it after a suggestion by his wife on the Latin word "certa," meaning "trust." Icertis was born. 🚀 5. The idea was simple ⏩build a "Salesforce of contract management". Companies can digitise all contracts into a single repository and extract critical insights for risk and compliance. But companies were still not convinced. 🤔 6. Companies found no strategic value, and he had to compete for a limited share of enterprise IT budgets. He then pivoted to software-based contract lifecycle management solutions, providing automated clause analysis, obligation tracking, and workflow enforcement. And it worked. ✅ 7. As it helped Mercedes-Benz achieve a cycle-time reduction of 80%, Microsoft, Airtel, Hyundai, and Johnson & Johnson became its customers. Samir was now scaling on a 100% cloud infrastructure. By 2015, he raised 37 CR led by Greycroft. 💰 8. Within a year, Icertis had scaled to 500,000 users and 2 million contracts. Samir was building "the foundation of commerce" through contract intelligence. As it scaled to 750,000 users and 2.5 million contracts by 2017, it raised 162 CR led by B Capital Group. 💸 9. Samir grew with his famous FORTE (Fairness, Openness, Respect, Teamwork, and Execution) principle to over 3.5 million users and 4 million contracts processed by 2018. Five of the world's ten most valuable companies were its customers. And finally, the big news came on July 17, 2019. 👇🏻 10. Icertis raised 792 CR led by Premji Invest at a valuation of 6890 CR. It became the first contract lifecycle management unicorn globally. 🦄 11. Today, Icertis manages over 10 million contracts in more than 40 languages and across 90 countries. It is valued at 20,300 CR. 📈 However, Samir Bodas still donates Rs 40 lakh to over 170 non profit organisations every year. 🙏 #startups #india #casestudy For more such stories, read my book - Startups of Bharat, on Amazon!

1,345

60 Comments

i’ve built 4 companies in 7 years. all profit making. yet i’ve spent most of that time quietly falling apart. 1. inVOID? we raised from YC. reached $1.5m ARR. felt like we made it. but scaling outside India? brutal. different compliance, slower sales cycles. we thought ambition could outrun geography. it didn’t. 2. Unfinance? 100 million views. 1.5 million followers but most nights i was refreshing analytics at 3am, wondering why virality still didn’t feel like validation. 3. NoApp? closed 100+ clients. but we built on Meta’s API - one policy change away from extinction. 4. SuperPen? our first paying user came from a meme i almost deleted because it felt too desperate. i wish i could say it gets better. what actually happens is: you get better at functioning inside the chaos. i’ve seen founders with perfect roadmaps and no guts. i’ve seen creators who look successful but live in fight-or-flight. and i’ve learned this: - startups don’t test your IQ. - they test your capacity to keep moving while everything around you is collapsing - quietly, politely, professionally. the truth is, there’s no playbook. so if you’re still building even after the investor silence, the product flops, the nights where you question your whole identity - congrats. you’re already doing the hardest part. you’re still here.

442

53 Comments

Search engines, like Google, rely heavily on search logs to understand search behaviors - but they are not enough. This is also true for vertical search engines like amazon, swiggy, spotify, etc. When I was building search at Unacademy, we started by relying heavily on the search logs to improve relevance, but our relevance plateaued. No matter what we tried, the needle was not moving by much. Then I read a paper and following key insights from it, started talking to our actual users, observed their behavior, and used tools that gave us insights on what they were 'actually' doing and struggling. this was completely non-scalable, but so worth it. The insights we could gather (talking to them, capturing detailed metrics, user interactions, etc) shaped the next 3 key features we launched, which shot up the relevance, CTR, and watch time. This weekend, I am re-reading the paper that inspired our actions. The paper is from Google (written in 2007) which talks about things we should augment the search logs that improve relevance. Spoiler alert - a bunch of them are non-scalable. This paper is a classic example of how human feedback is the most critical thing to build a superior product. There cannot be one generic solution that fits all when it comes to search, given people have very peculiar ways of looking for certain types of things - food, restaurant, course, topics, questions, etc. The paper is a breeze to read and would hardly take you a day to wrap. Give it a shot, if search engines or relevance systems excite you. You can always find the papers I read and the notes (of some) on my paper shelf and other write-ups in comment arpitbhayani.me/papershelf arpitbhayani.me/blogs By the way, I have 5 videos on my channel talking about search and how different products have been implemented and improved. If you find the search interesting, you will love the videos. Here they are - search at zomato - youtu.be/JL9x9N6YSUc - search at yelp - youtu.be/i8MweKYoG1U - search at twitter - youtu.be/dOyCq_mMtdI - search at flipkart - youtu.be/NcNCty7_3kc #AsliEngineering #SystemDesign

520

6 Comments

Zomato's monthly userbase - ~20 million Swiggy's monthly userbase - ~17 million Uber India's monthly userbase - ~17 million Cred's monthly userbase - ~13 million Declared income tax payers over 50 lakhs ($58k USD) are around 9 lakh (900,000). Another 24 lakh folks (2,400,000) earn over 20 lakh ($23k USD) . A further 69 lakh (6,900,000) people earn over 10 lakh ($12k USD). Now this is based on Income Tax Return data so evaluate these numbers as you will. So when you're building a startup, and I see TAM SAM SOM (god this sounds like a Thai Salad) showcasing over 3 crore paying users, my question is "Sure, where are you producing these folks from?" We've seen a ~500% growth in tax payers earning above 50 lakh from 2013 till 2024. Assuming even a similar trajectory, we've got 45 lakh folks by 2035. Still a bit short of 3 crore. Meanwhile, AI is most DEFINITELY impacting jobs, household debt sits at ~42% of GDP and inflation is high. Now none of this is meant to be discouraging. We've had a surge of brilliant brands rise in India, several which are making their way across the world today. Lenskart, Innovist, Mokobara, The Whole Truth, Blue Tokai, these are all part of my everyday usage, and I'm seeing them sold across the world. If you had told me this a decade ago, even I wouldn't have fully believed it was possible! The reason I'm writing this is to put out direct numbers. There's a finite audience you're reaching, so understand the cost of acquiring them, retaining them and creating a differentiated voice and product. #venturecapital #startups #d2c #consumer

