Interview guides · Vertafore

Vertafore Data Scientist Interview Questions & Guide 2026

Data scientist prep for Vertafore: practice defining a B2B software and infrastructure metric, picking its guardrails and defending the tradeoff.

The 60-second brief

As a Data Scientist at Vertafore, you play a pivotal role in harnessing data to drive better decision-making and enhance the overall user experience. This role is crucial for leveraging large datasets to derive insights that inform product development and strategy, ensuring that Vertafore remains at the forefront of the insurance technology industry. You will be involved in building predictive models, conducting exploratory data analysis, and collaborating across teams to translate complex data into actionable strategies.

How much statistics you need depends on the flavour of the seat. Experiment-facing work wants you deep enough to notice that repeated looks at accumulating data inflate the false positive rate of a fixed-sample test; modelling-facing work wants estimation and honest uncertainty intervals.

Focus areas

Separate contracted seats from actively used seats
Strip CI, retry and synthetic traffic first
Analyse at the account grain, cluster errors

Overview

The impact of a Data Scientist at Vertafore extends beyond mere analysis; you will contribute to creating innovative solutions that affect products like AMS360 and Vertafore's Agency Platform. These tools are integral to the insurance sector, and your work will help streamline workflows, improve user engagement, and ultimately drive business growth. Expect to address complex challenges involving data quality, model performance, and user behavior, making this role both demanding and rewarding.

The Vertafore interview process

Vertafore candidates report 2 rounds · ≈ 2-4 weeks. The stages below are what candidates describe, not a published process.

Cognitive Test

An extra round usually exists because something is still open after the standard loop: a skill the earlier interviews did not sample, a level decision, or two interviewers who disagreed. It is rarely a rerun of what you already did well. Ask the recruiter who you are meeting, what function they sit in, and how long the session runs. That is an ordinary scheduling question, and the answer changes what you should prepare. What separates a strong candidate here is treating the round as a fresh evaluation with its own bar, rather than assuming earlier performance carries you through or sinks you.

What to demonstrate

Whether you can answer well on ground the earlier rounds did not cover, without leaning on what you already said to someone else
Consistency of the facts in your stories: the same sample size, timeframe, team size and scope of your own role as in earlier conversations
How you handle an unfamiliar format live, including whether you ask what kind of answer is wanted before producing one

How to prepare

Ask the recruiter for the interviewer's function, the length, and whether to expect a coding surface, a discussion, or a presentation. Preparing for a 30 minute conversation with a partner team is not the same work as preparing for a 60 minute technical block.
Write out what each earlier round actually covered, then list the two or three areas nobody probed. That gap is the most likely subject of the extra round.
Re-read the numbers in the project stories you have already told, so a second telling does not quietly contradict the first.

Video Interview

Because the format is not fixed, prepare the reasoning rather than the ritual. Nearly every version of this round draws on the same underlying material: a design you can defend, a metric you can define exactly, an analysis whose assumptions you can state out loud. Only the wrapper changes, whether that is a take-home, a live case, a deep dive on past work, or a rough estimate on a whiteboard. Answers rehearsed to fit one shape stall the moment the shape differs. Practise naming the assumption behind a number, then saying how much the conclusion moves if that assumption is wrong.

What to demonstrate

Whether your justification for a method survives the question 'why not the simpler thing', including when the simpler thing would have worked
Precision under pressure: what exactly counts as an active user, a conversion or a success, over what window, with what exclusions
Whether you carry an argument through to a recommendation instead of stopping at a list of tradeoffs

How to prepare

For each project you plan to mention, write the metric definition in one sentence: numerator, denominator, time window, exclusions. Say it out loud once, because vagueness shows up in speech before it shows up on paper.
Rehearse the same project at three lengths: two minutes, ten minutes, and a deep dive on one technical decision. Cutting live is harder than it sounds.
For your headline result, write down what would have had to be true for it to be wrong, and how you ruled that out.

