Level A Quick Drills: 20 Timed Guesstimate Self-Attempts
Level A Quick Drills are 15-minute, medium-difficulty guesstimate prompts designed for timed self-attempts before you look at any solved structure. They help you practise the interview habits that matter most - clarifying scope, building an assumption tree, doing clean math, and sanity-checking the result. This sheet gives you the 20 unsolved prompts, a reusable attempt process, and a self-grading method so you learn from structure before judging the final number.
- Level A drills are medium guesstimates meant to be attempted in 15 minutes with no peeking.
- Each attempt should include scope, assumption tree, math, and a sanity check.
- The two-line nudge is only a thinking starter, not the answer.
- When reviewing, check scope, tree granularity, and sanity check before checking the final math.
- Level B skeletons show reference structures and defensible ranges, not single right answers.
- Level C master drills chain 2-3 sub-estimates and usually deserve 20 minutes, not 15.
How to Use This Drill Sheet
These 20 prompts are the Level A section of a graded guesstimate drill book. Every prompt is marked medium and should be tried as a 15-minute timed attempt. The goal is not to guess a perfect number; it is to make your thinking visible enough that an interviewer can follow and challenge it.
Use the hint only as a starting direction. If the hint says "population x users x messages/user" or "vehicle count x utilisation x distance", it is pointing you toward the type of tree, not giving you the answer.
Clarify scope - build the tree - state assumptions - do the math - sanity-check - then compare with a skeleton or external reference.
Before drilling individual prompts, keep the full guesstimate workflow in view.
The 20 Level A Quick Drills
Use this table as your timed practice bank. All 20 prompts are medium difficulty. Several are India-focused and use named everyday contexts such as Zomato, Swiggy, WhatsApp, Fevicol, and Delhi Metro, which makes them useful for placement interviews where the interviewer expects grounded business intuition.
What These Prompts Are Testing
Although the topics look different, most Level A drills test the same core muscles. A food delivery rider estimate tests volume and frequency. A kirana revenue estimate tests Total Addressable Market, or TAM, which means the full annual revenue pool available in a market. A WhatsApp estimate tests user segmentation and intensity.
Some acronyms in the drill bank are worth fixing early. FY26 means financial year 2026. FMCG means Fast-Moving Consumer Goods, such as everyday packaged categories; the source uses it while sizing kirana and biscuit markets. HH means households. WTP means willingness to pay, used when estimating what people are ready to spend, such as after-school tuition.
Attempt Process for a 15-Minute Drill
A 15-minute guesstimate should feel structured, not rushed. The easiest way to lose time is to start multiplying before deciding what exactly you are estimating. For example, "daily ridership of Delhi Metro on a typical weekday" requires you to lock the day type and unit before you think about lines, stations, peak, and off-peak.
Do not check the final number first. Check scope, tree granularity, and sanity check; only then look at the math.
Worked Example: WhatsApp Messages Sent in India Per Day
Situation: estimate the number of WhatsApp messages sent in India per day. The problem is not just population sizing; a single average can hide major behavioural differences, so the skeleton uses users x messages per user per day and segments users by intensity. The decision is to split India WhatsApp MAU into heavy, medium, and light users, then sum the daily messages and sanity-check against global WhatsApp volume.
The learning is that segmentation often beats a single average. If you used 55 cr users x one assumed message count, the answer might still be in range, but the interviewer would see less evidence that you understood usage intensity.
How to Compare With Solved Skeletons
Level B skeletons are not answer keys in the school-exam sense. They show the tree and assumption set, with the final number stated as a range. That is why the review order matters: structure first, arithmetic second.
For example, the kirana skeleton uses a top-down approach: total grocery + FMCG retail x kirana share. It uses Indian FMCG market of ~₹19 lakh crore (FY26), adds fresh produce + unorganised staples of ~₹16 lakh crore, reaches total grocery + FMCG TAM of ~₹35 lakh crore, applies a kirana share of ~75-80%, and gets kirana TAM of ~₹27 lakh crore with a range of ~₹24 - 30 lakh crore annually. The sanity check is ~1.3 cr kirana stores x ~₹20 L/yr each ≈ ₹26 L cr.
Choosing the Right Estimation Tree
Most candidates struggle less with multiplication and more with choosing the right first breakdown. A bottom-up tree starts from units on the ground, such as vehicle stock for petrol sold daily in Mumbai. A top-down tree starts from a larger market pool, such as grocery + FMCG TAM for kirana revenue.
When ownership may overlap, use the tree as a practical decision, not a rigid rule. For instance, annual helmet sales combines stock, ownership, replacement cycle, new two-wheeler sales, and pillion adoption. That is not purely top-down or bottom-up; it is a stock-and-flow estimate.
Nuance: Level A Is About Habits, Not Perfect Answers
The source is explicit that there is no single "right answer" in the solved skeletons. A defensible range is often better than a falsely precise number. In many interviews, a candidate who gives a transparent tree, clear assumptions, and a strong sanity check will score better than someone who lands near the answer through unclear arithmetic.
This is especially true when the same prompt can be approached in more than one acceptable way. Zomato/Swiggy delivery riders in Bangalore could be estimated from order volume and rider capacity, while Delhi Metro ridership could be built from lines, stations, and peak/off-peak usage. The interviewer is watching whether your structure matches the business reality of the question.
The most frequent error is grading yourself on the final number first. That costs points because guesstimates are judged on scope, tree granularity, assumptions, and sanity checks before arithmetic precision.
Conclusion
Level A Quick Drills build the core interview reflex: structure the unknown before calculating it. Attempt each prompt in 15 minutes, review the tree before the number, and use every skeleton comparison to improve how you scope, segment, estimate, and sanity-check.