Razorpay PM Interview — Phased Payments Launch

It tests whether you can sequence a phased launch of a new payments product for Indian merchants rather than describe a single big-bang release. The interviewer probes how you segment merchants and pick a beachhead, how you define a launch metric with a numerator and denominator, what release criteria gate each phase, and what number triggers a rollback. India payments depth matters: UPI economics, MDR, NPCI caps, settlement, and KYC are treated as part of the plan, not background. You are also tested on whether you hold your reasoning when the interviewer pushes back with a new constraint.

Start by clarifying the product and the merchant problem before proposing anything. Pick a specific beachhead merchant segment and say why that one and not the others. Then sequence the launch in phases with an explicit gate between each, name the metrics you will watch with their exact denominators, and state the rollback trigger before you describe scaling up. Close by naming owners and the dashboard you would instrument before launch. Depth on one launch path beats a broad tour of every possible step.

The frequent failures are jumping to a solution before segmenting merchants, naming an adoption metric with no denominator or guardrail, proposing one launch with no phase gates and no rollback trigger, and treating RBI aggregator rules and KYC as someone else's job. Another common miss is a launch plan with no named owner and no dashboard defined before launch. Candidates also lose ground when they cannot defend a metric or a gate under pushback and either fold instantly or get defensive instead of reasoning.

The AI plays a fictional Group PM at a fictional commerce company, not a real Razorpay employee, so there is no bias from a name or a referral. It probes every claim for the underlying number, never praises mid-interview, and stays in character as an operator with a launch on the line. It will introduce a mid-conversation constraint, such as an NPCI rule change, to see how you adapt. Unlike a real loop it gives no outcome signal during the session; the feedback arrives as a structured scorecard afterward.

Your transcript is scored against role-specific dimensions such as launch sequencing, metric definition with denominators, rollout gating and rollback discipline, India payments domain depth, and how you defend decisions under pushback. Each dimension has observable anchors, so two reviewers should land within a narrow band. The scorecard names the specific moment a metric lacked a denominator or a gate could not be defended, rather than giving a single overall grade, so you know exactly what to fix before the real round.

Do not start pitching. Spend the first two minutes clarifying what the new payments product is, which merchants it serves, and what success means to the business before you propose a launch sequence. Ask one or two sharp diagnostic questions about the merchant base or the constraint. State your beachhead segment and why. Signaling that you diagnose before prescribing is exactly what differentiates strong candidates in the opening, and the interviewer opens up with more context when you do.

Expect a curveball such as an NPCI volume cap or a six-week general-availability deadline introduced partway through. Do not abandon your goal or restart from scratch. Acknowledge the constraint, state which part of your plan it breaks, and rework the affected phase or metric while keeping the launch coherent. The interviewer is testing recalibration, not whether your first plan was perfect. Naming the tradeoff the new constraint forces, with a number, scores far higher than smoothing past it.

A strong answer names a specific beachhead merchant segment with a reason, sequences the launch as production testing then a design-partner merchant then a beta cohort then a percentage rollout, and attaches explicit release criteria to each gate. It defines a launch metric like payment success rate with an exact denominator and time window and pairs it with a guardrail. It states the number that triggers a rollback before scaling, names owners, and instruments a dashboard before launch. Under pushback it revises with reasoning rather than folding.

Enough to make the plan real, not encyclopaedic. You should be comfortable with payment success rate as the metric that drives merchant retention, MDR-thin acquiring economics, NPCI governance of UPI including market-share caps, settlement timelines, and that KYC and RBI payment-aggregator rules gate how fast onboarding can scale. You do not need exact regulatory clause numbers. You do need to weave these into the launch sequence and metrics rather than mentioning them once and moving on.

It is product sense applied to a launch with a strong business and operations layer. You are not designing a feature from scratch and you are not running a pure market-sizing case. You are deciding how to bring an already-scoped payments product to market: who first, in what phases, gated by what metrics, with what rollback. Expect business-acumen probes on unit economics and MDR alongside the launch sequencing, since a Razorpay PM owns both the rollout and the numbers it has to move.