465

88 Comments

Ankit Shukla and Aakash Gupta did an awesome study on PM salaries in India They found that average PM CTC grows from 17 Lakhs per annum as an APM to 130 Lakhs as a CPO. Of course, there's some bands around all of this. Here's the total comp 25th to 75th percentiles: Associate PM: 14 to 20 lakhs PM: 18 to 38 lakhs SPM: 37 to 62 lakhs PPM/GPM: 62 to 97 lakhs Director: 61 to 133 lakhs VP: 77 to 135 lakhs CPO: 52 to 194 lakhs But it's interesting to look at the average. For instance, here's the average split between Fixed/Variable/ESOPs Associate PM: 82%/ 11%/ 6% PM: 81%/ 6%/ 13% SPM: 76%/ 9%/ 14% PPM/GPM: 74%/ 13%/ 14% Director: 65%/ 7%/ 28% VP: 65%/ 7%/ 28% CPO: 55%/ 7%/ 38% The ramp up in variable and ESOPs is very dramatic at senior levels, from 17% to 45% of total compensation. And that leads to huge increases in Total CTC. For more details - mediums, locations, company, and other cuts - check out the full report: https://lnkd.in/djbctbAV

292

18 Comments

Groww’s 1,400+ employees have created ₹44 crore of value each at IPO market cap. 🔥 2× more leverage per employee than any other listed Indian internet company. Most people talk about network effects, margins, growth, CAC, operating leverage… But one under-discussed indicator of platform strength is value created per employee. 🚀 Groww is running one of the leanest teams among large venture-backed IPO scaleups — and that’s the part that should interest founders, operators, and investors the most. Fewer employees. Massive output. Maximum leverage. This is the real internet efficiency story — not user counts or vanity GMV. https://lnkd.in/gXp6GJ5F

110

2 Comments

10 indian companies that you might not know were yc companies: 1. groww (W18) 2. meesho (S16) 3. razorpay (W15) 4. zepto (W21) 5. orange health labs (S20) 6. khatabook (S18) 7. cashfree payments (S17) 8. lokal (S19) 9. okcredit (S18) 10. innov8 (S16)

126

1 Comment

Zomato / Swiggy are building what Paytm tried to build 5 yrs ago - but better. For anyone who thinks Zomato and Swiggy are still food delivery apps has been living under a rock. Food delivery (yes at high burn rates) was simply CAC. Now comes the monetization. Like everyone had / used paytm almost everyday, multiple times a day a few yrs ago and “Paytm karo” became synonymous with UPI payments. But with payment monetization getting regulated and Paytm unable to effectively become a “superapp” when venturing into ticketing, entertainment, mutual fund investments etc., the CAC never translated into monetization. Hard luck + not as sharp execution. Then came other innovations like the sound box, BFSI like longer-time taking monetization strategies. Might work but take time. Zomato / Swiggy on the other hand - move fast. Zomato bought Paytm’s ticketing business and “District” was born. Blinkit. Instamart. Now Scenes by Swiggy. In the future, flying buses by Zomato (if one can dream). Private labeled groceries (Noice by Swiggy) - the customer is here, now sell the high margin products. The monetization is kicking in thick and fast. You’re a captive customer and will remain so with the food delivery, and everything else will bring in the margins and the monetization. A Superapp without being labeled a superapp. That was always the plan. It is just playing out now, and how!

155

8 Comments

Anthropic just raised $13B at a $183B valuation. This is wild, that single round is more than all VC funding raised in India in 2024, and even beats the $11.6B Indian startups have raised in 2025 so far. Anthropic valuation tripled in 6 months (from $61.5B to $183B). But the real story isn’t the money, it’s the strategy shift happening behind the scenes. Key highlights: -Revenue went up 5x to $5B run-rate in just 8 months -They've got 300,000+ business customers now -Claude Code is already making $500M+ They're beating OpenAI in enterprise market share (32% vs 25%) Everyone's obsessed with ChatGPT for consumer stuff, but Anthropic has quietly been building this enterprise empire. Their "Constitutional AI" approach focuses on safety and reliability, which is exactly what big companies and their security teams want. Some interesting things to look at... 1. Enterprise vs Consumer Strategy Is Anthropic's approach (85% API/15% consumer) actually smarter than OpenAI doing the opposite? Which one wins in the long run? 2. The Valuation Reality Check A $183B valuation on $5B revenue equals a 36x multiple. Are we in another bubble, or does AI actually justify these crazy numbers? 3. Ethics vs Growth Dilemma Anthropic's CEO said they had concerns about Gulf state funding but took it anyway because they needed to compete. How do you balance ethics with the massive $$ needed to stay in the game? The AI landscape is changing so fast. While we're all debating which AI to use for personal stuff, big companies are making billion-dollar decisions on which platform to bet on. What do you think about Anthropic's enterprise-first strategy?

58

8 Comments