Practice

Find orders from the last 30 days with no matching refund record using the anti-join pattern.

Find duplicate email addresses with occurrence counts and signup date ranges using GROUP BY and HAVING.

Compare event volumes on weekdays vs weekends using CASE WHEN pivot pattern.

Find paying customers inactive for 90+ days with their subscription tier and lifetime spend.

Product analytics wants a quick engagement snapshot.

Find users exceeding average activity count.

Find rows whose value exceeds a joined reference.

The admin team needs all API tokens that have actually been used.

The security team is auditing API token usage.

The monetization team wants to know how many active users made at least one transaction in April 2026.

Two level GROUP BY to build a frequency histogram.

Finance is pulling annual ad revenue.

The infrastructure team is mapping alert hotspots.

The data catalog team needs a list of all distinct regions present in the infrastructure nodes table.

The cloud finance team needs total annual cloud spending.

The finance team wants total cloud spending and the number of distinct services billed in each fiscal year.

The data platform team is auditing pipeline failures.

The analytics team needs a list of users who were active in April or May.

The platform team needs all API endpoints that contain the word 'health' (case-insensitive).

The delivery team tracks successful deployment authors.

The compliance team needs a full audit trail for the 'auth-service'.

The ad analytics team needs a quick benchmark.

The platform team is benchmarking build performance.

The data quality team computes average check fail rates per table.

The infra team is sizing capacity for GPU nodes.

HR analytics is benchmarking compensation.

The ML platform team needs the average accuracy for models scoring between 91 and 100 inclusive.

For a quarterly reliability review, the team needs the average latency broken out by health status for 'healthy', 'degra

The SRE team wants the average latency for each API call status.

The cloud finance team needs average CPU and memory utilization broken down by region and node type.

The product analytics team wants the average rating for products in each category.

The SRE team needs average response time broken down by hour of day, sorted chronologically.

The performance team wants the average latency for the '/api/search' endpoint.

The search relevance team is benchmarking query volume per user.

The performance team wants to benchmark session lengths by device.

The marketplace team needs the top 3 sellers by total amount in each product category for January.

Join tables and label rows as over or under budget.

The product team wants all distinct product names that are budget-friendly, meaning priced between 5 and 20 inclusive.

The marketing analytics team wants to measure campaign reach.

The finance team needs total ad revenue grouped by campaign.

The merchandising team is reviewing category performance for 2026.

Sum transactions for a specific payment type.

The networking team needs all CDN log entries related to image requests.

The DNS team is filtering lookup records.

The growth team is debugging URL patterns.

The cost optimization team is finding the lowest cloud spend per region.

The commerce team wants to identify each customer's cheapest purchase.

The network team needs to identify CDN edges that serve cached content with no errors.

The data quality team found that the latency column in service health records contains some non-numeric strings.

The product analytics team needs all ad impression records where the user clicked.

The FinOps team is reviewing spend by team.

The data quality team suspects duplicate records in the users table.

The batch processing team needs all completed priority-1 jobs.

The infra team is inventorying compute nodes across key regions.

The content team wants a quick audit of items created by 'alice' or 'bob'.

The content team wants a popularity snapshot.

The product analytics team is investigating historical engagement trends and needs all content items published during 20

The content team is preparing a catalog sorted by length.

The data catalog team needs a count of records by content type, sorted alphabetically by type.

The content moderation team needs to audit a specific creator.

Finance is calculating efficiency ratios for cloud services.

The platform team wants to understand the breadth of the service mesh.

The ops team needs a quick count of how many infrastructure nodes are in the 'us-east-1' region.

For the quarterly infrastructure report, the team needs the minimum, average, and maximum CPU utilization across all nod

For the internal directory, the HR team needs each customer's first and last name combined into a single full name field

Count distinct users per day and per week.

The platform analytics team needs daily unique user counts across mobile and web sessions.

The ops dashboard needs daily deployment counts.

HR is benchmarking compensation across teams.

The release management team needs a deploy frequency report showing how many deployments each service has had, sorted fr

The ML platform team is cataloging model frameworks by deployment status.

The ops team is reviewing deployment outcomes.

The ML platform team is auditing model versions.

The growth team is profiling power users by device type.

The SRE team needs to audit disabled feature flags.

The analytics team needs a deduplicated list of all referrer sources for blog content.

The product catalog team needs a quick inventory check.

The data governance team needs a historical audit.

Count members grouped by location.

The product team wants to classify error severity for reporting.

