Interview guides · Inc.

Inc. Data Scientist Interview Questions & Guide 2026

Causal inference drills for Inc. data scientist prep: diff-in-diff, matching, instruments, and the limits of observed data.

The 60-second brief

As a Data Scientist at Inc., you sit at the crucial intersection of advanced analytics, machine learning, and strategic decision-making. This role empowers you to drive high-impact initiatives that shape product development, optimize operational reliability, and influence business trajectories at scale. Whether you are building predictive models for fleet health, analyzing complex consumer and operational datasets, or deploying cutting-edge natural language processing pipelines, your work directly informs how Inc. innovates and delivers value to its users.

A large share of questions open as "how would you measure X", where the real work is choosing the metric, fixing its denominator, and defining the population it applies to. Any computation comes last and is frequently not required at all.

Focus areas

Size an experiment before anyone launches it
Turn a vague request into a measurable question
Define numerator, denominator and window precisely

Overview

The work environment at Inc. demands both technical excellence and a total ownership mentality. You will collaborate closely with engineering, product management, and business leadership teams to translate ambiguous problems into structured, data-driven solutions. Success in this position requires a balance of rigorous statistical methodology, efficient coding practices, and the ability to communicate complex findings compellingly to non-technical stakeholders.

Expect a fast-paced yet supportive environment where original thought is prized. You will be encouraged to challenge assumptions, explore novel methodologies, and take accountability for end-to-end model and analytics lifecycles. If you thrive on solving complex, multidisciplinary problems and want your analytical insights to drive tangible business outcomes, the role at offers an exceptional platform for your career.

The Inc. interview process

Inc. candidates report 4 rounds · ≈ 3-5 weeks. The stages below are what candidates describe, not a published process.

Recruiter Screening Call

A screening call is a matching exercise run by someone who will not evaluate your statistics. They are checking that the work described on your resume is work you personally did, and that its scope matches the level the role is written for. Logistics get settled in the same half hour so nobody spends an interviewer's afternoon on a mismatch. The answer that fails is the one narrated in the plural. If every sentence is 'we built' and 'the team decided', there is nothing specific to write down about you. Name the piece that was yours, the decision you made inside it, and what changed after.

What to demonstrate

Whether the ownership implied by your resume survives one round of follow-up about who actually did which part
Whether your described scope (data size, stakeholders, what shipped) matches the seniority the role is written at
Whether timeline, location and compensation expectations make the rest of the loop worth scheduling

How to prepare

Rewrite your top three resume bullets in the first person singular, each with the decision you made and what moved afterwards, then say them out loud once so the 'we' does not return under pressure
Attach one number to each project: the baseline, the change, and the window it was measured over. Where impact was never measured, say that plainly rather than inventing a figure
Settle your compensation range before the call and give it as a range with a reason behind it, such as current total comp or a competing timeline, instead of deflecting the question twice

Online Technical Assessment

A handful of shapes account for most of what gets asked in this format: a ranking or deduplication inside groups, a running or rolling total, a period-over-period comparison, and a cohort tracked forward over time. Recognising the shape quickly is most of the speed here; deriving it from scratch while a clock runs is where the time goes. Know that a window function keeps every row while a GROUP BY collapses them, and know which one the question needs. If the exercise is in Python instead of SQL, the same shapes arrive as groupby with transform, shift and merge, and the same grain mistakes are available.

What to demonstrate

Whether you reach the right construct without a detour, such as ROW_NUMBER over a partition to deduplicate instead of a self-join against a MAX subquery
Whether you know what your window frame actually is, since adding ORDER BY inside OVER changes the default frame and silently changes a running total
Whether the thing runs. A near-miss that throws an error scores below a plainer query that returns the right rows.

How to prepare

Write each of the four shapes once from memory against a small schema and keep the working version somewhere you will reread it: dedupe with ROW_NUMBER, a running total, a month-over-month change with LAG, and a retention table
Compute one running total twice on data with tied timestamps, once on the default frame and once with ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW, and look at where the two disagree
If Python is on the table, rebuild the dedupe and the running total with groupby and cumsum, then assert the two implementations return identical rows

Technical and Behavioral Rounds

Most of the weight in this round sits on the disagreement questions. Data work routinely produces an answer someone senior did not want, and the interviewer is trying to learn what you do in that hour. Both failure modes are common: folding as soon as a director pushes back, and treating the pushback as ignorance to be corrected with a better chart. A strong answer usually contains a specific thing the other person knew that you did not, and describes how you found out whether it changed the conclusion.

