Loading...
Please sign in to view your progress
Estimation interviews: market sizing, capacity, revenue and mental math. Train with Coach Tara, then take a mock with Owen.
| # | Name | ||
|---|---|---|---|
1 | The Method Document | ||
2 | ![]() | Learn the five steps behind every estimation question. Build a tree for the cafés in a made-up
city in Tara's 3D lab, tap to predict each move, watch the numbers flow up into the answer,
check it two other ways and say a range. Then answer two real loop questions out loud with
feedback.
| |
3 | ![]() | The estimation round of a PM loop with Owen. Estimate the restaurants in San Francisco: define
it, build a tree, defend each number, check it another way and say what it means for a product.
Then the same question in Bengaluru, and a quick one on Gmail users. Written report after.
| |
4 | Top-Down and Bottom-Up Document | ||
5 | ![]() | Build one number two ways: from the people who buy and from the places that sell. In Tara's 3D
lab, tap to predict, watch a tree's values flow up into the answer, and see two routes meet on
one summit. Then answer two real estimation questions out loud with feedback.
| |
6 | ![]() | The estimation round of a PM loop with Owen. How many shampoo bottles are sold in the US each
year? Build it from the people who use it, then from the stores that sell it, and defend one
number when the two disagree. Then a curveball and a quick second question. Written report
after.
| |
7 | Market Sizing Document | ||
8 | ![]() | Size a market honestly: count units, then dollars; narrow from TAM to SAM to SOM with reasons;
give a range for a market that barely exists. Tap to predict in Tara's 3D lab and watch the
market cake narrow. Then size two real markets out loud with feedback.
| |
9 | ![]() | A consulting-style sizing case with Owen. Size the US market for driverless ride-hailing in
2030: ask for the case facts, define it, build it, narrow it honestly and check it against the
fleet. Then a curveball, the so-what for the client, and a quick second question on GPUs.
Written report after.
| |
10 | Population Funnels Document | ||
11 | ![]() | Count people the way strong candidates do: pick the right unit, narrow with filters you can
defend, segment where it matters and never count anyone twice. Predict in Tara's 3D lab as a
funnel narrows and segments add up. Then answer two real questions out loud with feedback.
| |
12 | ![]() | The estimation round of a PM loop with Owen. How many people in India use a credit card? Build
a funnel of filters you can defend, keep people and cards apart, segment where it matters and
check it another way. Then a curveball country and a quick second question. Written report
after.
| |
13 | Capacity and Throughput Document | ||
14 | ![]() | Size anything that serves a flow: lanes, drivers, registers, elevators. In Tara's 3D lab, tap
to predict a checkpoint's capacity, its worst hour and its bottleneck, watch a queue pile up,
then count the planes in the sky with one line of math. Then answer two real loop questions
out loud with feedback.
| |
15 | ![]() | The estimation round of a PM loop with Owen. How many drivers does Uber need for the San
Francisco Bay Area? Size the busiest hour, turn rides into drivers, check it another way and
defend one number, then handle a Friday-night stadium crowd. A quick second question on
subway trains. Written report after.
| |
16 | Volume and Weight Document | ||
17 | ![]() | Crack the "how many fit" and "how much does it weigh" questions. In Tara's 3D lab, fill a box
and then a bus with golf balls, weigh the load two ways, find how dense a skyscraper is and
stack a million sheets of paper. Then answer two real loop questions out loud with feedback.
| |
18 | ![]() | A consulting first round opens with Owen's physical estimate: how many ping pong balls fit
in a Boeing 747? Ask for the figures you need, subtract what's inside, pack the rest and
defend one number, then handle tennis balls instead. A quick second question: what does a
school bus weigh? Written report after.
| |
19 | Revenue and Pricing Document | ||
20 | ![]() | How much money does it make? Build revenue as a tree in Tara's 3D lab, tap to predict each
total, separate what customers spend from what the company keeps, find the lever that moves
the answer most, and work out what a fee cut costs. Then two real revenue questions out loud,
with feedback.
| |
21 | ![]() | The estimation round of a PM loop with Owen. How much money does the Play Store make in a
year? Build it, say one number with a range, then work out what a five-point fee cut does and
whether you would make it. Then a quicker one: Airbnb's revenue. Written report after.
