@Sujayath @sujayath-blog - Tumblr Blog

AI concepts that every founder / PM should know. The no-jargon version.

AI: AI is software that can do things we used to think only humans could do. It is a software that can learn by itself from data and uses it to predict or generate something.

Predictive AI vs Generative AI: Predictive AI looks at data and says "what's likely to happen next?" - fraud detection, demand forecasting, lead scoring. Generative AI looks at data and says "let me create something new" - text, images, code, music.

Machine Learning & Deep Learning: Machine Learning is AI that learns patterns from data instead of being manually programmed. Deep Learning is a subset of ML that uses neural networks (layers of math inspired by the human brain). Deep Learning works well on messy data like text, images, and audio.

Large Language Model (LLM): Think of it as an incredibly well-read intern who has read most of the internet. Under the hood, it just predicts the next word extremely well that it can write essays, code, and reason through problems. GPT, Claude, Gemini - all LLMs.

Foundation Model: Instead of building a separate AI for each job, you train one giant model and fine-tune it. Like building one strong engine and putting it in cars, trucks, and boats.

Training vs Inference: Training = teaching the AI. Expensive. Takes weeks. Costs a lot. Indirect / R&D Cost. Inference = using the trained AI. Cheap per query. Milliseconds. Direct Cost.

Prompting / Prompt Engineering: The art of asking AI the right question to get the right answer. A vague prompt gives a vague answer. A specific prompt with context, examples, and constraints gives a brilliant answer. Levels of Prompting: Zero-shot: Just ask. No examples. Few-shot: Give 2-3 examples in your prompt. Chain-of-thought: Ask the AI to think step by step.

Tokens: LLMs don't read words - they read tokens. A token ≈ ¾ of a word. You pay per token. Every API call costs money based on tokens in (prompt) and tokens out (response).

Context Window: The amount of text an AI can "see" at once. Its working memory. Small context window = AI forgets earlier parts of your conversation.

Hallucination: When AI confidently makes up facts that sound completely real. It doesn't "know" it's lying. It's pattern-matching and sometimes the pattern leads somewhere wrong.

RAG (Retrieval-Augmented Generation): Instead of relying on what AI memorized during training, you feed it your own documents at query time. Reduces hallucination.

Fine-Tuning: Taking a pre-trained model and training it further on your specific data. Like hiring a generalist doctor and sending her to specialize in cardiology. More expensive than prompting. Cheaper than training from scratch.

Embedding: Converting text (or images) into numbers that capture meaning for AI. "Happy" and "Joyful" → similar numbers. "Happy" and "Refrigerator" → very different. This is how AI "understands" similarity.

Vector Database: Traditional databases find exact "word" matches. Vector databases find "similar" things. User searches "shoes for flat feet" → vector DB finds relevant products even if no description uses those exact words.

Agent: An AI that doesn't just answer questions - it takes actions. Browse the web, call APIs, write code, book meetings.

Agentic Workflow: Instead of one big prompt → one big answer, you break the task into steps. Agent 1 researches. Agent 2 drafts. Agent 3 reviews.

Tool Use / Function Calling: Giving AI the ability to use external tools - search engines, databases, APIs. The AI decides when to use which tool.

Coding Agents: AI that writes, tests, debugs, and ships code autonomously. You describe what you want in plain English, the agent writes the code. Ex: Claude Code, Codex, Cursor.

Computer-Using Agents (CUA): AI that can see and use your screen - click buttons, fill forms, navigate apps - just like a human would. Ex: Claude Cowork, OpenAI's Operator.

Multimodal AI: AI that understands and generates all modes - text, images, audio, video, and code. Your product can accept a photo and return text, or accept text and return an image.

Diffusion Model: The tech behind AI image generation (Midjourney, DALL-E, Stable Diffusion). Starts with noise, gradually removes it to create an image.

Synthetic Data: Training data generated by AI itself, not collected from the real world. When you don't have enough real data, or real data has privacy issues, you generate synthetic data.

Guardrails: Rules and filters around AI to prevent it from going off-script. Prevent harmful content, off-topic answers, leaking confidential data.

Human-in-the-Loop (HITL): AI does 90% of the work. A human reviews the last 10%. Needed when mistakes are costly - medical, legal, financial.

AI-Native vs AI-Enabled: AI-enabled = existing product with AI bolted on (chatbot on your PowerPoint). AI-native = built from the ground up around AI (Gamma).

Temperature: Controls how "creative" vs "predictable" the AI is. Low temperature = safe, consistent, factual. Ex: Customer Support AI. High temperature = creative, varied, crank it up. Ex: Brainstorming.

Overfitting: When AI memorizes training data instead of learning patterns. Like a student who memorized past exam answers but can't solve a new problem.

Eval (Evaluation): QA for the AI world. How you check if your AI is actually doing what it was supposed to do in all cases. Checks for Accuracy, Relevance, Helpfulness, Safety. Eval is pre-ship. Post-ship, you need what's called monitoring to catch drift, cost spikes, and latency.

MCP (Model Context Protocol): A standard that lets AI connect to external tools and data in a plug-and-play way. Think USB-C for AI - one standard connector instead of custom integrations for every tool.

Explainability (XAI): Can you explain why AI made a decision? In finance, healthcare, insurance - "the AI said so" is not acceptable.