What to demonstrate

Whether you can state the other side's argument accurately before you explain why you disagreed
What you treated as evidence during the disagreement, such as a rerun under their assumption or a holdout check, rather than persuasion technique
Whether you distinguish being overruled from being wrong, and can give an example of each

How to prepare

Write out one disagreement where you turned out to be wrong, and say what in the data misled you. Candidates prepare the story where they were right, and the follow-up asks for the other one.
For your main disagreement story, be ready to say what result would have made you drop your position. If no such result exists, you were not arguing from the data.
Practise stating the opposing position out loud in one sentence the stakeholder would accept, then continue the story.

Final Interview Loop

A loop is not scored one interview at a time. The people you meet compare notes afterwards, usually in a meeting you are not in, and the outcome turns on what each of them can say about you when asked. That rewards something other than survival: every room needs one specific thing worth repeating, and none of them can contradict another. The common way to lose is to tell the same project four times with different numbers in it, or to be uniformly fine in a way that leaves nobody with anything to argue for.

What to demonstrate

Whether your account of a project survives being told twice, with the same scale, the same metric definition and the same numbers each time
Whether each interviewer leaves with one concrete claim they could make on your behalf later, rather than an absence of complaints
Whether a question you already answered in an earlier room gets the same answer at the same depth, without visible impatience

How to prepare

Write a one-page fact sheet for your two or three main projects that fixes the numbers you will quote: rows of data, the metric as a single sentence, the effect you measured and how long the work took. Say them aloud from the sheet until they come out identical every time
For each kind of room you expect, decide the one sentence you want that interviewer repeating in a debrief, then check during the mock that you said it outright instead of implying it
Rehearse answering the same project question twice in one sitting, the second time as though you had not just answered it, because the thing that needs fixing is the flatness that creeps into a repeated story

Where candidates lose points

Slicing a flat experiment until a segment reaches significance

Testing one metric across twenty segments at a nominal 5% level produces a significant result about two thirds of the time when nothing is happening anywhere, and the segment that surfaces is by construction the one with the most favourable noise. The reported effect in that slice is then badly overstated, because selection on significance conditions the estimate on being large. What makes it dangerous rather than merely wrong is that a post-hoc segment always has a plausible story attached, so it survives the meeting. The controls are declaring the small number of segments of interest before launch, correcting across the ones tested, and treating anything discovered afterwards as a hypothesis that needs its own adequately-powered test rather than a finding.

Watching an experiment daily and stopping when it crosses significance

A fixed-sample test controls type I error at one pre-declared look. Checking repeatedly and stopping at the first p < 0.05 inflates the false positive rate to roughly 0.15 to 0.20 for ten looks, and it rises further with more frequent checks, because the p-value takes a random walk that will eventually dip below the threshold under the null. The usual defences are a fixed horizon declared before launch, group-sequential boundaries such as O'Brien-Fleming that spend alpha across a planned number of looks, or always-valid confidence sequences that are correct under continuous monitoring. Compounding it, the effect size reported conditional on having crossed the threshold is biased away from zero, and the bias is larger the lower the power was, so an underpowered test that 'won' typically overstates the lift it found.

Naming a model class before naming the deployment constraints

Set out the latency budget, the label delay, the retraining cadence, the interpretability requirement and the number of labelled examples, then pick the model that fits them. A boosted-tree answer to a problem where each decision must be explained to the affected user is a well-executed answer to the wrong question.

Never asking what decision the analysis will inform

Open with who makes the decision, what the options are, and by when. The answer determines the precision you need, the segments worth cutting, and whether an observational read suffices or an experiment is required.

Questions & practice

How do you test for and handle non-normal distributions when comparing two sample populations?

