company-specific prep

Generic interview prep gets you generic results. I have watched thousands of candidates prepare for “PM interviews” as though every company runs the same process. They do not. Google cares about things Meta does not. Amazon tests for things Flipkart takes for granted. And a Series B startup in Bangalore evaluates you on dimensions that would never show up in a Big Tech loop.

The difference between getting rejected at a company and getting an offer often comes down to one thing: did you prepare for that company, or did you prepare for “PM interviews” in general?

This page is your company-by-company field guide. Not exhaustive — companies change processes constantly. But the underlying DNA of what each company values? That changes slowly. And that is what you need to understand.

The fundamental mistake

Most candidates prepare horizontally. They practice product sense, then metrics, then behavioral, then strategy — and they apply that same preparation to every company.

Smart candidates prepare vertically. They pick a target company, understand its interview DNA, and then map their preparation to what that specific company actually tests.

Google: Product sense is the whole game

Google’s PM interview is designed to test one thing above all: can you think about products with depth and nuance?

The loop (typically 4-5 rounds):

• 1-2 rounds of Product Sense (design a product, improve a product)
• 1 round of Analytical / Metrics
• 1 round of Leadership & Drive (behavioral)
• 1 round of Technical (varies by role — sometimes strategic)

What Google actually evaluates:

Google interviewers score on four dimensions: Googleyness & Leadership, General Cognitive Ability, Role-Related Knowledge, and Product Sense. But the weight is disproportionately on product sense. You can have a mediocre behavioral round and still pass. You cannot have a mediocre product sense round.

Product sense at Google means:

• Defining the user with uncomfortable specificity. Not “young professionals” but “solo commuters in cities with poor public transit who use Google Maps for the same route daily.”
• Identifying the right problem before jumping to solutions. Google interviewers let you talk. If you jump to features in the first two minutes, they have already started scoring you down.
• Trade-off articulation. “We could do X or Y — here is why I would pick X, and here is what we lose by not doing Y.” This is the move that separates strong hires from borderline ones.

What trips up Indian candidates at Google:

Richa Pareek, a Google PM in San Francisco, told us that product sense and risk management are the two dimensions that matter most. Indian candidates tend to over-index on solution breadth — listing many features — and under-index on solution depth. Google wants you to pick one direction and go deep. Three well-reasoned features beat fifteen surface-level ones every time.

The Google-specific prep move: Pick 5 Google products you use daily. For each one, identify the top user pain point, design one solution, and articulate the trade-off you are making. Practice explaining why you chose this solution over alternatives. That exercise, repeated 20 times, builds the muscle Google is testing.

Amazon: Leadership Principles are the scoring rubric

Amazon is the most legible interview process in tech. They tell you exactly what they test. Most candidates still do not prepare for it properly.

The loop (typically 5-6 rounds):

• 2-3 rounds of Behavioral (mapped to Leadership Principles)
• 1 round of Product Sense / Product Design
• 1 round of Metrics / Analytical
• 1 Bar Raiser round (mostly behavioral, cross-functional interviewer)

What Amazon actually evaluates:

Amazon has 16 Leadership Principles. Every interviewer in your loop is assigned 2-3 principles to evaluate you on. The Bar Raiser — a specially trained interviewer from outside your hiring team — has veto power and evaluates across all principles.

This is not a metaphor. Your interviewer literally has a scorecard with principle names on it. They are listening for evidence that you have demonstrated those principles in your past work.

The principles that matter most for PM roles:

• Customer Obsession — Every PM answer should start with the customer. Not the business. Not the technology.
• Bias for Action — Amazon rewards speed. “We had incomplete data, so I made a decision with what we had and set up a mechanism to course-correct” is a strong Amazon answer.
• Dive Deep — They want to see that you go beyond the summary. If you managed a metric, they want to know the specific SQL query you ran, the specific segment that was underperforming, the specific root cause.
• Earn Trust — Stories about admitting mistakes, disagreeing with your manager, or delivering bad news. This is the one most candidates do not prepare for.

The Amazon-specific prep move: Create a spreadsheet. List all 16 principles in column A. For each principle, write 2-3 stories from your experience in column B. Tag each story with the STAR structure. You need at least 25 stories total — because a 6-round loop with 2 questions per round is 12 behavioral questions, and some stories will not fit the specific question asked.

People who get Amazon offers prepare this spreadsheet. People who wing behavioral rounds do not.

Meta: Execution speed and impact measurement

Meta’s PM interview has shifted over the years, but the core thesis remains: can you ship things that move metrics?

The loop (typically 4-5 rounds):

• 1 round of Product Sense
• 1 round of Execution (this is Meta-specific)
• 1 round of Leadership & Drive
• 1-2 rounds of Analytical / System Design

What makes Meta different:

The Execution round. This is not about “how would you build X?” It is about “you are the PM on this product, this metric dropped 15% this week — what do you do?” Meta wants to see your debugging instinct, your ability to triage, and your prioritization under pressure.

Meta PMs live and die by metrics. The company culture is deeply quantitative. If you cannot define a success metric for your proposal, and explain how you would instrument it, you are not speaking their language.

The Meta-specific prep move: Practice metric decomposition. Take any product — Instagram Reels, WhatsApp Status, Facebook Marketplace — and break down the primary metric into a tree. DAU = New users + Retained users. Retained users = (Last week active) x (Retention rate). Now the metric drops. Where in the tree did it break? Practice this until you can decompose any metric in under 60 seconds.

Flipkart: The Indian e-commerce gauntlet

Flipkart’s interview process borrows from both Amazon and Google, but has its own distinct character. It is the most common “big company” PM interview in India, and it is harder than most candidates expect.