Where candidates lose points

Randomising an experiment at the user level when users share an account

Two problems fire at once. Colleagues in one workspace see each other's work and talk to each other, so a treated user changes the behaviour of a control user in the same account, which violates the no-interference assumption and biases the estimate toward zero. Separately, outcomes within an account are strongly correlated, so the effective sample size is roughly n / (1 + (m - 1) * rho) for m users per account and intra-class correlation rho, not n. With rho around 0.3 and twenty users per account that is a design effect near 6.7, meaning a user-level confidence interval is about two and a half times narrower than it should be and results cross significance thresholds on noise alone. Randomise the account and cluster the standard errors.

Comparing accounts that received a sales or customer-success touch against those that did not

Assignment of coverage is deliberate and pulls in both directions at once: the largest accounts get a named owner because they are valuable, and the accounts showing distress get one because they are at risk. The comparison therefore mixes a strong positive selection with a strong negative one, and the naive estimate can come out with either sign depending on which assignment rule dominated during the period examined. Nothing about matching on observed size fixes this, because the risk signal that triggered coverage is usually the same signal that predicts the outcome. It needs either an actual randomised or staggered rollout of coverage, or a design built on a capacity constraint or territory boundary that assigns coverage for reasons unrelated to account health.

Extrapolating a first-week lift inflated by novelty effects

Plot the treatment effect by days since first exposure instead of quoting one pooled average. A lift that decays toward zero across the test window is behaviour that will not persist, and annualising it produces a forecast that misses by an order of magnitude.

Dropping rows with missing values without naming the mechanism

Say whether the values are missing at random, missing by a known process, or missing in a way that depends on the outcome, and handle them accordingly. Deleting incomplete rows silently redefines the population whenever missingness correlates with what you are measuring.

Questions & practice

What metrics would you use to evaluate the performance of a classification model?

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

Where could label leakage enter this setup?
How would you choose the decision threshold, and who owns that choice?

Discuss the importance of feature engineering in predictive modeling.

Medium · machine learning and modelling

Approach

Pick an evaluation metric that matches the cost of each error type, not a default.
Set a baseline first, so any model has something honest to beat.
Check what information would not exist at prediction time, and exclude it.

Follow-up

How would you choose the decision threshold, and who owns that choice?
Where could label leakage enter this setup?

Permutation-test a consumption experiment randomised at account level

Hard · permutation-test · heavy-tails · cuped · experiment-readout

An experiment randomised 900 accounts into two arms. You have one row per account: account_id, arm, consumption_28d (billable units after launch) and consumption_pre (the 28 days before). Consumption is heavy-tailed and the largest account is several percent of the total. Write a permutation test from scratch: winsorise at the pooled 99th percentile as a pre-registered rule, use the difference in arm means of the winsorised outcome as the statistic, and obtain a two-sided p-value from 20,000 relabellings of the account-level arm vector. Report the observed effect, the p-value, and the same test on a CUPED-adjusted outcome.