Medium · statistics and probability

Approach

Sanity-check the answer against a simple bound or a simulated case.
Say what the estimate is of, and over what population it generalises.
Quantify uncertainty explicitly rather than reporting a point estimate alone.

Follow-up

How would you explain this result to someone who does not know statistics?
Which assumption here is most likely to be violated in practice?

How do you monitor for and mitigate data drift and concept drift in deployed machine learning models over time?

Medium · machine learning and modelling

Approach

Frame the prediction: the label, the moment of prediction, and the action it triggers.
Pick an evaluation metric that matches the cost of each error type, not a default.
Say how the offline result would be validated online before it is trusted.

Follow-up

What would you monitor after launch to know the model is still valid?
How would you choose the decision threshold, and who owns that choice?

Cluster bootstrap for a per-session rate randomised on users

Hard · cluster bootstrap · randomisation unit · variance

An experiment randomised on user_id reports a per-session conversion rate, so sessions inside a user are correlated. Input: one row per session with user_id, variant in {control, treatment} and converted in {0,1}. Write a cluster bootstrap from scratch: resample users with replacement within each arm, keep every session of a drawn user, recompute each arm's ratio of converted sessions to sessions, and take the difference. Return the point estimate, a 95 percent percentile interval from at least 2,000 resamples, the naive session-level interval that ignores clustering, and the ratio of their widths.

Approach

Name the estimand precisely: it is a ratio of sums, sum(converted) over sum(sessions) within an arm, not the mean of per-user rates. Those differ whenever session counts vary across users, and the ratio is what the reported metric is.
Resample the cluster, not the row. Draw n_users user ids with replacement inside each arm and take every session belonging to each draw, including duplicate draws of the same user. Keeping the user count fixed per arm rather than the session count is what preserves the sampling design.
Precompute per-user (converted_sum, session_count) once, so each resample is two vector lookups and a division rather than a repeated filter over the session frame. That turns 2,000 resamples from minutes into under a second.
Take the 2.5th and 97.5th percentiles of the 2,000 differences for the interval, and report the point estimate from the full data rather than from the bootstrap mean, since the bootstrap mean carries the resampling bias.
Compute the naive interval from the session-level binomial standard error and compare widths. The expected inflation is roughly sqrt(1 + (m-1)*rho), with m the mean sessions per user and rho the intraclass correlation of converted within users, so a computed ratio far from that value points at a bug in one of the two intervals.

Worked solution (35 min)

per_user = df.groupby(['variant','user_id'])['converted'].agg(['sum','size']); split into two arrays per arm.
point = (t_sum.sum() / t_n.sum()) - (c_sum.sum() / c_n.sum()).
For b in range(B): idx = rng.integers(0, len(t_sum), len(t_sum)); ratio_t = t_sum[idx].sum() / t_n[idx].sum(); same for control; store the difference. Vectorise by drawing a (B, n) index matrix if memory allows.
ci = np.percentile(diffs, [2.5, 97.5]); naive_se = sqrt(p_t*(1-p_t)/n_sessions_t + p_c*(1-p_c)/n_sessions_c); naive_ci = point +/- 1.96*naive_se.
width_ratio = (ci[1]-ci[0]) / (naive_ci[1]-naive_ci[0]).

Expected result: A point estimate identical to the full-data ratio difference, a percentile interval containing that point estimate, and a width ratio greater than 1 whose value approximates sqrt(1 + (m-1)*rho) for the data's mean cluster size m and intraclass correlation rho. A ratio of approximately 1.0 means sessions were resampled instead of users.

Verify it yourself

Construct data where every user has exactly one session: the cluster and naive intervals then agree to within Monte Carlo error, since m = 1 makes the design effect 1.
Construct data where converted is constant within each user: rho is 1, and the width ratio should approach sqrt(m).
The bootstrap distribution's mean sits close to the point estimate; a large offset means the resampled arms have drifted in size.
Doubling B narrows the Monte Carlo noise in the percentile endpoints without shifting the interval's centre.

Follow-up

Users average 3.4 sessions and the intraclass correlation is 0.12. What width ratio do you predict before running it, and does your bootstrap land there?
Give the delta-method standard error for this ratio and say when you would prefer it to the bootstrap.
Half the users in the treatment arm have exactly one session. What does that do to the cluster bootstrap's coverage, and how would you check it?