The "Last Mile" Problem: AI can get you 80% of the way shockingly fast. That last 20% - handling edge cases, weird inputs/outputs, industry-specific nuance, getting output just right - is where all the time and money goes. Most failed AI products didn't fail at the AI. They failed at the last mile.

#startup

The India I Knew

Yes, our streets were dirty,

Maybe our homes were too!

But I can assure you,

Our minds were still clean!!

Yes, money was tight,

Yet life was so full!

There was no need for flight,

Train journeys were cool!!

Yes, the system was corrupt,

Yet people were good!

GDP was small,

But dreams were big!!

Yes, parents were strict,

Yet family was fun!

Even distant relatives

Still felt so close!!

Yes, the country was crowded,

Yet there was space for all!

Even strangers were kind,

With no hate to blind!!

Who cast the Evil Eye?

Who rolled the wrong die?

Why did we all leave

Our motherland so dry?

My father was born here,

My grandfather roamed here!

Even before gods arrived,

My ancestors prayed here!

And just like that, one fine day—

I no longer belong here …

Yes, the temple was not there,

But God was still there!

Love was in the streets,

Religion stayed at home!!

Yes, my country was not perfect,

But it was still mine!

Until you came to save us,

We were just fine!!

Grit

There will come a time,

When all seems to be lost!

There will be no path to go!

There will be no energy left!!

Nothing can be done!

You are all alone!

All there is silence,

And a still blackness!!

You can't see anything!

It's all Dark!

The only purpose of your eyes,

Is not to see, but cry!!

Closed ones would have left you!

Trusted would have stabbed you!

When you see your loved ones,

Their sadness will sink you!!

Every door you knock,

You will find it closed!

After hundreds of knocks,

Still hundreds to go!!

You will now find out,

You were never a Hero!

All you thought you did,

Was merely a luck!!

You will be shivering!

Stomach will be churning!

You might even think,

Death might be relieving!!

Loss after loss,

Will make you numb!!

There will seem no purpose,

Why even try?

Your brain says it's over!

The world says it's over!

But there is that faint voice -

'Maybe It's not over,

Until YOU say it's over!!'

You will get up!

With great effort,

You will take that next Step!

There is no oxygen!

You will still be breathing!!

There is no floor below!

You are still not falling!!

When you can't be running,

You will still be walking!

When your legs give up,

You will start crawling!

When you can't crawl,

How are you still Moving?

It will look like God's act!

It might look like magic!

Things that are impossible,

Will now look like sure shot!!

You are breaking nature's laws!

But universe will bend for you!!

Where one task was uphill,

Thousand tasks are cakewalk!!

Hope might only be slim!

But what can a mere God do?

When a mighty mortal

Stands against Him!!

Purpose

One day, You will be very good at something! Very good Indeed!

All the hard work your opponent does, Whatever courage he could muster, However prepared he could be, Will not seem to matter against What comes natural to you!!

You will not sweat! You will not have stress! It will feel neither like work, Nor like play! But like breathe You will do it all day!!

You will follow no rules! You will follow no rituals! 'This is how things are done' People won't come near you!

No one can understand how you do, Or for that matter, what you do! Grammar will not apply to you, As it will be rewritten by you!!

No opponent will envy you! No competitor will hate you! They will have no ego! There will be no anger! You will just see their awe, Even when they are demolished by you!!

Some will think you are an alien, As no one on earth can do what you do! Some will think you are an animal, As your sheer strength is unexplainable! Some will think you are a God, And just choose to kneel before you!!

That good you will be at something, That I am very sure of! Now you have just one job to do, Find what is that something that you will do!!

#startup

Strategic Snippets for founders / PMs

A mega-thread on strategic concepts that every founder / PM should know.

1. Kernel of a Strategy: A process to create a strategy. It contains three elements: diagnosis, tenets (guiding principles), and action items. You spend half of the time on diagnosis, another 40 percent on the tenets, and 10 percent on coherent actions.

2. BHAG: It is a big, daring, ambitious goal that pushes the company beyond its boundaries defined clearly with no ambiguity. People get it right away. It has a sense of urgency. It has a purpose. When you first hear it, you will feel it's a joke and that you’d never achieve it.

3. Objectives and key results (OKR): A goal-setting framework that helps companies set an objective, which is “what I want to have accomplished,” and the key results, which are “how I’m going to get it done.” OKRs must be specific and include a way to measure achievement.

4. Operating Plan (OP1) is an annual planning document that covers the strategy for the next 12 months, ways of executing the strategy, and the budget required. BHAG answers WHAT, OKR answers HOW, OP1 answers WHY.5. Flywheel: There is no single action, no grand program, no killer innovation, no lucky break, no miracle moment that creates momentum. Rather, the process resembles relentlessly pushing a giant, heavy flywheel, turn upon turn, building momentum until a point of breakthrough.

6. SMaC (Specific, Methodical, and Consistent) Tenets: A set of operating principles that is the first step in turning strategic concepts into an execution plan. SMaC guides you on what not to do in addition to what to do. SMaC tenets don't change more than 20 percent per year.

7. Blitzscaling: An execution framework that prioritizes speed over efficiency and allows a company to go from "Startup" to "Scaleup" at a furious pace that captures the market. For a startup to move very fast, it must take on far more risk than a company going through the normal.