The observability team needs to enrich error data with service health context.

The analytics team is running a signup cohort analysis and wants to isolate February signups with odd-numbered user IDs.

HR is auditing onboarding patterns and wants to look at June signups whose user IDs are even numbers.

The product team is tracking event volume on key days.

The product analytics team wants to see total event volume for each calendar month (1 through 12), combining all years t

The product team wants to find versatile users who have triggered both 'click' and 'purchase' event types in event_data.

The procurement team needs a count of distinct AWS services that have at least one cost entry of 200 or more.

HR needs to find employees with extreme compensation.

The SRE team is tracking failed deployments.

The data quality team is investigating missing data.

The HR analytics team needs a cohort snapshot of all users who signed up during February 2026.

Select rows matching a text suffix pattern.

The people ops team needs a clean roster for the all-hands.

The HR team needs to find all deploy authors whose last name is 'Chen'.

The devops team wants to know when each repository first appeared in the CI pipeline.

The migrations team wants the very first migration ever applied, identified by the smallest migration ID.

The batch operations team needs each job's initial row count.

The ops team needs a full roster of customers alongside their order history, including customers who have never placed a

The SRE team is hunting for specific error patterns.

The SRE team wants to see the distribution of health check results for auth-service.

The observability team needs a quick service inventory showing how many health checks exist for each service.

The search relevance team wants to flag power searchers.

The alerting team wants to review stale alerts.

The A/B testing team wants to know whether the control or treatment variant produces a higher average metric value.

Self join to find employees outscoring their manager.

The executive dashboard needs peak cost metrics per team.

The SRE team wants the 3 endpoints with the highest peak latency, showing each endpoint and its maximum latency value.

The content team is evaluating high-engagement creators.

The procurement team needs a list of all products priced above 100.

The sustainability team needs a quick metric: what percentage of products are both in stock and rated 4 or above?

The marketing team wants to find ad campaigns that generated more than 100,000 in total revenue during 2026.

The product team wants to know how many distinct endpoints received 100 or more total API calls during February 2026.

The data platform team needs all batch jobs that processed more than 3 million rows.

The growth team wants to find the year in which the influencer ad campaign surpassed 2000 total clicks.

Product analytics wants to spotlight the holiday sale campaign.

The engagement team wants to find dormant trial users.

The inventory team needs a count of how many products are currently in stock.

Max minus min metric per group, ranked by gap.

The regional sales team is pulling last month's numbers for the APAC region.

Combine two related tables with a join.

Find the group with the highest member count.

The data pipeline team needs a quick tail check.

The migrations team wants the most recently applied migration, identified by the largest migration ID.

The infra team wants to know the last time each server reported a log entry.

Engineering leadership wants to compare each service's latency against the regional average.

The data platform team suspects stale records in employee metrics.

For each user in user_sessions, find their most recent session start date.

The deployment dashboard needs the latest version deployed for each service.

The SRE dashboard needs a count of server log entries at each log level, sorted by level alphabetically.

The SRE team needs incident frequency by day of week.

The compensation benchmarking team wants the five highest distinct metric values from employee metrics.

The deploy report needs full identification for the deployment(s) with the longest duration.

The search team is analyzing query patterns.

The CDN team is investigating low-traffic edges.

The product team is sizing a migration cohort.

The cost optimization team wants to find the cheapest service category.

The SRE team wants to validate that certain endpoints are performing well.

The security team needs all data quality checks classified as 'low' severity.

The data quality team wants to identify low-throughput data pipelines.

The SRE team is investigating underperforming services.

Find the highest metric in each group.

The capacity team is scoping memory-intensive workloads.

The release engineering team wants to know how many CI builds were triggered by merges in 2026.

The content moderation team is scanning for flagged terms.

The support team is investigating specific user complaints.

Find the max-min spread within each group.

The data integrity team is spot-checking records where two metrics coincidentally match.

The capacity team needs to scope mid-tier nodes.

The compensation team needs a filtered report of cost allocation entries where the amount falls between 500 and 10,000 i

The leaderboard team needs all batch jobs that rank between positions 8 and 10 by rows processed.

The notification platform team found an anomaly: some users from 2026 have no email on file but their account status is

The product analytics team is measuring engagement by device.

For the platform health dashboard, count the distinct active users per endpoint during January 2026.

Sum amounts by category and month.

The ops team tracks monthly deployment frequency.

The growth team needs to track user registration trends by month.

Count transactions per user per month.