How do you optimize a slow-running query that joins multiple large tables in a distributed data warehouse?

Medium · sql and data manipulation

Approach

State the window function and its partition and ordering out loud before writing it.
Check whether any join is one-to-many before aggregating, or the sums inflate.
Handle the rows that do not match: a LEFT JOIN with a NULL check is usually the question.

Follow-up

How would you verify this result without re-running the same query?
How does the query change if the join becomes one-to-many?

Given a complex event log, write a query to identify user session streaks and drop-off points.

Medium · sql and data manipulation

Approach

Check whether any join is one-to-many before aggregating, or the sums inflate.
State the window function and its partition and ordering out loud before writing it.
Say which table is the grain you start from, and join outward from it.

Follow-up

How would you verify this result without re-running the same query?
What breaks if events arrive late or out of order?

Seven-day activation rate by weekly signup cohort

Easy · cohorts · activation · date-bounds

dim_user holds user_id, account_created_at_utc, is_internal. fct_event holds user_id, occurred_at_utc, is_core_action. A user is activated when core-action events fall on at least two distinct UTC dates inside [account_created_at_utc, account_created_at_utc + 7 days). Return, for the last twelve complete weekly signup cohorts, the cohort week, cohort size, activated users and the activation rate. Exclude is_internal users. Every signup in the cohort week stays in the denominator, including users who never returned.

Approach

Start from dim_user as the denominator spine with is_internal = FALSE and DATE_TRUNC('week', account_created_at_utc) as the cohort key. Driving the query from the event table instead would silently condition on having events and delete the entire non-activating population.
Join fct_event on user_id with is_core_action = TRUE and a per-user bound, occurred_at_utc >= u.account_created_at_utc AND occurred_at_utc < u.account_created_at_utc + interval '7 days'. The bound is correlated to each user's own signup timestamp, not a single global date range.
Aggregate per user with COUNT(DISTINCT occurred_at_utc::date) >= 2, then LEFT JOIN that back onto the spine and COALESCE the flag to FALSE so non-activators contribute a zero rather than vanishing.
Restrict the published cohorts to those whose week ended at least eight days ago. A cohort younger than that has not finished its seven-day window, so its rate is mechanically low and reads as a decline.
Roll up by summing the numerator and denominator per cohort week, and state the two-distinct-days threshold next to the number since it is a choice that re-bases the whole history if changed.

Worked solution (20 min)

Write the cohort spine and confirm its total equals the count of non-internal signups in the date range.
Write the per-user distinct-active-days CTE with both interval bounds and inspect a handful of users manually.
LEFT JOIN, COALESCE the flag, aggregate to cohort week.
Apply the eight-day publication lag and drop the incomplete cohort.
Re-run with a closed upper bound (<= +7 days) and note how many users change state, to show the boundary is doing work.

Expected result: Twelve rows, one per cohort week, each with cohort_size, activated and rate. Cohort sizes sum to the total non-internal signup count across those twelve weeks. No cohort is younger than eight days.

Verify it yourself

activated is never greater than cohort_size in any row.
A user with three core actions on a single day is not counted as activated.
Removing the LEFT JOIN in favour of an inner join raises every rate, which demonstrates the denominator was being conditioned on survival.
Cohort sizes sum to an independent COUNT(*) over dim_user for the same range.

Follow-up

Why two distinct days rather than one event? What happens to the published history if someone changes it to three?
Invited seats and SSO-provisioned users get an account_created_at_utc at provisioning and may never sign in. Should they be in this denominator?
The rate rose 3 points this week. What do you check before believing it?

How do you define and track key performance indicators for operational reliability and downtime reduction?

Medium · product sense and metrics

Approach

Restate the decision this analysis has to support, and who acts on the answer.
State what result would change your recommendation, so the answer is falsifiable.
Fix the population and the time window before naming any metric.

Follow-up

What would you do if the primary metric and the guardrail moved in opposite directions?
How would you detect that the metric is being gamed rather than genuinely improving?