8. FIRE BULLETS, THEN CANNONBALLS: When you see the enemy ship, you take a little bit of gunpowder and keep firing bullets until one bullet hits the ship. Now, you take all the gunpowder and fire a big cannonball along the same line of sight, which sinks the enemy ship.

9. Play-to-win canvas: Use this to explain your strategy to people who don't have time to read our entire OP1. It contains 1. What is your winning aspiration? 2. Where will you play? 3. How will you win? 4. What capabilities must be in place? 5. What systems are required?

10. Aggregator vs Platform: Aggregators such as Google and Facebook help you get things done. Think of them like Cars. While Platforms are Bicycles. Platforms such as Microsoft and Apple are an aid to humans, not their replacement.

11. Growth Loops: User acquisition funnels are now being replaced with a system of loops. Loops are closed systems where the inputs generate output that can be reinvested in the input. Similar to flywheel but for acquisition/growth.

12. Viral Coefficient (K Factor): The number of new customers the average customer generates. The virality should cover the churn of users. In other words, if k-factor > churn, more users come than users leave, and our product is going to have exponential growth.

13. It's AND. Not OR. The ability to embrace both extremes at the same time. Instead of choosing between X OR Y, they figure out a way to have both X AND Y.Profit AND Growth. Great Customer Exp AND Great Margins. Great Control AND Lean Operations. It's possible to do both.

14. Product-Market Fit: PMF is achieved when your users love your product so much they spontaneously tell other people to use it. It's a binary test. You can always feel product-market fit when it is happening. The customers are buying the product just as fast as you can make it

15. Network Effect: The idea of a network effect is that every additional user increases the value of a good or service. The Internet is an example of the network effect. There could be “internalized” versus “externalized” network effects.

16. Moat: Like the moat that surrounds a castle to provide it with a preliminary line of defense, companies need to have moats or the ability to maintain competitive advantages over their competitors in order to protect their long-term profits market share.

17. High-expectation customer (HXC): is the most discerning person within your target demographic. It’s one who will acknowledge and enjoy your product for its greatest benefit. She is also someone who can help spread the word. Your early adopters are not always your HXCs

18. Bullseye framework: To systematically find the most promising channel. The first step is brainstorming every single traction channel. The second step is running cheap traction tests. The third step is to focus solely on the channel that will move the needle for your product.

19. Level 5 Leaders: Display a powerful mixture of personal humility and stubborn will. They're incredibly ambitious, but their ambition is first and foremost for the cause, for the organization and its purpose, not themselves. They are often quiet, reserved, and even shy.

20. Did he say No?: Usually we pitch what we want, follow up 3-4 times more and then move on if we don’t hear anything positive. Don't move on until we hear the affirmative "NO". Lack of "Yes" is not good enough. We should keep knocking until we hear a strong and clear "NO".

21. Burn Multiple: Calculated as Net Burn / Net New Revenue. How much is the startup burning in order to generate each incremental dollar of revenue? The higher the Burn Multiple, the more the startup is burning to achieve each unit of growth. Burn multiple under 1X is good.

22. Efficiency Score: This is nothing but reverse of burn multiple. It’s a catch-all metric. Any serious problem will eventually impact the Burn Multiple / Efficiency Score by either increasing burn, decreasing net new revenue, or increasing both but at disproportionate rates.

23. Contribution margin (CM): is a product's net sales minus all associated variable costs. The total contribution margin represents the total amount available to pay for fixed expenses and to generate a profit. It can be further divided into CM1, CM2, and CM3.

24. CM1 is sales minus the basic cost of goods sold, discounts and coupons. This is the same as Gross margin.CM2 is CM1 minus logistics, warehouse, CS, payment gateway fees and any other operational variable costs.CM3 is CM2 minus Marketing. EBITDA is CM3 minus indirect costs.

25. Managerial Leverage (aka High Output Management): A manager’s output is the output of all of the people and the teams that report to her. A manager's activity with high leverage will generate a high level of output; an activity with low leverage, a low level of output.

26. Cohort Analysis: Track specific groups of users, known as cohorts, to understand how users engage with your product in the days, weeks and months after you acquire them and, in turn, understand how resilient your growth is.

27. Simon Sinek Circle (aka Golden Circle): There are three parts of the Circle: Why, How, and What.The WHAT represents the products or services a company sells. The HOW is an explanation of why their products/ services are better. The WHY is about what a company believes in.

28. First Who, Then What: Make sure you have the right people on the bus and the right people in the key seats before you figure out where to drive the bus. Always think first about who and then about what. Great vision without great people is irrelevant.

29. Prospecting Pyramid: Arrange your list of leads with high-yield prospects on top and low-yield prospects on the bottom. Start by prospecting from the top of the pyramid.

30. Minimum Viable Product (MVP): Is a proof of concept product that validates your idea before you build a full, mature, stable product. MVP is not just a product with half of the features chopped out but a process that you repeat over and over again till you get it correct.

31. Free cash flow (FCF): measures how much cash is generated after capital expenses such as buildings and equipment have been paid. Operating cash flow is your cash flow from operating revenue minus operating expenses. If you subtract capital spending from this, you get FCF.

32. Cash conversion cycle (CCC): How long it takes your customers to pay you minus how many days it takes you to pay your suppliers. Super-efficient companies have their CCC down to the single digits. At Amazon last year, the CCC was negative 30.6 days.

33. Working Backward: A practice where you start by writing the documents you will need at launch (a Press Release and an FAQ) first and then work backward from there to the product requirements.