The engagement team tracks monthly reach per ad campaign.

The SRE team needs all warning-level server log entries that occurred before noon.

The analytics team wants the single most common status among batch jobs whose name contains 'etl'.

The security team needs each user's most recent API token activity.

Product analytics wants a quick directory of users sorted alphabetically by username, with ties broken by age bucket in

The platform team wants to find users who have logged in from devices running different operating systems across session

The finance team needs cloud cost amounts for the providers AWS, GCP, and Azure.

The experimentation platform team needs to find users who participated in both an 'onboarding' variant and a 'retention'

The inventory team wants to find products that exist in the catalog but have never been ordered.

The infra team is triaging node issues and needs all infrastructure nodes located in 'us-east-1', 'us-west-2', or 'eu-we

The infrastructure team needs a region-level summary showing the total number of nodes and the number of distinct node t

The on-call team is reviewing alert acknowledgments.

The content team is filtering out draft content.

The marketing team needs the total number of push notifications that were opened on January 2, 2026.

Find the 3rd highest amount from an employee table. A classic consulting firm screener.

Select the row with a specific rank position.

Understand how NULLs affect join results.

The growth team wants to spotlight extremes in the user base.

The product team wants to see review volume for the lowest-rated products.

The platform team needs the overall average latency across all API calls.

The platform team tracks user activity windows per device type.

The ad revenue team is looking for the peak earning moment.

For the quarterly investor deck, the data team needs peak metrics per department.

The product team wants to find the month with the most users who had sessions but never made a purchase.

The cost optimization team wants to identify the single billing month with the highest total cloud spend.

The batch processing team needs to audit stuck jobs.

The data catalog team needs a lineage summary.

Compute current-to-initial value ratio per period.

The data team is auditing feature flag ownership and needs all feature flags owned by a platform team, along with each f

The feature flags team needs all flags owned by the 'platform' team where the flag name contains the word 'beta'.

The reliability team wants to understand low-restart pods.

The SRE team is auditing deployment volume.

The DevOps team wants to count how many deployments went to the production environment during April or later (month 4 on

The localization team needs a quick text transform on the product catalog.

For a compact dashboard, the team needs just the first three characters of each product name, sorted by product ID.

The search quality team wants to count search queries originating from the products page where the search term contains

The content team wants to find stale inventory.

Finance wants to spotlight the most profitable product categories.

The marketing team needs cost-per-acquisition for campaigns advertised via push notifications (campaign names containing

The FinOps team tracks cost fluctuations between the start and middle of the year.

The product analytics team needs Q2 2026 search volume.

The release engineering team needs deployment counts broken down by quarter.

The SRE team wants to identify recurring error types.

The infrastructure team needs to flag busy regions.

The campaign team wants all ad impressions tied to retargeting campaigns.

Finance needs a quick revenue total for users alice and bob.

The engineering manager wants a workload summary.

The infra team needs to find pairs of servers in the same region that are both healthy.

The platform team ran a developer satisfaction survey but forgot to collect region data.

The QA team needs the distribution of HTTP status codes for API calls to the health endpoint.

The product analytics team is reviewing search behavior for a specific user.

The search relevance team wants to audit queries.

Find the second largest distinct value using DENSE_RANK.

Find the second largest value without LIMIT.

The platform team wants to know how frequently each service appears in the alert events table.

The reliability team needs to surface the noisiest services.

The growth team tracks user expansion by service.

The security team needs to identify users who logged in during a specific window.

The product analytics team is segmenting user activity by device.

The hiring team wants to track recent growth.

HR is counting new hires during a specific window.

Find orders delivered after the promised time.

The SRE team needs to find production deployments that took longer than 150 seconds.

The data governance team needs to sort API token scopes for a compliance report.

The infra team is auditing critical nodes and needs to find all nodes with a database-type designation, along with each

The SRE team needs a count of successful API calls to the deploy endpoint, where success means a 200 status.

The reliability team wants to know which data pipelines have completed successfully and how often.

The operations team is reviewing successful production deployments that have duration data.

The trust and safety team flags sessions shorter than 5 seconds as potential bot activity.

The ad ops team needs to find high-value campaign impressions.

The data platform team needs to find the third largest batch job by total rows processed.

The social graph team needs to count chat message threads that user 1 is not part of.

The cost optimization team wants the three lowest distinct cloud cost amounts.

Compute multiple date windowed aggregates in a single query.

The data engineering team suspects some service health records have an unknown status marked as '?