What steps do you take when test variants show a statistically significant lift in one secondary metric but a drop in the primary guardrail metric?

Medium · product sense and metrics

Approach

Name one primary metric, then the guardrail that stops it being gamed.
State what result would change your recommendation, so the answer is falsifiable.
Restate the decision this analysis has to support, and who acts on the answer.

Follow-up

How would you detect that the metric is being gamed rather than genuinely improving?
What would you do if the primary metric and the guardrail moved in opposite directions?

How do you evaluate the success of a feature when standard conversion metrics do not tell the full story?

Medium · product sense and metrics

Approach

State what result would change your recommendation, so the answer is falsifiable.
Restate the decision this analysis has to support, and who acts on the answer.
Decompose the metric into the rates that drive it, and say which one you would check first.

Follow-up

Which segment would you cut first, and what would that rule out?
What would you do if the primary metric and the guardrail moved in opposite directions?

A key business metric has experienced a sudden drop over the past week; walk us through your diagnostic framework to find the root cause.

Medium · product sense and metrics

Approach

Restate the decision this analysis has to support, and who acts on the answer.
Name one primary metric, then the guardrail that stops it being gamed.
Decompose the metric into the rates that drive it, and say which one you would check first.

Follow-up

How would you detect that the metric is being gamed rather than genuinely improving?
What would you do if the primary metric and the guardrail moved in opposite directions?

How do you account for network effects or interference between treatment and control groups in a marketplace setting?

Medium · a/b testing and experimentation

Approach

Decide the analysis before seeing data, including how long it runs and when you look.
Name the randomisation unit first; it decides the variance and what the test can detect.
Name the guardrails that would stop a launch even on a positive primary result.

Follow-up

What would you do if you could not randomise at all?
What would you conclude if the result is positive but the test is underpowered?

Walk us through the end-to-end design of an A/B test for a new user onboarding flow, including sample size estimation and power calculation.

Medium · a/b testing and experimentation

Approach

State the primary metric and the minimum effect worth shipping, then size the test.
Decide the analysis before seeing data, including how long it runs and when you look.
Say whether units interfere with each other, and switch design if they do.

Follow-up

What would you conclude if the result is positive but the test is underpowered?
What would you do if you could not randomise at all?

A completion rate the owning team can move without fixing anything

Medium · gaming · instrumentation · funnel

A team's target is core-flow completion rate: distinct fct_event.flow_instance_id with a 'flow_completed' event within 30 minutes of its 'flow_started' and no 'error_shown' carrying the same flow_instance_id in between, over distinct flow_instance_id with a 'flow_started' in the window, split by surface and app_version. The same team owns the client that emits those events and the tracking plan that defines them. List the ways this rate rises without any user completing more flows, then redefine the metric and its guardrails so those routes are closed. Deliverable: the hardened definition.

Approach

Work the emission side first, because that is what the team controls: delay minting flow_started until after the first screen so the highest-dropping attempts leave the denominator; stop emitting or rename error_shown; mint a fresh flow_instance_id on each retry so one failed attempt becomes several attempts whose last one completes; move flow_completed earlier in the flow.
Sort those moves by where they are visible. None of them shows in the rate itself; three of them show only in volume, which is why the denominator has to be published on the same chart as the rate.
Re-anchor the numerator on something outside the flow's own instrumentation: require a downstream is_core_action = TRUE event for the same user_id within 30 minutes of flow_completed, so a completion only counts when it produced the thing the flow exists to produce.
Add the guardrail that catches the retry route specifically: mean and p90 flow_instance_id per user per day, with the rule written down that a rising completion rate alongside rising attempts per user is a regression and not a win.
Make definition changes visible instead of forbidden: stamp a tracking-plan version on the series and re-base the history when event semantics or the 30-minute window change, rather than splicing two definitions into one line.

Worked solution (25 min)

Enumerate at least four emission-side moves and mark for each whether it shows in the rate or only in volume.
Pin the denominator: publish flow_started per 1,000 sessions and per distinct user beside the rate.
Rewrite the numerator to require a downstream core action for the same user within 30 minutes.
Add the attempts-per-user guardrail with its interpretation rule stated as a sentence.
Version the definition and write the re-base policy for the historical series.