34. Free Parking' Business Model: To bootstrap of the “chicken and egg problem”, give away one side of the market for free. Typically it is best to offer the free side to consumers since no one loves “free” more than a consumer. Or offer the service that has lowest marginal cost.

35. Freemium: A variant of the “free parking” model where the company transforms code into the equivalent of marketing spending and “gives away for free” service X to generate qualified leads for interlinked service Y. Freemium works best if service X has network effects.

36. Cash multiplier (CMX): Revenue generated from customer segments over limited time frames or payback window.CMX = Total Rev / Total Customers.Ex: Your total rev for 60 days is $140. If new customers were 1000, your CMX ( 60-day LTV) would be $1,400 in total.

37. Customer lifetime value (LTV, CLTV, or CLV): The revenue generated from the average customer over the course of an average customer lifespan. LTV is the future cash flows over her entire relationship with the company.

38. High Output Meetings: Is a medium through which managerial work is performed. It is a way to supply information and know-how, to explain the way of doing things, and to help make decisions. We need to make meetings as efficient as possible. Not fight the need for the meetings.

39. Task relevant maturity (TRM): for a team member is a combination of the degree of their achievement orientation and readiness to take responsibility as well as education, training, and experience. All of this for a particular task.

40. UI Complexity Score: UI needs to be as simple and functional as possible. To calculate the complexity score, you add up a point every time you used a new font, font size or colour in the UI. The total score is the complexity score. A single page needs to stay below five point.

41. Conversion rate optimization (CRO): Science behind understanding why your visitors are not ‘converting’ into customers, and then improving your messaging or value proposition to increase this rate of conversions.

42. Blue ocean strategy: Red oceans are all the industries in existence today – the known market space. Cut-throat competition in existing industries turns the ocean bloody red. Blue oceans are all the industries not in existence today – the unknown market space.

43. Productive Paranoia: You assume that conditions can unexpectedly change, violently, and fast. You obsessively ask, What if? By preparing ahead of time, building reserves, preserving a margin of safety, you handle disruptions from a position of strength and flexibility.

44. Inbound marketing: A marketing strategy that attracts customers by creating valuable content and experiences tailored to them. While outbound marketing interrupts your audience, inbound marketing forms connections they are looking for and solves problems they already have.

45. Ramen profitable: A startup makes just enough to pay the teams' expenses. Traditional profitability means a big bet is finally paying off, whereas the main importance of ramen profitability is that it buys you time.

46. Category Management: The process of managing categories as independent business units, in a way that enables maximum consumer appeal while maximizing profits. Category Management aims to provide customers with what they want, where they want it, and when they want it.

#productmanagement #startup

SIM - Growth without Burn

Want to grow your product fast without burning money? Here is a simple framework for that. I just call it SIM (Scalable Input Module) framework.

1. A SIM is an input unit that you can deploy to get to your output goals. For example, a sales representative (SR) is a SIM who gets you the output which is Sales or Revenue.

2. A SIM should be input and not an output. So a salesperson is a SIM but deals closed is not. A distributer is a SIM but the distribution GMV is not.

3. A SIM should be completely within your circle of control. Again an SR is a SIM as hiring one is within your circle of control. Getting featured in the App Store is not as that is not under your control.

4. A SIM should scale proportionately. If one SR gets 10K sales daily, 10 salespeople can get ~80-100K daily. If $1 spend in Facebook ads gets 10K daily, $10 spend in FB ads will not get you 80-100K (try it and see). So FB ad is not a SIM while an SR, a Lenskart store or a Swiggy hub is.

5. A SIM is replaceable. If one sales rep doesn't work, you can replace them with another. If FB ads don't work or if FB blocks you, you can't replace FB.

6. A SIM should have revenue directly attributable to it. So an SR is a SIM but an Engineer is not.

7. Sometimes, a SIM can be a combination of input units. For example, one SIM could be a team of a market research person, a lead generation person, a salesperson and a closer.

8. It's a good practice to combine both supply and demand for a SIM wherever applicable. For example, one sourcing associate and one sales rep together can constitute a SIM.

9. A SIM should be Contribution Margin positive. For example, a sales rep should generate enough margin that could pay their salary before they can be considered a SIM. A store, distributor, franchisee or a hub should be profitable for it to be a SIM.

If you can create a SIM for your product, then it's just a matter of cloning more SIMs to grow your product. If you don't have a SIM for your product, then there is a good chance that your growth might not be under your control. So don't start scaling until you have identified a SIM.

#startup

Managing PMs

There are a lot of tips available for product managers. What about founders and CEOs who are managing PMs. Here is a thread for them -

1. Think of your PM as the CEO of the product and you as a board member. Allow them to operate independently and poke your nose only when absolutely required. Better to agree on items upfront where they need to take your input (similar to AVMs in SSHA).

2. Have a working agreement with your PMs on reports, metrics, communication, prioritisation framework etc. Don't surprise them every time with a new modus operandi.

3. Allow your PMs to be straight-shooters. They shouldn't need to sugar coat their message to you. This pretty much depends on how you react when they bring bad news. If you bark at them every time they bring you unpleasant news, they will just stop bringing them to you.

4. Don’t make your PM as your messenger. If you want to give some bad news to your customers, your boss or other leaders in your company, you do it yourself.

