Practice the SQL patterns that recur in Meta (formerly Facebook) data analyst and data engineer interview rounds. Window functions, aggregations, date functions — with AI tutoring and instant feedback.
What is sourced: Meta does not publish its interview format, so every specific about their process on this page is what candidates and prep guides described publicly — each one carries its source and the date it showed. Formats change; treat them as reported, not official. What is ours: the practice questions are SQL Quest challenges, picked because their SQL matches the patterns those sources report. They are not questions Meta has asked.
Meta does not publish the format. Each row is what candidates and prep guides have described publicly, with its source; where sources disagree, the row says so. Formats change — treat every specific as reported, not official. The sources below describe Data Engineer; other teams at Meta may run a different loop.
| The technical screen | Two candidate reports agree on five problems in the screen: three SQL and two Python.[1][2] |
|---|---|
| Environment | You are given the schema, and in both the screen and the onsite the code is not expected to be runnable — one report adds that the onsite also gives you the expected output. Interviewers differ in how strict they are about syntax.[1] |
| What the SQL covers | Reported as medium difficulty: window functions, subqueries, CTEs and date/time work.[2] |
| The onsite | Described as several rounds spanning product sense, data modelling, SQL, Python and behavioural. Neither report gives a length for the individual rounds, so this page does not state one.[1] |
Sources: Blind — "Meta Data Engineer 2025 Interview Experience" (candidate report, E4 offer) (5 Apr 2025); Blind — "Meta Data Engineer: Technical Screen-Help" (poster plus two replying candidates) (17 Mar 2025). Accessed 14 Sep 2026.
The 22 challenges tagged Meta in the SQL Quest bank, with every raw challenge tag resolved to the 9 canonical skills. Each share is the portion of those 22 challenges that exercise the skill — a challenge exercises several, so the shares do not sum to 100%. This is the composition of the practice set on this page, not a measurement of Meta’s interview.
These are SQL Quest challenges chosen because their SQL matches the work — Querying Basics, Aggregation & Grouping, Window Functions. They are not questions Meta has asked, and this page does not claim to know its questions. All of the six play free; each card opens the challenge itself.
Analyst Round
Updated September 2026 · 22 Meta-tagged challenges in the bank · 7 play free, 15 are Pro
Analyst-round SQL questions keep returning to six shapes: rank inside a group, accumulate across rows, compare a row with its neighbour, count who is still here, find the repeat, and filter by existence. Each family below lists every Meta-tagged challenge that drills it. All 22 run in the browser and show the solution once solved.
DENSE_RANK versus ROW_NUMBER on ties, NTILE quartiles, PERCENT_RANK, and a median with no percentile function. Say which tie-break the question wants before you write it — that sentence is usually worth more than the query.
Second Highest Salary per Department
DENSE_RANK inside PARTITION BY, so tied salaries share a rank; one-person departments drop out.
Median Salary Without PERCENTILE
ROW_NUMBER plus COUNT(*) OVER (), then average the middle one or two rows.
Salary Percentile Ranking
PERCENT_RANK over the whole company, rounded to two places.
Fare Percentile Ranking
NTILE(4) into fare quartiles, with a name tie-break in the ORDER BY.
Department Salary Percentile Buckets
NTILE(4) per department, then CASE to label each quartile.
The frame clause and the empty OVER (). A running SUM, a three-row rolling average, and a category's share of the whole without a second query.
Cumulative Distinct Customers Over Time
First-order date per customer in a CTE, new customers per day, then SUM() OVER (ORDER BY date). The one Hard Meta-tagged challenge that plays free.
3-Movie Rolling Average Revenue
AVG() OVER (… ROWS BETWEEN 2 PRECEDING AND CURRENT ROW), rounded.
Revenue Share by Category (Window %)
SUM(SUM(total)) OVER () as the denominator — a window on top of a GROUP BY.
Compare a row with its neighbour. LAG finds the gap, a running SUM of the gap flags becomes the session id, and the ROW_NUMBER-difference trick turns consecutive dates into islands.
Order Sessionization by Customer
LAG on order_date per customer, CASE for a 30-day gap, running SUM for the session number.
Island Length Classification
Date minus ROW_NUMBER groups consecutive days; MIN, MAX and COUNT per island, then a CASE label.
Count who is still here. Daily actives, customers active in two or more months, days since the last order, and a first-order-month cohort table.
Daily Active Customers
COUNT(DISTINCT customer_id) per order_date — DAU on an orders table.
Multi-Month Active Customers
JOIN, then HAVING COUNT(DISTINCT strftime('%Y-%m', order_date)) >= 2.
Customer Recency Analysis
MAX(order_date) per customer and julianday arithmetic for days since the last order.
Customer Retention Cohort
MIN(order_date) month as the cohort, JOIN back to activity, count per cohort month and order month.
Quarterly Hiring Cohort Report
CASE on the month to build the quarter, GROUP BY year and quarter.
Find the repeat. The orders table joined to itself on the customer with the date window in the ON clause, and the employees table joined to itself on id + 1.
Detect Repeat Buyers Within 7 Days
orders o1 JOIN orders o2 on the customer, o2 strictly later, at most 7 days apart; DISTINCT the pairs.
Self-Join: Repeat Orders Within a Week
The same shape carrying both products — watch for the duplicate pair.
Consecutive IDs
JOIN employees e2 ON e2.emp_id = e1.emp_id + 1, same department.
Filter by existence. Correlated EXISTS, the every-category relational-division question, and a LEFT JOIN with COALESCE so empty groups still show.
EXISTS vs IN: Departments with Top Performers
Correlated EXISTS on the same table, then AVG per department.
Departments With High Earners
EXISTS with a salary threshold; DISTINCT department names.
Customers with Orders in ALL Categories
HAVING COUNT(DISTINCT category) = (SELECT COUNT(DISTINCT category) FROM orders).
Membership Tier Revenue Analysis
customers LEFT JOIN orders, COALESCE(SUM, 0), and a COUNT DISTINCT on each side of the join.
General guidance about analytics SQL work — not sourced from Meta and not a description of its process. We have no dated, citable source for how Meta runs its SQL round, so this page states none.
Skillmap
Ten questions, no signup. You get a readiness score weighted to the SQL this page covers, your Skillmap across joins, window functions, aggregation and the rest, and the weakest skill to practise first.
Drill the skills the Meta set leans on, one at a time: GROUP BY exercises · Window function practice · CTE practice · JOIN practice · Date function practice — or browse every SQL practice question.
Every question in the Meta set, one page each with the schema and a hint: EXISTS vs IN: Departments with Top Performers · Quarterly Hiring Cohort Report · Departments With High Earners · Customer Recency Analysis · Membership Tier Revenue Analysis · Multi-Month Active Customers · Cumulative Distinct Customers Over Time · Salary Percentile Ranking · Order Sessionization by Customer · Detect Repeat Buyers Within 7 Days · Consecutive IDs · 3-Movie Rolling Average Revenue · Fare Percentile Ranking · Daily Active Customers · Revenue Share by Category (Window %) · Median Salary Without PERCENTILE · Customers with Orders in ALL Categories · Self-Join: Repeat Orders Within a Week · Island Length Classification · Department Salary Percentile Buckets · Customer Retention Cohort · Second Highest Salary per Department.
Practice the 22 Meta-tagged challenges with AI tutoring, instant feedback, and real datasets. Free to start, no signup.
Practice the Meta Set →Works on Chrome, Firefox, Safari, Edge · No plugins · No downloads
Interviewing at more than one company? The same patterns carry: Google · Amazon · Apple · Netflix — or the full company-by-company interview guide.