The loop (typically 4-5 rounds):

• 1-2 rounds of Product Design / Case Study
• 1 round of Guesstimates + Analytical
• 1 round of Behavioral + Leadership
• 1 round with a senior leader (VP / Director)

What Flipkart actually evaluates:

Flipkart interviews are heavily context-dependent. They expect you to understand the Indian e-commerce landscape — not in theory, but in specifics. If you are asked to improve Flipkart’s app, they expect you to know that tier-2 users on Android devices with limited storage are a real segment, that cash-on-delivery dominance creates specific product challenges, and that vernacular language support is not a nice-to-have but a growth lever.

The senior leader round is where most candidates fail. This is not a product question round — it is a “do I want to work with this person?” round. They test for intellectual curiosity, the ability to hold a nuanced opinion, and comfort with ambiguity. Rehearsed answers fail here. Genuine thinking survives.

Flipkart also has its own leadership principles. They are less codified than Amazon’s, but the interviewers evaluate on similar dimensions: customer-first thinking, ownership, bias for action, and the ability to operate in India’s complex market.

The Flipkart-specific prep move: Study Flipkart’s last two earnings calls or investor presentations. Understand their strategic bets — quick commerce, travel, grocery. When you answer product questions, ground them in Flipkart’s actual strategic context. “If Flipkart’s priority is growing in tier-3 cities, then the right user segment for this feature is…” — that sentence wins rounds.

Razorpay, Zerodha, CRED: The fintech and B2B interview

Indian fintech and B2B companies interview differently from consumer companies. The questions sound similar, but what they are evaluating is fundamentally different.

What fintech companies test:

• Regulatory awareness. If you are interviewing at Razorpay and you do not know what UPI autopay is, or what RBI’s tokenization mandate means for merchants, you are not prepared. Fintech PMs operate within regulatory constraints. The interview tests whether you understand that constraint landscape.
• B2B product thinking. Consumer PM interviews ask “who is the user?” Fintech PM interviews ask “who is the buyer, who is the user, and how does the integration work?” You need to think in terms of APIs, developer experience, merchant onboarding funnels, and enterprise sales cycles.
• Platform thinking. Razorpay is a platform. Zerodha is a platform. CRED has platform elements. The questions test whether you can reason about multi-sided systems — incentives for each side, chicken-and-egg problems, and network effects.

The fintech-specific prep move: Sign up as a merchant (test mode) on Razorpay or Cashfree. Go through the onboarding flow. Try integrating their API. Read their developer documentation. This gives you more interview ammunition than any amount of abstract preparation. When the interviewer asks “How would you improve merchant onboarding?” you can speak from actual experience, not imagination.

Early-stage startups: The unstructured interview

If you are interviewing at a Series A or B startup in India — say a D2C brand, an edtech company, or a vertical SaaS player — throw out everything above. The process is entirely different.

What the interview looks like:

• No fixed format. You might talk to the founder for two hours. Or do a product teardown on a whiteboard. Or get a take-home case study.
• The founder or CPO typically has the final say. There is no committee, no Bar Raiser, no structured rubric.
• They are testing for fit as much as skill. “Can this person operate in our chaos?” matters more than “Can this person structure a product sense answer?”

What actually matters:

• Ownership stories. Not “I contributed to a cross-functional initiative.” Rather: “I noticed this was broken, I built a prototype over a weekend, I showed it to three users, and here is what happened.” Startups want people who do not wait to be told.
• Speed of thinking. The founder will throw curveballs. “What if we pivot from B2C to B2B?” “What if our CAC doubles next quarter?” They want to see you think on your feet, not recall a framework.
• Domain knowledge. If you are interviewing at a healthtech startup, know the India healthcare landscape. If it is edtech, understand the NEP implications, the tutoring market dynamics, the content-vs-platform debate. Generic PM answers die in startup interviews.

The company research protocol

Regardless of which company you are targeting, there is a research protocol that takes four hours and gives you more leverage than twenty hours of generic practice.

Hour 1: Product immersion. Use the product. Not casually — systematically. Go through the core user flow three times. Note every friction point, every delight moment, every design decision that surprised you. Write these down. They become your interview ammunition.

Hour 2: Business model and strategy. Read the last earnings call transcript (public companies) or the last funding announcement (startups). Understand what the company is betting on for the next 12 months. Every product question you answer should be grounded in this strategic context.

Hour 3: Interview process research. Talk to people who have interviewed there recently. Glassdoor reviews are fine for structure, but conversations are better for nuance. “The Bar Raiser asked follow-up questions about metrics for fifteen minutes” is intel you cannot get from a blog post.

Hour 4: Culture and values mapping. Read the company’s engineering blog, product blog, and leadership talks on YouTube. Amazon writes about leadership principles constantly. Google publishes papers about product development. Flipkart’s leaders speak at conferences. This material tells you the language the company speaks. Use that language in your answers.

The meta-skill: reading the room

Every company has a culture that shows up in the interview room. Amazon interviewers will probe your stories for three levels of detail — they want the specific number, the specific date, the specific outcome. Google interviewers will stay quiet and let you talk for five minutes to see if you can structure your own thinking without prompting. Startup founders will interrupt you mid-answer to test if you can pivot.

None of this is in the job description. All of it determines whether you get the offer.

The candidates who prepare company-by-company — who treat each interview as a distinct challenge with distinct success criteria — are the ones who collect multiple offers. Everyone else is playing a lottery with better odds.

Company-specific prep is not more work. It is less wasted work.

Where to go next

• Nail the product sense round: Product Sense — the five-step structure that works for every product question
• Tell your story: Tell Me About Yourself — the first 90 seconds set the tone
• Handle behavioral rounds: Behavioral & STAR — structuring stories that show judgment, not just activity
• Know the interview types: Interview Types — what each round actually tests