5. Don't force your PMs to give optimistic rosy pictures to your customers and sales team. Your PMs are always paranoid. They constantly think about what can go wrong and don't want to hide issues. If a GM stops the PMs from giving visibility, they have to go against the GM's wishes.

6. Remember that the main job of your PM is to be customer advocate. They do injustice to their customers if they are not strongly arguing for their customers. Understand and coach their adjacent teams to not take this personally.

7. PMs are not your assistants. Don't expect them to take notes in your meetings and send them back to you. Don't always expect them to keep you in the loop. Don’t waste their time asking to find a suitable time that works for everyone and send calendar invites. Don't ask them to spoon feed you.

8. Don't expect your PM to butter up people and get things done. Don't make them feel that other team members are doing a favour to them when those members are just doing their job.

9. Allow your PMs to use sticks if carrots don't work. They have full responsibility without any authority and sometimes they need to escalate, complain, make noise. So be it.

10. PMs are here to get things done, whatever it takes. Sometimes, they need to bulldoze some lethargic blockers on their way. Its' a sad reality but sometimes they need to behave like a**holes. Allow that as long as it's exception and not the norm.

11. Take feedback on PMs from other teams with a pinch of salt. Only bad product managers try to please everyone.

12. PMs are not here to baby-sit Engineers. They need not worry about Engineers working on maintenance and bug fixes instead of new development. They are not going to use hot tech just to please Engineers. They don't care about the career progression of the Engineers. That's the job of Engineering Managers.

13. Engineers don't get to prioritise. It's the job of PMs.

14. Your PMs are not magicians. They can't add your last minute features without affecting deadlines. They can't deliver a project earlier than what Engineering Manager has quoted. They can't debug issues themselves. They do sleep sometimes.

15. Don’t bring your org politics to your PMs. PMs don't care about team boundaries. They go to any team within any org to get things done. They don't mind bypassing managers and work with the individual contributors.

#startup #product management #product managers

Who is a TPPM?

My first title when I joined Amazon (15 years ago) was TPPM - Technical Product Program Manager. What does that mean? Well, here is a thread on types of PMs. I have generalised a lot for brevity - obviously, there are a lot of nuances and different companies use these titles differently.

1. Product Managers own the definition of the product. They decide the "what" and "why" to build. They communicate the requirements through product specifications.

2. Project Managers own the execution of a project. They own launching the product on time as per specifications.

3. Program Managers own a program. A program could span across multiple projects and products and so they work with multiple project managers and product managers. Also, a program is ongoing while a project has a definite start and end.

4. Marketing Managers own the GTM (Go To Market) strategy and ongoing adoption of the product. They own taking the feedback from the market back to the product team.

5. Engineering Managers own hiring, training and allocating engineers to multiple projects. They own the people management and career growth of their Engineers.

6. Business managers own the P&L of a business or a business unit. Category Managers are a type of business managers who own the P&L of a category.

7. Technical Program Managers create technical requirements for a product. They take the product specifications and convert into more detailed technical specs. 'Program' here refers to 'programming' and not the collection of projects. They are usually part of Engineering. (In some tech companies, this role is called 'Program Manager').

8. Product Marketing Managers are product managers who are more outward facing. They own the 'Brand' of the product but leave the actual definition of the product to a different product manager. Sometimes they own the 'Why' while product managers own 'What' and program managers own 'How'.

9. Technical Product Managers are product managers who are more inward facing. They have a tech background and work on technical aspects of a product. They could own a Tech Platform that multiple products use or own products internal to the company. Sometimes, they are just product managers working on a core tech product.

10. Product Program Managers own both the definition and the delivery of a product. They create requirements and also work with engineering to deliver them.

11. TPPMs (Technical Product Program Managers) have a tech background and own both the definition and delivery of a product.

12. There are other titles such as Delivery Manager and Development Manager. They might refer to Engineering Manager or Program Manager depending on the company.

13. Amazon also used the title 'Project Program Manager' which I never understood the need for. I think that title is no longer used. In fact, Amazon has simplified the PM titles a lot recently.

So who is the mini-CEO? In some cases, Program Managers are the mini-CEOs hiring Product Managers to define products. In other cases, Product Managers act as mini-CEOs hiring program and project managers to manage the delivery. There are cases where business managers or category managers are the mini-CEOs hiring program managers and product managers under them. Their case became strong when a Head of Category became the CEO of the largest startup in India (wink).

#startup

Mental Model for Founders

As founders are going through difficult times now due to the COVID-19 situation, it helps to have the right mental model. A good mental model avoids the rollercoaster emotions of fear/euphoria. It prevents bad decisions arising from those emotions. Here are some pointers on that.

1. Accept responsibility. YOU are the reason for everything that is happening in your company. It’s not the market, partners, investor, investment climate, or team but YOU and how YOU handled them. Even if you believe otherwise, just assume and accept this.

2. Be in 'Form'. The objective of the mental model is to get YOU in Form (aka 'Flow'). Like a cricketer, however great your skills, you will still fail if you are not in Form. By Form, I mean you can execute effortlessly without fear or excitement. Effortless is the keyword here.

3. Leave Beliefs at bay. It doesn't matter what YOUR beliefs are or what you learned from the past. It doesn't matter what your 'Gut feeling' says. Don't predict the future based on your beliefs. It's just your Ego. Truth is unfolding in front of you. Just observe and react.