| |
22 | Rates and Growth Document | ||
23 | ![]() | Handle numbers that move. In Tara's 3D lab, tell a stock from a flow, turn a day into a
second, watch the evening peak push servers to the edge, compound five years of growth and
bend an adoption curve, then forecast a launch with no history. Then answer two real loop
questions out loud with feedback.
| |
24 | ![]() | The estimation round of a technical loop with Owen. How many YouTube videos are watched each
day? Then per second at the peak, then in five years. Define what counts, check it from watch
time, compound the growth honestly, then handle India only. A quick second question on text
messages. Written report after.
| |
25 | Technical Estimates Document | ||
26 | ![]() | Storage, requests a second, bandwidth and cost. Place files on a bytes ladder in Tara's 3D
lab, build the Google Photos storage tree and watch it multiply to exabytes, catch a classic
cost slip, and size links for the peak. Then two real technical questions out loud, with
feedback.
| |
27 | ![]() | A technical estimation round with Owen. As the PM for Gmail, how much storage will you buy for
next year? Ask for the constraints you need, size it with copies and headroom, defend your
buffer, then react when the free quota doubles. Then: messages per second at peak. Written
report after.
| |
28 | Mental Math Document | ||
29 | ![]() | Make interview math fast and safe. Predict where numbers land on Tara's 3D powers-of-ten track,
count zeros, round one up and one down, turn per day into per second, and never lose a unit.
Then answer two real loop questions out loud with feedback.
| |
30 | ![]() | A rapid round of eight quick calculations with Owen, then the quarters-to-the-Empire-State
question: scope it, break it down, keep every unit, check it in metres, and adapt when the
quarters stand on edge. Then a quick one on rice. Written report after.
| |
31 | Sanity Checks and Triangulation Document | ||
32 | ![]() | Catch a broken estimate before the interviewer does. In Tara's 3D lab, tap to predict how far
off a first guess is, find the node that breaks it, check it against a known number and watch
three routes meet. Then answer two real estimation questions out loud, with feedback.
| |
33 | ![]() | A PM estimation round with Owen. Estimate Zomato's delivery riders in Bengaluru, then check it
from the other side and hold up when a rainy evening changes the numbers. Then catch a
teammate's number that is ten times too big. Written report after.
| |
34 | Sensitivity and Ranges Document | ||
35 | ![]() | One number is fragile; a range with its driver is strong. In Tara's 3D lab, predict which
assumption moves an answer most, watch errors multiply, pick what to test first and see the
range tighten. Then answer two real forecasting questions out loud, with feedback.
| |
36 | ![]() | A PM estimation round with Owen. If AR glasses go mainstream by 2031, how many ship that year?
Give a low, base and high, find the assumption that moves it most, say what you'd test first,
then hold up when the price triples. Then: forecasting a product with no history. Report after.
| |
37 | Tailoring by Interview Document | ||
38 | ![]() | One method, four rooms. In Tara's 3D lab, take one product through a consulting case, a PM
loop, a technical round and a finance question: ask for data, give a recommendation, plant
the flag in a range, get the units right and find break-even. Then two real questions out
loud, with feedback.
| |
39 | ![]() | A consulting case with Owen: a grocery chain weighs its first store in Montreal. Ask for the
data you need, size the opportunity, work out break-even, and give the client a clear
recommendation, then react when the site changes. Then a quick one on checkout lanes. Written
report after.
| |
40 | Presenting Your Answer Document | ||
41 | ![]() | A good estimate can still be lost in the telling. In Tara's 3D lab, practise the answer-first
close: the number, the tree in one breath, the two assumptions that matter, the check and the
so-what. Then present a real estimate in sixty seconds and hold it when Tara pushes back.
| |
42 | ![]() | A PM estimation round with Owen. Size the in-flight ads market, then present it to a VP in
sixty seconds and defend it when he pushes back with an airline's number. Then a quick one,
answer first: how many planes are over the US right now? Written report after.
| |
43 | ![]() | The whole estimation round in one sitting. Warm up, size a bike-share launch in a city you
choose, hold your number under pushback, work out the data the bikes send at the morning peak,
and decide whether it pays at three dollars a ride. Hiring report after.
| |