Expected result: A hardened definition in which each enumerated gaming route is either blocked by the numerator, exposed by a published volume series, or forced to surface as a definition version change.

Verify it yourself

Replay one untouched historical week under the old and hardened definitions. A large gap means the hardening changed the meaning of the metric, not only the incentives around it.
flow_started per 1,000 sessions should be roughly continuous across an app_version boundary unless the flow's entry point genuinely moved; a step change there is an instrumentation change.
Attempts per user and completion rate should not rise together within one build.

Follow-up

How do you distinguish a genuine instrumentation fix from gaming, given both look like a step change confined to one release?
The jump appears in exactly one app_version. Does that exonerate the team or implicate it?
What do you do with eighteen months of history once the definition is hardened?

Attribute a flow-completion drop to one client build

Medium · release-regression · instrumentation · staged-rollout

Core-flow completion rate on surface = 'ios', defined as distinct flow_instance_id reaching 'flow_completed' within 30 minutes of 'flow_started' with no 'error_shown' on the same id in between, fell from 78% to 71% over four days. A new iOS build began a staged rollout on day one. Using fct_event columns flow_id, flow_instance_id, app_version, event_name, user_id and occurred_at_utc, produce the completion rate per build per day, size the loss in absolute completions, and say whether the fix is a rollback. Treat the possibility that the instrumentation changed rather than the flow.

Approach

Cut the rate by app_version and day, each build against its own denominator. A pooled series during a staged rollout is a weighted average whose weights move daily, so it declines in proportion to the rollout share even when the old build is perfectly flat; confirming that proportionality is itself the evidence the regression is build-specific rather than environmental.
Test the denominator before believing the rate. Compute flow_started events per distinct user_id on each build: if the new build re-mints flow_instance_id on retry, or fails to carry the id from start to completion, the denominator inflates and the numerator falls with nothing changing for the user.
Normalise to a user-level outcome that is immune to the id question, namely completions per distinct user_id per day on each build. This is the number that says whether anyone actually failed to finish.
Check the error path: pull error_shown counts and their properties for flow_instance_id values on the new build. A genuine regression produces errors; an instrumentation break produces missing completions with no corresponding error volume, and the two prescribe different fixes.
Size the real loss as the per-user completion gap multiplied by daily users on the new build, and give the rollback recommendation conditional on the error evidence rather than on the rate.

Follow-up

The rollout is at 45% and product wants to go to 100% tomorrow. What do you say, and what would you need by when?
If the flow_instance_id is genuinely being re-minted, what is the correct historical treatment of the four affected days?
How would you have caught this on day one instead of day four?

Suggested practice checklist

For a candidate whose interviews will centre on A/B testing, metric movement and causal claims. Design comes before arithmetic, arithmetic before analysis, and the week ends by rehearsing the readout rather than the derivation.

Day 1 — Design one test end to end on paper

Take a single feature change and write the full design: randomization unit, the exact point of exposure, the primary metric with its grain, guardrails, allocation, planned duration, and the decision rule committed before any data exists.
Write why the randomization unit must sit at or above the level where treatment can spill over, and give one case where user-level randomization is still contaminated (shared accounts or devices, or two participants in the same marketplace).
State in advance what you will do if the primary metric is flat while a secondary metric is significant.

Deliverable: A one-page test design with a decision rule written before launch.

Day 2 — Power arithmetic until it is automatic

Compute required sample size per arm for a binary metric with the normal approximation, n is approximately 2 times (z for alpha/2 plus z for power) squared times p(1 minus p) divided by delta squared, for baselines of 2, 10 and 40 percent at a 5 percent relative lift, and note that for a fixed relative lift the requirement falls as the baseline rises because delta grows proportionally with p.
Redo the calculation for a continuous metric using variance in place of p(1 minus p), and show why a heavy-tailed quantity such as revenue per user needs either far more traffic or a capped version with a stated cap.
Convert one of the results into weeks given a weekly eligible traffic figure, then list the two honest ways to shorten it (accept a larger detectable effect, or reduce variance) and write why quietly lowering the power target is a decision to miss more real wins, not a speedup.