4. Don't have 'Favorites'. Scrap your favorite project. Chuck your favorite framework. Leave your favorite team member. React objectively, honestly, and truthfully.

5. In uncertain times like this, take one day at a time. What matters is 'Now' and how you respond to today's events. Forget the past and forget the future. Focus on 'Present'. Forget your 5-year business plans and plan for today.

6. Don't be a naive Optimist. You are not seeing the Reality and also preventing your team from seeing the Reality. And you will end up making hefty promises to partners that you can never fulfill. Hope or optimism is not going to solve your problems.

7. Self-pity is a crime. A Founder should quit instead of getting into this vicious loop of 'Self Pity'. You will never win without a winning attitude and you can never get that attitude with Self-pity drowning you.

8. Fuck Fear. If you are afraid of every step, how can you ever take a bet? Your job continues to be taking calculated bets and doubling down on winning bets. Don't let fear stop you in these uncertain circumstances.

9. A winning attitude is not Greed. It's not the same as being hopeful. It's about playing to win. It's about execution without the fear of failure or excitement of success.

10. Stick to the input process. The output is outside your circle of influence. Input is how you should measure yourself. Would you do anything better if you have to re-do a process or re-live a day? If no, you have succeeded irrespective of what the outcome was.

11. Unshackle the constraints. Don't promise anything to team, partners, or investors. Just tell them your process and get alignment there. Freedom gives you a lot of options in these circumstances.

12. Last - Don't retaliate or take revenge steps. Be it Employees, Partners, or Investors. They are also facing tough times and might hurt you in the process of saving themselves. Revenge will cost you more than to them.

#startup

Unconventional Lessons from Shailendra

Sharing a few unconventional lessons we learned working with Shailendra Singh, our investor/mentor. He is the MD of Sequoia India and is the only Indian VC to be featured in the list of top 100 VCs in the world.

1. "Did he say No?" Usually, when I meet investors or potential partners, I follow up immediately and then 3-4 times more and then move on if I don’t hear anything positive. Shailendra turned this on its head by suggesting to not move on until I hear something negative. Whenever I said the person doesn't seem interested, Shailendra will say "Did he say No? If not, you can't give up".

2. "Take one day at a time." Voonik has gone through a lot of tough phases. We have been on the verge of giving up multiple times. Shailendra narrated how he dealt with a personal tragedy by taking one day at a time. Among VCs asking for 5-year business plans, he asked us to think just about the next day. That has helped us immensely when we could not see how the next month/year was going to turn out.

3. "I have seen too many companies succeed after being at the brink of shut down for us to give up". Another message that kept us going during tough times. The main lesson is that even great companies had tough times and such times are not indicative of the potential of the company.

4. "You will be surprised how elastic teams are." We did the same mistake which every startup makes in its growth phase - Over-Hiring. When Shailendra first mentioned about how same speed and efficiency can be done with one-tenth of the team size, I thought he was joking. But I am genuinely surprised about how true this is. I have even seen one person doing some task (with a lot of tech tools, of course) that required 25 people at some point.

5. "It's AND. Not OR." While I read this concept in Jim Collins' books, I didn't internalize it until I saw how Shailendra used this principle. Profit AND Growth. Great Customer Experience AND Great Margins. Great Control AND Lean Operations. It's possible to do both and we have seen it first-hand at Voonik.

#startup

Founder Tenets

Years ago when we started Voonik, I and @hinava came up with tenets to ensure founder integrity. I am glad that till date we have stuck to these principles without a single exception.

Company first. Stakeholders second. Founders last.

100% of time should be dedicated to Voonik and only Voonik. Don't officially take up any advisory roles, mentorship roles or consultant roles. Exceptions allowed: Spending few hours to help budding entrepreneurs, preferably over weekends.

Be fully invested in Voonik and only Voonik. Never own conflicting shares. (I sold Amazon shares as they conflict while Navaneeth can hold FreshWorks shares)

Avoid any related party transactions. If super-beneficial to the company, have the other founder deal with the related party.

Don't treat venture money like your own money. You should be more careful. This is other people money.

Never take any money off the table before your employees and investors have made money. Have a conversation with the other founder if you need money for family needs. Instead, feel free to ask for more ESOPs based on successful milestones.

Never be the highest paid person in the company. (We had 5% of people getting more salaries than the founders)

Whenever you ask for a sacrifice from your employees, you be an example by making that sacrifice first.

Go through as much pain as you can personally but never let any pain or stress leak to the family. Safeguard your family from workplace stress and financials.

#startup

Why can't we follow advice, even if we know it's good?

Nowadays founders have a lot of resources to learn what makes a startup successful. Whether it's Paul Graham, Ben Horowitz or Peter Thiel - the advice from them are mostly consistent and even we founders acknowledge them as a universal truth. Yet, we don't follow them. Why?

At first thought, it might look like we did not follow them because they were not applicable to our unique startup. But the reality is that most of the advice is generic and *does* apply to our startups. Then why don't we follow them?

Let me give a couple of personal examples on advice I did not follow before and ended up following much later on, once I internalized what that advice meant.

------

Example 1: I got an advice that only losers have an advertising strategy and winners have a product strategy. Dominant brands like Google, Facebook, YouTube, Twitter, Zara, and WhatsApp have never needed to advertise to get dominance. Either your product has to be talk-worthy or at least stories about your product need to me remark-worthy.