Approach

Be precise about what the permutation test needs. Under the sharp null of no effect for any account, the outcomes are exchangeable across arm labels, and the test is valid for ANY statistic T(outcomes, labels) provided the identical function is applied to the observed labels and to all 20,000 relabellings. The pooled 99th percentile is a function of the outcome vector alone, so recomputing it inside the loop returns the same number 20,000 times: that is wasted CPU, not a bias, and hoisting it out is an optimisation rather than a correctness fix. Say plainly that capping at all changes the estimand from mean consumption to mean capped consumption; it is not a neutral cleaning step.
The mistake that does invalidate the test is an asymmetry between the observed statistic and the permuted ones, and the easiest way to create it is to derive the cleaning rule from the observed arm labels and then freeze it — winsorise each arm at its own observed 99th percentile, hold those two caps fixed, and permute. The observed value is then computed with caps matched to its own partition while every relabelling is scored with caps belonging to a different one, so the null distribution no longer answers the question the p-value claims to answer. A per-arm cap recomputed consistently inside every permutation is a valid test, but it estimates a contrast whose two sides are capped at different thresholds, so prefer the pooled cap on estimand grounds and pre-register it.
Permute the account-level arm vector, because the account is the randomisation unit. Relabelling anything finer — users, workspaces, requests — generates a null distribution narrower than the design actually supports and returns p-values that are anti-conservative.
Vectorise the null: tile the treatment indicator into a (B, n) matrix and permute along axis 1 with rng.permuted(..., out=...). The statistic is a difference of means, so the treated sum alone determines it and the whole null is one matrix-vector product. Use the two-sided p-value (1 + count(|stat_perm| >= |stat_obs|)) / (B + 1); the plus-one on each side is not cosmetic, it keeps the p-value away from exactly zero and keeps the test valid at finite B.
For CUPED, fit theta = cov(y, x) / var(x) on the pooled data and use that same theta for the observed statistic and every relabelling. Pooled theta, like the pooled cap, carries no label information, so where in the loop you compute it is again only a performance question; fitting theta within arms is what goes wrong, because the adjusted outcome then depends on the labels and an observed-label fit frozen across all 20,000 relabellings breaks the match between observed and permuted statistics. x must be measured entirely before launch, which consumption_pre is. Expected variance reduction is about 1 - corr(y, x)^2; measure the achieved reduction from the two null distributions rather than asserting it.

Worked solution (45 min)

cap_y = np.quantile(df.consumption_28d, 0.99); y = np.minimum(df.consumption_28d.to_numpy(float), cap_y); cap_x = np.quantile(df.consumption_pre, 0.99); x = np.minimum(df.consumption_pre.to_numpy(float), cap_x)
t = (df.arm == 'treatment').to_numpy(); n1 = int(t.sum()); n0 = len(t) - n1; obs = y[t].mean() - y[~t].mean()
rng = np.random.default_rng(11); L = np.tile(t.astype(np.int8), (20_000, 1)); rng.permuted(L, axis=1, out=L); s1 = L @ y; stats = s1/n1 - (y.sum() - s1)/n0
p = (1 + int(np.sum(np.abs(stats) >= abs(obs)))) / (20_000 + 1)
theta = np.cov(y, x, ddof=1)[0,1] / np.var(x, ddof=1); y_adj = y - theta*(x - x.mean()); repeat steps 2 to 4 on y_adj and compare stats.std(ddof=1) between the two runs.

Expected result: A two-sided p-value strictly between 1/20001 and 1, an observed effect expressed in capped billable units per account, and a CUPED null whose standard deviation is smaller than the unadjusted one by roughly sqrt(1 - corr(y, x)^2). The CUPED point estimate stays close to the unadjusted one, since the adjustment removes variance rather than shifting the effect.

Verify it yourself

The permutation null is centred near zero: |mean(stats)| is a small fraction of std(stats).
An A/A run, relabelling arm at random before the test and repeating over seeds, gives roughly uniform p-values with about 5% below 0.05.
std(stats_cuped) < std(stats) whenever corr(y, x) is materially non-zero, and the ratio matches sqrt(1 - r^2) to within simulation noise.
Moving the pooled cap computation inside the permutation loop reproduces the same p-value bit for bit at the same seed — the direct demonstration that a pooled cap carries no label information — while setting the cap to infinity reproduces the uncapped test exactly, so the winsorisation switch is independently verifiable.

Follow-up

The p-value is 0.04 with the cap and 0.31 without it. What do you report, and what did you pre-register?
Colleagues in a shared workspace can see the treated behaviour. How does that change the design and the estimate?
How many accounts would you need to detect a 5% lift given this outcome's distribution?

Sessionise interactive API traffic with a thirty-minute inactivity gap

Medium · window functions · sessionisation · gaps and islands

fct_api_request carries account_id, user_id (null for service accounts), request_at, traffic_class and http_status. Using only traffic_class = 'interactive' rows with a non-null user_id, group each seat's requests into sessions on a 30-minute inactivity threshold: a request more than 30 minutes after the previous request from the same (account_id, user_id) opens a new session. For one ISO week return, per account, the session count, the median session duration in minutes and the median requests per session. A single-request session has a duration of zero.