Deliverable: A small script or sheet that maps baseline, minimum detectable effect, alpha and power to sample size and weeks, cross-checked against a published calculator.

Day 3 — Variance and the unit-of-analysis problem

Take a ratio metric whose denominator is not the randomization unit (clicks per session, randomized by user) and compute the standard error twice, once naively at session level and once by the delta method or a user-level bootstrap, then record how much the naive version understates it.
Implement CUPED on simulated data: choose a pre-period covariate X measured before assignment, estimate theta as Cov(Y, X) divided by Var(X), and analyse Y minus theta times (X minus its mean) in place of Y. Confirm the variance of the adjusted outcome equals the raw variance multiplied by one minus the squared correlation between Y and X, so a correlation of 0.45 removes about 20 percent of the variance and not 80.
Now run that simulation a few hundred times and confirm the adjusted effect estimate is unbiased for the same effect rather than numerically identical to the raw one. Within any single run the two differ, sometimes by a large fraction of the true effect, because the two arms' pre-period covariate means never coincide exactly in a finite sample; they agree in expectation, which is the property that matters and the one to state out loud.

Deliverable: A notebook showing the adjusted estimator with a measurably smaller variance than the raw one, plus a repeated-simulation table showing the two estimators agreeing on average while differing run by run.

Day 4 — Validity threats you can actually test for

Run a sample ratio mismatch check as a chi-square goodness-of-fit test against the intended allocation, and write the three causes you would chase first (assignment logged before exposure, an arm-specific redirect or load failure, bot filtering applied asymmetrically).
Simulate peeking: generate A/A data, test daily at alpha 0.05 across 14 looks, record the inflated false positive rate, then apply an alpha-spending boundary or commit to a fixed horizon and confirm the rate returns to nominal.
Write how you would separate a novelty effect from a durable lift using the treatment effect plotted against days since first exposure, and what shape would change your recommendation.

Deliverable: One table showing the peeking false positive rate before and after correction, plus a written SRM triage list.

Day 5 — When randomization is not available

Write the identifying assumption for difference-in-differences (parallel trends in the absence of treatment), then plot pre-period trends for two candidate control groups and justify rejecting one of them.
Design a switchback test for a change where user-level randomization would leak across participants, choosing a time-block length against the carryover you expect and saying how you would detect carryover in the data.
List what an interrupted time series or a synthetic control buys you and the one thing neither can rule out: an unobserved shock that coincides with the launch.

Deliverable: A one-page memo recommending a single quasi-experimental design and naming its weakest assumption explicitly.

Day 6 — The readout query

Write the assignment-to-exposure join that returns exactly one row per unit per experiment, and handle units appearing in both arms by excluding and counting them rather than silently keeping one.
Compute the per-arm metric, its variance and the relative lift with a confidence interval in SQL, then reproduce the identical numbers in a notebook as a cross-check.
Add a segment breakdown and write the sentence that keeps it from being p-hacking: segments declared in advance, everything else reported as exploratory and corrected for multiplicity.

Deliverable: A single query that outputs the full readout table, matched to a notebook recomputation.

Day 7 — Present it to someone who will not read the appendix

Give a 10-minute readout of a real or simulated experiment in the order decision, number, uncertainty, caveat.
Have your listener ask "can we ship it" in the case where the primary is flat and a guardrail moved, and answer with a recommendation rather than a request for more data.
Rewrite your opening line so the recommendation lands before any methodology.

Deliverable: A one-page readout whose first line is the recommendation.

Where your data came from

Data people depend on systems owned by other teams, and much of the job is negotiating for instrumentation, access, or a fix to a broken pipeline. Prepare an example of getting something changed upstream that you did not control. Describe what you asked for, what you traded, and how you worked while you waited.

Give an example of how you took ownership of an ambiguous problem space with minimal initial direction.

Medium · behavioural and stakeholder questions

Approach

Name the disagreement or constraint, and how you resolved it with evidence.
Pick a story where you drove the decision, not one where you observed it.
Quantify the outcome, including what you would not claim credit for.

Follow-up

How did you know the outcome was caused by your change?
What did you decide not to do, and why?