This is an advice that sounds right. Yet, I did not follow it. I did not believe my product is as remarkable as Google or Facebook that it can grow just through word of mouth. That does not mean I can give up on my startup. I need to find quick ways of getting the word out and get traction.

So, I started advertising. Initially, it seemed fine. We then realized that we have to keep increasing our advertising budget to scale our business. We did not have that kind of money. So we started focusing on improving conversion and retention. We also simultaneously kept increasing advertising budgets in small increments. We somehow were getting the growth rates we wanted.

When I looked back after a few months, I realized that the big bumps in adoption came from product features and never from marketing tactics. I could have just focused on the product and I would have got the same growth, without needing to spend ad dollars. Actually, the growth could even have been better, as the energy, resources and focus from ads would also have been spent on product

I got the aha moment - I need not be a Google or Facebook; for any product, the growth is going to come from product changes and not from advertising dollars.

Example 2: Paul Graham's advice that having few users love you is better than having a lot of users who sort-of like you.

Again, an advice that I chose not to follow. We started our service for both men and women (I also wanted to include kids ;) ). I wanted to launch it in all platforms - desktop, Android, iPhone, windows. I even wanted to go international.

The thought was that the word of mouth or ads are going to reach across the segments - why waste the reach that we got for a segment when getting the word out was the biggest challenge? For example, if there is a press coverage about us, 50% of those who read will be men and we will lose them if we don't have products for men. That press coverage took a lot of effort from my side, why waste half of that reach? Similarly, the users reading the coverage might have a windows phone or iPhone and what if I don't support that platform.

We learned the hard way that it was impossible to build a product that satisfies all the segments. We decided to focus only on Indian women and focus only on Android. Still, there are miles to go in addressing Indian women in android before we could move to other segments. Now I understand the meaning of starting small and having few users who love us.

----------

Nowadays, whenever I hear or read an advice, I smile - I know that I have just read the advice but don't understand what it means until that "aha" moment arrives.

#startup

How to use metrics to woo Investors, build your business, and live happily ever after?

This article is based on a talk I gave at Microsoft Ventures under the topic “Metrics that Matter”.

Metrics, why Should I care?

People track metrics for multiple reasons. Most use it to keep a tab on the business. I use it for a related, but slightly different reason.

As founders, we are primarily story-tellers. We are constantly selling our story to investors, partners, employees, press, and who not. For me, Metrics is nothing but a story, told in numbers. This story is much more interesting and credible to investors and business stakeholders and so totally worth it.

It is important to understand the story that metrics are telling you – every day, every week, every month. Yes, this is the best way to keep a close tab on your business. But, it is even more important to practice how you will convey that story to the outside world — every day, every week, every month. We should be picking the interesting stories and should be planting seeds for better stories.

The Keystone Metric

There should be one metric that defines your company. This is the one metric why the investor is interested in you. This is the one metric around which the entire organization rallies around.

Let's say, your metrics A, B and C are on par with industry. D and E are below par. F is above par. Which one would you focus on?

Well, what’s the use in putting all the hard work, just to become on par with industry? Pick a metric should make you stand out from the crowd.

Remember that Keystone metric might not be in the metrics that you are currently tracking. Look deeper. Look everywhere. Also, remember that Keystone metrics change over time. My company, Voonik, had different keystone metrics across stages: MVP stage, Building Block stage, and Scaling Stage. Keystone Metric could also change in the same stage, based on new information and changing priorities.

And your keystone metric should grow. As Paul Graham says, “ Pick a growth rate and then just try to hit it every week. If you hit that number, you’re successful for that week. But if you don’t hit it, you’ve failed in the only thing that mattered.”

You might have a best-in-class conversion rate, awesome growth rate but those are meaningless, without the scale. You can grow from 1 to 2 orders and claim a 100% growth rate. So, pick stories/metrics that makes sense for your stage. Percentage of positive reviews or repeat usage is a better story than the growth rate for a startup without the scale.

What should I track?

In addition to keystone metric, you should be tracking a lot more metrics. You might not know all the metrics you would need in future. So better enable tracking for all possible metrics you could think of. Send events for every action.

If possible, try creating a single repository. At Voonik, we dumped all the data from Google Analytics (GA) and FreshDesk into our database so that we can combine them with Voonik data to create meaningful reports that are not possible in GA.

Remember, a sub-par metric is not bad, as long as you planned it that way. A phenomenal metric is not good if you don’t understand why is it that way. Don’t miss human metrics such as Productivity metrics, Effort Burn rate etc. And don’t miss metrics outside your product such as Metrics for the platforms and services you use

Sometimes, you need to invent a metric. At Voonik, we have a custom metric to measure the effectiveness of our algorithms. At Amazon as well, we had a number of invented metrics.

The Metrics Roadmap

Before creating a project roadmap, create a metrics roadmap. Assign an owner for every important metric. Have them report the story every day. Assign projects/tasks to the person whose metrics will be affected by the project. Make sure every metric owner understands their part of the story and the goals.

Don’t go by industry benchmarks. Different companies measure the same metric differently. For example, How do you measure conversion rate? And don’t believe the analyst reports. Go by what is needed for your business model.

Input vs. Output Metrics

Metrics that are under your circle of influence are input metrics, such as Number of products added, Number of steps in checkout, the design of the website.