Approach

Filter first: traffic_class = 'interactive' and user_id IS NOT NULL. Machine traffic has no sessions in any useful sense, and leaving CI or batch rows in produces sessions that are really cron schedules.
Get the previous timestamp with LAG(request_at) OVER (PARTITION BY account_id, user_id ORDER BY request_at). Partitioning by user_id alone stitches one person's work across two different accounts into one fabricated session, because a human holds memberships in several accounts.
Flag a boundary where the lag is NULL or request_at - lag > interval '30 minutes'. Decide and state whether exactly 30 minutes continues the session; strictly greater is the conventional choice and needs to be written down either way.
Assign session ids with SUM(boundary::int) OVER (PARTITION BY account_id, user_id ORDER BY request_at ROWS UNBOUNDED PRECEDING), the standard running-count construction for islands.
Roll up to sessions with min(request_at), max(request_at) and count(*), then to accounts with percentile_cont(0.5) WITHIN GROUP (ORDER BY ...). Use medians, not means: session length is strongly right-skewed and one long-running client dominates the average.

Follow-up

Sessions belonging to colleagues in one account are correlated. What does that do to a t-test on session length across an experiment arm?
A long-poll or streaming endpoint keeps a connection open for hours. How do you stop it reading as one twelve-hour session?

Seven-day activation rate by signup cohort week

Easy · funnel · cohorts · censoring

dim_account carries account_id, created_at, is_internal and is_current; fct_api_request carries account_id, request_at, http_status, api_key_id and traffic_class. Build a signup cohort by the ISO week of created_at over accounts with is_internal = false. An account counts as activated when it issues a request with http_status < 400, a non-null api_key_id and traffic_class <> 'synthetic_monitor' within 168 hours of its own created_at. Return cohort accounts, activated accounts and the rate per week, and exclude any week that has not yet fully elapsed its 168-hour window.

Approach

Collapse dim_account to one row per account_id before anything else. It is a type 2 dimension, so a plan or status change gives the same account several rows; filtering to is_current = true is the cheapest correct choice here because created_at does not change across versions.
Express the window as interval arithmetic on the timestamptz column: request_at >= created_at AND request_at < created_at + interval '168 hours'. A date-difference of 7 days is a different and wrong condition for accounts created mid-day.
Test activation with EXISTS rather than a join to MIN(request_at). EXISTS short-circuits, keeps the cohort at one row per account, and cannot fan out.
Aggregate by date_trunc('week', created_at AT TIME ZONE 'UTC'), counting accounts and activated accounts, and divide as a ratio of counts.
Drop unreportable weeks: the last account in a cohort week is created just under week_start + 7 days, so the week is only complete once now() >= week_start + interval '14 days'. Without that filter the newest week always looks like a regression.

Worked solution (20 min)

Write the cohort CTE: SELECT account_id, created_at FROM dim_account WHERE is_current AND NOT is_internal AND created_at >= <window start>.
Add the activation predicate as a correlated EXISTS over fct_api_request on account_id with the four conditions: http_status < 400, api_key_id IS NOT NULL, traffic_class <> 'synthetic_monitor', and the 168-hour bracket.
Group by date_trunc('week', created_at AT TIME ZONE 'UTC'); select count() AS cohort_accounts, count() FILTER (WHERE activated) AS activated_accounts, and the ratio cast to numeric.
Add HAVING or an outer WHERE that keeps only weeks where week_start + interval '14 days' <= now().
Spot-check one account that activated on hour 167 and one that activated on hour 169 to confirm the boundary is exclusive at the top.

Expected result: One row per ISO cohort week with cohort_accounts, activated_accounts and activation_rate between 0 and 1; the newest row is at least 14 days old, and accounts with no rows at all in fct_api_request appear in the denominator with activated = 0 rather than disappearing.