Defend a flat experiment readout against a post-hoc segment

Medium · experiments · segmentation · pushback

A feature you evaluated is flat on seven-day activation: +0.05pp with a 95% interval of [-0.47pp, +0.57pp], from 61,000 exposed users per arm in fct_experiment_exposure joined to dim_user and fct_event. Baseline activation is 32%. The launch team asks you to drop every surface except mobile_web, where the point estimate is +1.1pp, and re-run. You have ten minutes in their planning meeting. Deliver a spoken position: what you will and will not do, and the decision you recommend.

Approach

Recognise what is being probed: whether you hold a statistical position under social pressure without becoming either rigid or apologetic. A generic answer says the segment is not significant; a strong one separates the request into a question that is answerable (is the mobile_web number real?) and one that is not (can we ship on it?), and answers both.
Price the multiplicity out loud. The slice was chosen after seeing the results, so its estimate is selected on favourable noise and is biased away from zero. With k independent looks at a nominal 5% level, the chance of at least one false positive is 1 - 0.95^k: 26% at six segments, 64% at twenty. Quote the k you actually inspected, not the k you reported.
Use the arithmetic already in front of you. On the point estimates, a +1.1pp mobile_web effect combined with a pooled +0.05pp implies the remaining surfaces average negative in proportion to mobile_web's share of exposures. State that as a testable implication of their story rather than as a rebuttal of it.
Ask the one question that settles the category: was mobile_web named in the analysis plan before launch? If it was, it is a planned comparison and gets a corrected reading. If it was not, it is a hypothesis, and the honest move is to size the test that would confirm it.
Convert the refusal into a cost. Size a mobile_web-only confirmatory test at the claimed effect, state the weeks of mobile_web traffic it needs, and close with the recommendation: do not ship this as a lift, and note that the interval already rules out anything at or above +0.6pp, which is itself a useful input to the roadmap.

Follow-up

The confirmatory test you sized needs nine weeks of mobile_web traffic and the team has three. What do you recommend instead?
Suppose mobile_web was pre-registered. How does your reading change, and what correction do you apply?
Your interval excludes +0.6pp. Is that the same as saying the feature does nothing?

Turn an ambiguous onboarding question into a measurable metric

Easy · scoping · metric definition · stakeholder

Two days before a planning review, a director asks whether onboarding is working. You have dim_user (account_created_at_utc, signup_surface, is_internal), fct_event (is_core_action, flow_id, flow_instance_id, event_name, occurred_at_utc, received_at_utc) and fct_session. No further meeting with the director is possible before you start work. Deliver three clarifying questions you would send in writing, the metric you will compute in the meantime with its numerator, denominator, window and exclusions, and one sentence naming the question you are deliberately not answering.

Approach

Recognise what is being probed: whether you convert a goal into a computable predicate without stalling for requirements or guessing in silence. Listing clarifying questions is the generic answer; shipping a defensible default alongside them is the strong one, because the review is in two days and it will happen with or without you.
Infer the decision behind the request. A question about whether onboarding works, arriving before a planning cycle, usually means whether to staff it next quarter. That points at a rate with visible headroom over several cohorts, not at a descriptive dashboard.
Write the three questions so that each one changes the SQL. Which population, all signups or only self-serve from dim_user.signup_surface. What counts as working, reaching a core action or completing the onboarding flow_id. Against what bar, last quarter's cohorts or a stated target.
Propose the default explicitly: seven-day activation on weekly signup cohorts. Numerator, users with is_core_action = TRUE events on at least two distinct UTC dates inside [account_created_at_utc, account_created_at_utc + 7 days). Denominator, the signup cohort with is_internal = FALSE. Publish with an eight-day lag, and state that the two-distinct-days threshold is a frozen choice rather than a discovery.
Name the exclusion in the same breath as the number. The series shows whether users activate; it does not establish that onboarding caused the level, which needs a staged rollout or an experiment.

Follow-up

The director replies that they meant the onboarding flow specifically, not activation. What changes in the query and in the caveats?
Your cohort metric needs an eight-day lag and the review is in two days. What do you present, and how do you label it?
Two of your three questions come back unanswered. Which one do you refuse to proceed without?