PharmEasy PM Interview — Daily Medicine-Delivery Order Estimate

It tests whether you can size a messy real-world number out loud under interruption. The interviewer asks you to estimate how many medicine-delivery orders PharmEasy fulfils per day across India, then watches how you scope the question, build a visible calculation path, segment chronic versus acute and prescription versus over-the-counter demand, justify assumptions with India-specific reasoning, and sanity-check the final figure. The number itself matters far less than the structure and the reasoning. The senior PM running it grades transparent decomposition and composure when an assumption is pushed back on.

Start by clarifying the exact quantity: orders not users, India only, an average day not a peak. Then build one visible path, either from population down to buyers and orders, or from PharmEasy's known scale up to daily volume. Split demand into segments that behave differently, especially recurring chronic refills versus one-off acute purchases, and prescription versus over-the-counter. State each assumption before you use it and tie it to something real like population or online adoption. Finish with a sanity check against a known anchor and a stated confidence range.

The biggest one is announcing a final number before showing any structure. Close behind: treating all medicine orders as one pool instead of separating chronic refills from acute purchases, ignoring that prescription items need verification and convert worse than over-the-counter, assuming PharmEasy owns the entire online pharmacy market when it shares it with Tata 1mg and Apollo, and using India's full population without scaling down for smartphone reach and online-pharmacy adoption. Skipping the sanity check entirely is the last common way candidates lose this round.

It behaves like the senior PM candidates report at PharmEasy: it interrupts on shaky assumptions, probes one thing at a time, and never praises you mid-answer. The differences are that it is consistent, it never reacts to your accent or delivery, and it produces a transcript-backed scorecard afterward naming the specific assumption your structure could not defend. It will not coach you during the round or tell you whether your number is right, exactly as a real loop interviewer would not.

You are evaluated on observable behaviours, not the accuracy of the final number. The scorecard weighs how clearly you scoped the quantity, whether your calculation path was visible and followable, whether you segmented demand by behaviour like chronic refills versus acute orders, whether each assumption was justified with India-specific evidence, how you reworked the model when an assumption was challenged, and whether you closed with a sanity check and a confidence range. Each dimension is scored from the transcript so two evaluators would land within a narrow range.

Do not start computing. Spend the opening confirming what is being asked: orders rather than active users, India-wide, an average day rather than a festival peak, and whether tele-consultation or diagnostics count as a medicine order. Then say out loud which direction you are taking, building up from population or down from PharmEasy's known scale, and name the segments you intend to split before you touch any arithmetic. This signals structure first, which is exactly what the interviewer is listening for in the opening.

Expect it. Roughly a third of the way in, the interviewer will challenge one assumption deliberately to see whether you adapt or freeze. Do not defend it stubbornly and do not abandon the whole model. Acknowledge the challenge, swap in a revised number with a reason, and show how it propagates through the rest of your tree. Recovering cleanly under this pushback is one of the strongest signals you can give in this round, more than getting the original assumption right.

A strong answer narrates a visible path: here is my buyer base, here is why I scaled it down for smartphone and online-pharmacy reach, here is how chronic refill patients order monthly while acute buyers order rarely, here is the prescription-versus-over-the-counter split and why verification suppresses one of them, here is the share I am giving PharmEasy against Tata 1mg and Apollo, here is my daily number, and here is the benchmark I am checking it against with a confidence range. It stays narrated, segmented, and India-grounded throughout.

Either works, and the interviewer will not tell you which to pick because choosing and defending a path is part of what is graded. Top-down from India's population scaled by smartphone reach, online-pharmacy adoption and PharmEasy's market share is clean. Bottom-up from PharmEasy's reported customer base and an order-frequency assumption per chronic and acute user is equally valid and often easier to sanity-check. The strongest answers pick one explicitly, run it transparently, and then cross-check the result against the other direction.

Because the two segments generate completely different order volumes. A chronic-therapy patient on diabetes, hypertension or thyroid medication reorders roughly every month through automated refills, so a relatively small base of chronic users can produce a large, predictable share of daily orders. An acute buyer purchases once for a fever or infection and may not return for months. Treating both as one undifferentiated pool either massively over- or under-counts daily orders, which is why the interviewer pushes hard on whether you separated them.