Metrics that are outside your circle of influence are output metrics. They are influenced by input metrics, such as Number of orders, Sales Volume, Contracts.

Assign only Input metrics to subordinates. They will be frustrated if their efforts don’t impact the output metrics and will start making excuses for not meeting goals.

And that’s it for today. Will update the article as I learn more on this topic.

#startup

Customer Acquisition - Free Channels vs. Paid Channels

From the time I started up, I have met many folks and they can typically be classified into two schools of thoughts regarding customer acquisition. The first set of people believe religiously that you create remarkable product & content, and that will get you the distribution you need. They also believe that continuous experiments and growth hacks are the only fuel required to push your remarkable product & content. (Note: Remarkable customer service is part of a remarkable product). These people are usually the darlings of the media and certain VC firms, which makes sense as being active on media and social networks are part of their growth strategy.

The second set of folks think that building free channels is a waste of time (except for direct traffic). Their reasoning is that free channels (including SEO) don't scale beyond a point. So you will be stuck at a point where you might not be able to grow further. Even viral channels might give you a peak and then start dropping. So this second set of folks' goal is to find a scalable paid channel where you can get back the money that you invested through the profit you make through that channel, aka operational break-even. Unlike free channels, this will allow you to scale faster by investing more in that channel.

Direct Traffic is a separate category by itself and should not be clubbed with other free channels. Sometimes, when most of the people talk about a remarkable content/product, they club direct traffic complicating the discussion. Even those who purely follow the paid channel strategy cannot survive without direct traffic, which in turn is driven by a remarkable product. So assuming direct traffic is a must (and so, in turn, remarkable product in a must), the discussion should be between other free channels and paid channels.

I think we need a combination of paid channels with scalable free channels. By scalable free channels, I mean channels such as Facebook Pages and Newsletters which can be scaled easily.

#startup

Stop being a startup

[Published in Yourstory on 19th Apr 2017]

It is important for a startup to act and execute like a mature company. In a world that is glamorizing startups, why is this important? Are there any particular downsides to acting like a startup? Here is a take on why not being a startup might be a good thing.

“Voonik is no longer a startup. It’s a mature company and let’s act like that.” When I said this, it triggered a series of discussions and pushbacks. How will a company benefit from not being a startup? Am I encouraging bureaucracy and office politics? Will things move slowly now and will faster startups overtake us? How can Voonik talk about not being a startup when Jeff Bezos is talking about still being day one for Amazon?

The first aspect of not being a startup is about growth. According to venture capitalist Paul Graham, a startup is a company designed to grow fast. Every startup creates a culture and systems that foster growth. However, chasing growth blindly can only lead to a crash. Product managers will start compromising customer experience to improve conversion. Category managers will start promoting beautiful looking products that might look a lot different when delivered. Account managers will go after whales instead of nurturing upcoming accounts. Logistics managers will be recklessly adding PIN codes without worrying about the capabilities of the delivery partners. The sad part is that the same things that give growth in short-term will always hurt growth in the long term. We will end up with low retention, bad economics and with being dependent on a few suppliers. What we need is a high-quality growth that can be sustained over time. This requires a culture of patience than a culture of restless growth. Patience is a moat by itself.

The second aspect is about bets. A startup roadmap is always full of silver bullets. Startups try tens of things hoping some feature will make them the next Facebook. There is nothing that is too risky or technically not feasible. Startups have to try everything. They are satisfied only when they see the data that no one cares about the latest feature launch. Go big or go home.

Contrary to what startups think, a good roadmap is mostly consisting of lead bullets with one or two silver bullets. There should be conviction behind roadmap. The roadmap should lead you to where you want to go. Roadmap can’t be a hit or miss list of kitchen sink features. Every feature will need to be launched for a small set of users, groomed based on feedback, and launched to the wider audience in a careful and thoughtful manner. Iteration is key to startup roadmap and not bets.

The other aspect is the process or lack of it. Being a startup is a ticket to be undisciplined and not to follow the process. We don’t need heroic engineers who can fix any production issue; we need proper monitors and alarms. System downtime due to load is not a good problem to have. It doesn’t work that designers work at day and engineers code at night. Asking for status is not bureaucracy. QA is not optional. Anything that is not scalable and not repeatable is not useful. Anything that is repeatable should be a process with proper monitors and alarms. And a good plan is a pre-requisite for good execution.

Then there is the focus on outputs. In startups, everyone is chasing the output metrics such as GMV, DAU, MAU, revenue, gross margin, and what not. Since everyone in a startup has visibility into company metrics, they become concerned when output metrics not grow and start a vicious cycle of deploying all their efforts on keeping the output metric up and to the right. The reality is that most team members are clueless as to how to affect an output metric. By definition, the output metric is outside their circle of influence. A good, confident team focusses on input metrics. You grow the selection, improve the quality, lower the price and you will see that GMV will automatically improve. Selection, quality, and price are all within the influence of the team members, and the team can be rightfully made accountable for those.

Having said that, there is one grave danger in not being a startup, and that is the danger of stopping to dream. If we avoid that pitfall and still be dreamers and be nimble to external factors, we should be good. Patience, a roadmap of lead bullets, process, and focus on input metrics can do only good. If that doesn’t come under the definition of a startup, then so be it. We will stop being a startup.

#startup

My Talk to Students

#interview

Trending Blogs

Recently Viewed Blogs

@Sujayath