Verify it yourself

Summing cohort_accounts across weeks equals the count of non-internal current-version accounts created in the same window.
Re-running the query a day later leaves every previously reported week byte-identical; any movement means the censoring filter is wrong.
activated_accounts is never greater than cohort_accounts, and the count of distinct account_id in the activated set matches a separate COUNT DISTINCT over the fact table.

Follow-up

The median time-to-first-successful-call is more informative than a fixed-window rate. Why can you not compute it from this query, and what estimator does it need?
How would you separate accounts that never called from accounts that called and got only 4xx responses, and which of those is a product problem?

How would you approach building a recommendation system for our products?

Medium · product sense and metrics

Approach

Fix the population and the time window before naming any metric.
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

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?

What steps would you take to analyze the churn rate of our customers?

Medium · product sense and metrics

Approach

Decompose the metric into the rates that drive it, and say which one you would check first.
Restate the decision this analysis has to support, and who acts on the answer.
Fix the population and the time window before naming any metric.

Follow-up

Which segment would you cut first, and what would that rule out?
How would you detect that the metric is being gamed rather than genuinely improving?

If you were given a dataset with thousands of records, how would you summarize the key insights?

Medium · product sense and metrics

Approach

State what result would change your recommendation, so the answer is falsifiable.
Name one primary metric, then the guardrail that stops it being gamed.
Fix the population and the time window before naming any metric.

Follow-up

How would you detect that the metric is being gamed rather than genuinely improving?
Which segment would you cut first, and what would that rule out?

How do you prioritize tasks when working on multiple projects?

Medium · product sense and metrics

Approach

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.
Restate the decision this analysis has to support, and who acts on the answer.

Follow-up

Which segment would you cut first, and what would that rule out?
How would you detect that the metric is being gamed rather than genuinely improving?

Explain the difference between supervised and unsupervised learning.

Medium

Approach

State your assumptions explicitly before working the problem.
Work from the decision backwards to the evidence you would need.
Clarify what is being asked and what a complete answer would contain.

Follow-up

How would you know your answer was wrong?
What assumption would you test first?

Stop metering retried requests: design the metric that decides it

Medium · guardrail conflict · billing integrity · retries · expected value

The platform meters accepted requests. A proposal carries three clauses: stop metering fct_api_request rows where is_retry = true, stop metering rows with a 4xx status, and stop metering rows with http_status >= 500. Its author has attached one figure to all three together, roughly 4% of requests_thousands volume. You have fct_api_request (account_id, is_retry, idempotency_key, http_status, traffic_class, billable_units, request_at) and fct_usage_daily (billable_quantity, net_amount_cents, cogs_cents). Size each clause separately before arguing about any of them, then define the primary metric, the guardrail that genuinely conflicts with it, and how you resolve that conflict for a decision that has to be made this quarter. Revenue falls this quarter with certainty; any benefit appears at renewals up to twelve months out.

Approach

Size the three clauses before accepting the headline 4%, because one of them is a no-op. billable_units is defined as zero for requests that failed with a 5xx, so the third clause removes no metered volume at all. Confirm that in the data rather than trusting the column comment: if sum(billable_units) over rows with http_status >= 500 is not zero, the metering pipeline contradicts its own definition and that is a billing defect to file before any pricing conversation happens. The two live clauses are 4xx failures, which are metered in full, and retries that did not themselves end in a 5xx.
Size the two live clauses as a union, not a sum. A retry can return 4xx, so the clauses overlap and adding their volumes counts that intersection twice. Partition the trailing 90 days into four mutually exclusive buckets instead: clean (is_retry = false, http_status < 400), non-retry 4xx, retry with http_status < 500, and http_status >= 500. Report the removable share per account as a distribution; if the mass sits in a handful of accounts this is a commercial conversation with those accounts rather than a platform-wide pricing change.
State the conflict rather than dissolving it. The primary metric, net metered revenue per paying account, and the integrity guardrail, the share of metered volume that is retried or failed traffic, move in opposite directions by construction. No redefinition removes that. The job is to price the trade-off, not to make it disappear.
Show the perverse coupling with data, and be exact about its mechanism. Because a 5xx already carries zero billable_units, the platform is not paid directly for its own failures; it is paid for the retries and the client-side 4xx traffic those failures provoke, which is one step removed and therefore easy to miss. Cross-tab each account's trailing 28-day 5xx rate against its metered volume in the same window. If metered volume rises with error rate, that indirect coupling is the actual argument for the change.
Resolve on expected value with the uncertainty stated. The revenue loss is computable and near-certain; the renewal benefit is not, so invert it and state the break-even: how many basis points of gross logo retention on the renewal-eligible base would offset the loss. That converts an argument about values into an argument about one number. Then propose the measurement that would settle it instead of claiming a readout you do not have: stage the rollout by renewal cohort so accounts whose terms end soonest are treated first, read out on gross logo retention on the renewal-eligible base, and say honestly whether the number of annual renewals in the window can support that estimate at all.

Worked solution (30 min)

Test the third clause first: over the trailing 90 days compute count(*) and sum(billable_units) from fct_api_request where http_status >= 500. The sum must be zero, because billable_units is defined as zero for 5xx failures. If it is zero the clause removes nothing and drops out of the analysis; if it is not, stop and raise a metering defect, because every volume figure downstream of that column is then suspect.
Compute trailing-90-day metered volume per account in the four mutually exclusive buckets: clean, non-retry 4xx, retry with http_status < 500, and http_status >= 500. Roll the total up to the requests_thousands SKU and reconcile it against fct_usage_daily billable_quantity for the same window.
Convert the two removable buckets to money using each account's realised rate, net_amount_cents / billable_quantity from fct_usage_daily, because list rate overstates revenue for every discounted account.
Annualise the revenue at risk and divide it by the ARR of the renewal-eligible base to express the break-even as an improvement in gross logo retention, in basis points.
Cross-tab account 28-day 5xx rate deciles against metered volume per account to establish whether the error-to-revenue coupling, which can only run through retries and 4xx rather than through the failed requests themselves, is real or a story.

Expected result: Three clause sizes rather than one blended figure: zero for the 5xx clause, since billable_units is already zero there, and separate volumes for 4xx failures and for retries that did not 5xx, combined as a union so the retry-and-4xx intersection is counted once. On top of that, a single break-even figure stated as the improvement in gross logo retention on the renewal-eligible base needed to offset the metered revenue actually removed, and the per-account concentration of that removable volume. The answer says that the primary metric and the integrity guardrail conflict by construction, and resolves it with the break-even number and a staged rollout rather than by moving a definition.

Verify it yourself

sum(billable_units) over rows with http_status >= 500 must be exactly zero; any other result contradicts the documented definition of the column and invalidates the volume figures derived from it.
The four buckets must be mutually exclusive and must sum to total billable_units, which must in turn reconcile to fct_usage_daily billable_quantity within the known restatement tolerance; a large gap means the request-to-SKU mapping is wrong.
The two removable buckets must be reported as a union; their arithmetic sum will exceed the union by exactly the retry-and-4xx intersection, and a write-up that quotes the sum has overstated the revenue at risk.
Requests sent with no idempotency_key cannot carry is_retry = true, so the measured retry share is a lower bound and the write-up must say so.

Follow-up

Suppose the two live clauses turn out to remove 2.6% of consumption revenue. How much improvement in gross logo retention on an annual-contract base pays that back, and over what horizon does the payback land?
A retry sent without an idempotency_key cannot be flagged as a retry. Which direction does that bias your estimate of the removable volume, and how can you bound it?

Consumption per account always dips in the last four days

Easy · pipeline lag · restatement · reporting windows

Your consumption dashboard reads billable units per paying account from fct_usage_daily (account_id, sku_code, usage_date, billable_quantity, net_amount_cents, is_restated, first_written_at, restated_at, updated_at). Every refresh shows the final three to five days declining, and a review is scheduled on that shape. Establish empirically how long a usage_date takes to settle, separately per sku_code, and specify the trailing exclusion window every reported figure should use. Derive the number from the data; do not adopt a convention.

Approach

Treat settlement as a measurable curve rather than a belief: for each historical usage_date, compare the total captured as of age d days against that date's final settled total.
Do it per sku_code, because metering paths differ. A per-request SKU lands within hours, while a storage-month SKU is produced by a daily sweep and lands later, so one global lag number is wrong for at least one of them.
Pick the age at which a stated percentile of dates reaches a stated completeness threshold, for example the tenth percentile date reaching 99.5 percent of final, and take the slowest SKU's age as the dashboard's exclusion window.
Separate late arrival from restatement. Late rows raise totals and are fixed by waiting; restatements can move either way and are not, so report the share of the gap from each.
Encode the exclusion inside the query rather than in the chart, so anyone reusing the SQL inherits the rule, and never compare a fresh partial period against a settled one.

Follow-up

Month-end invoicing needs a number before settlement completes. How would you publish an early estimate with an honest uncertainty band attached?
What monitor would tell you the settling time has changed, without anyone remembering to re-run this analysis?

Suggested practice checklist

For someone who can already write the query and train the model but stalls when asked what to measure or whether a change is worth making. Metric definition and case structure come first; the technical work is kept as maintenance rather than the centre of the week.

Day 1 — Metric anatomy

For three products you use daily, write one primary metric, two input metrics that plausibly move it, and one guardrail that would catch a cheap way of moving the primary at the cost of the product.
For one of them, specify the metric precisely enough that two analysts would return the same number: numerator, denominator, unit of observation, time window, and how returning and deleted accounts are treated.
Pick a ratio metric and write what happens to it when the denominator shrinks for reasons unrelated to the numerator, with a concrete example of that happening.

Deliverable: A one-page metric tree for one product, with the primary metric written as an unambiguous spec.

Day 2 — Diagnosing a drop without guessing

Take the prompt "weekly active users fell 8 percent week over week" and write the segmentation plan before proposing any cause: platform, region, tenure cohort, acquisition channel, and whether the movement sits in the numerator or in a changed denominator.
List the instrumentation failures that manufacture fake drops (a client release that stopped firing an event, a bot filter change, a shifted date boundary or timezone) and write the query that rules out each one.
Rehearse stating the boring explanations first, seasonality and day-of-week composition, before reaching for a product cause.

Deliverable: A drop-diagnosis checklist short enough to recite from memory in under a minute.

Day 3 — Should we build it

Take a feature idea and write it as a bet: what you believe is true, what would have to be true for it to pay off, the metric that would confirm it, and the effect size that would justify the engineering cost.
Size the opportunity top-down and bottom-up, then reconcile the two numbers in writing instead of quoting whichever is friendlier.
Write the counter-metric that would make you kill the feature even if it wins on the primary metric.

Deliverable: A one-page product memo ending in a decision rather than a list of considerations.

Day 4 — The places aggregate numbers lie

Construct a Simpson's paradox numerically: two segments where the treatment wins within each segment yet loses overall, and identify the shift in segment weights that causes it.
Take a heavy right-tailed quantity such as revenue per user and write why the mean is the wrong summary, which percentile you would report instead, and what a moving mean with a stable median tells you.
Write your definition of a session for the product from day one, then name two real behaviours it misclassifies.

Deliverable: One page holding a worked Simpson's paradox table and a session definition with its two known failure cases.

Day 5 — Technical maintenance, aimed at metrics

Solve four timed SQL prompts that all end in a ratio metric, so the question of grain stays live in every answer.
Compute a 95 percent confidence interval for a proportion on a small sample, and state why the normal approximation is unreliable when either np or n(1 minus p) falls below roughly 10, along with which interval you would use instead.
Take one metric from your day-one tree, write the query that computes it correctly, then write the query that computes it wrong in the most plausible way and explain how you would notice.

Deliverable: Four solved prompts plus a matched correct and plausible-wrong query for one metric.

Day 6 — Turning engineering work into data science stories

Write three project stories as situation, decision, trade-off, outcome, each carrying one number and one thing you got wrong.
For the story you will lead with, prepare an answer to "what would you do differently" that names a decision you made, not a constraint you were handed.
Practise the sentence that reframes a systems project as a question project: the question the work answered, ahead of the pipeline it shipped.

Deliverable: Three written stories with the lead story delivered aloud and timed under four minutes.

Day 7 — Mock case and gap list

Run a 40-minute mock case with someone playing a product manager who pushes back on your metric choice, and record it.
Listen back and mark every moment you proposed a solution before the success metric existed.
Rewrite those moments as the question you should have asked, and rehearse the first 90 seconds of the case until scoping comes before solving.

Deliverable: A recorded case plus a rewritten opening 90 seconds.

Ownership when the result is inconvenient

Sometimes the honest read is that the initiative did not work, and the person who commissioned the analysis was hoping otherwise. Interviewers want to know whether you softened it. Prepare the case where you delivered an unwelcome result, how you presented the uncertainty without hiding behind it, and what the team did next.

How do you handle missing data in a dataset?

Medium · behavioural and stakeholder questions

Approach

State the situation in two sentences and spend the rest on your reasoning.
Quantify the outcome, including what you would not claim credit for.
Name the disagreement or constraint, and how you resolved it with evidence.

Follow-up

What would you do differently if you ran that project again?
What did you decide not to do, and why?

What motivates you to work in data science?

Medium · behavioural and stakeholder questions

Approach

Close with what you would do differently, concretely.
Quantify the outcome, including what you would not claim credit for.
Name the disagreement or constraint, and how you resolved it with evidence.

Follow-up

What would you do differently if you ran that project again?
What did you decide not to do, and why?

Disagree with a product manager about an adoption claim

Medium · conflict · telemetry hygiene · adoption metrics

A product manager is about to present that a new SDK release drove a 40 percent rise in requests among adopting accounts, computed from fct_api_request counts grouped by sdk_version. You find the rise is concentrated in traffic_class equal to ci, that rows with is_retry true grew alongside it, and that restricting to interactive non-retry traffic leaves a 3 percent lift. The launch review is in two days. Decide how you raise this, with whom and in what order, and what you propose the claim becomes.

Approach

The interviewer is probing whether you can correct a colleague without ambushing them, and whether your own counter-analysis carries the caveats theirs lacked. Go to the product manager privately before the review. A correction delivered in the room is a status move and loses the argument you are actually trying to win.
Bring a decomposition rather than a verdict: the same accounts and window, requests split by traffic_class with retries held out as their own column, so their 40 percent and your 3 percent reconcile line by line and neither has to be taken on trust.
Reproduce their figure exactly first. If you cannot land on 40 percent with their method, you do not yet know what you are disagreeing with.
Ask whether the continuous-integration lift is itself valuable. An account wiring the SDK into its pipeline has increased integration depth, which is the dominant switching cost in this domain, so the honest claim may be that integration depth rose while interactive usage moved 3 percent. Improving the claim beats deleting it.
Name the mechanism that makes the raw count dangerous: clients retry when the platform degrades, so retry volume climbs exactly when the customer is most at risk. Pull the 5xx rate for the same accounts and window before anyone concludes anything, and note that billable_units is zero on 5xx rows, so request counts and billable quantities diverging is itself the signal.
Close with a standing definition for launch metrics so the next release does not repeat the exercise.

Follow-up

The product manager argues that continuous-integration traffic is real usage and declines to split it out. Is that position defensible, and under what metric definition?
Suppose the 5xx rate for those same accounts also rose 40 percent. What is the claim now?
The review happens and the raw number is presented regardless. What do you do next, and what do you not do?