The SRE team is reviewing server logs for incident postmortems.

The content catalog team is auditing naming conventions.

The data platform team needs the total rows processed by the top 100 batch jobs ranked by rows completed.

The platform team wants the top 10 batch jobs by rows processed.

The ML team wants to spotlight top model performance.

For the quarterly product review, rank endpoints by their peak single-request latency in 2026.

The network team is auditing the longest DNS lookups.

The revenue team wants to rank ad campaigns by total earnings.

Engineering management wants to identify the scope associated with the highest-value API tokens.

Rank regions by their average metric value.

The ML team wants to surface their best-performing deployed model.

The product team wants to know which device type generates the most user sessions.

The content team wants to find all content items that held the number one position (highest rating) in their category at

The growth team is studying engagement by device type and wants to know which mobile operating system has the highest av

The ML ops team needs all models considered top-performing, meaning those with accuracy at 95 or above.

The product team wants to identify the highest-grossing categories.

The content recommendation engine needs to suggest content that fits within a user's average session duration.

Rank products by sales in a recent window.

Rank items by total revenue.

The product team ranks users by total transaction value.

The FinOps team needs the total cloud spend for the 'compute' service category.

The FinOps team needs total spend per category.

The finance dashboard needs the total allocated budget for the engineering team.

The data pipeline team needs row-level context alongside aggregates.

Finance needs a summary of each customer's total spend.

The executive dashboard needs a quick snapshot: the total number of distinct users who have at least one transaction and

The ML team needs all distinct feature names reviewed by the pipeline_v2 source that have a recorded average value.

Simple select progressing to a join

The data quality team found trailing whitespace in endpoint names.

The search team discovered leading whitespace in some search terms.

The content catalog team needs a quick count of how many content items have 'tutorial' somewhere in the title.

Find rows with no matching entry in another table.

The capacity planning team needs a quick headcount.

The product team needs a quick count of search engagement.

The product team wants to know how many unique users have performed a search.

The data catalog team needs a clean inventory of all streaming topics.

LEFT JOIN anti pattern to find categories with no products.

The compliance team needs to flag models that have never been evaluated.

The security team needs to identify single-scope API tokens that are currently active.

The KV store team needs a quick inventory of all entries stored in the us-east-1 region.

The platform team wants to rank users by age bucket in descending order.

The product analytics team wants per-user engagement totals.

The engagement team needs unique event type counts per user, including users with zero events.

The analytics team needs session data for user 4, but only sessions that started on a Saturday or Monday.

For each device type, count the number of distinct users who have at least one session.

The engagement team needs all users who clicked on at least one ad, along with their account status.

The engagement team needs all users who have triggered at least one purchase event, along with their account status.

The data engineering team suspects duplicate commit IDs in the repo.

The content analytics team needs view counts per page.

The ops team needs a weekly transaction volume report.

The growth team needs a simple count of how many users signed up in 2026.

Predict the row count for INNER, LEFT, RIGHT, and FULL OUTER joins when one table has 0 rows. From an EY walk-in interview.

Track monthly revenue trends with MoM percentage change and flag drops >10% using LAG window function.

Find active content readers who never made a purchase using anti-join pattern with mixed temporal scopes.

Build a weekly cohort retention report showing what percentage of users return after 7 days.

Join products to transactions and use ROW_NUMBER to find each product's most recent sale.

The SRE team wants the 10 services with the lowest uptime scores.

The security team tracks 2FA verification via push notifications.

Compute a moving average over the last 7 entries.

The subscription analytics team tracks daily active premium users.

The analytics team is benchmarking API latency percentiles.

The operations team is measuring prediction accuracy for ML models trained within the first 14 days of 2026.

The cloud cost optimization team wants to identify services with above-average spending.

The pricing team wants to find products whose base price (minimum transaction amount) exceeds the average base price acr

The compliance team needs to identify token owners who were active on November 1, 2026.

The people analytics team wants to know whether there are more active users or more users in the us-east-1 region among

The marketplace team needs total ad revenue broken down by customer country.

The compliance team needs to flag API calls made outside business hours in December 2026.

The incident response team needs alert counts by severity tier.

The SRE team needs a pivot of alert counts by severity for each service.

The platform team needs a unified, deduplicated list of all endpoints that appear in either the API calls table or the r

The data quality team needs a breakdown by endpoint showing how many API calls have an error message versus how many do

The platform team suspects duplicate latency patterns.

The platform team needs the overall churn rate for API tokens.

The marketplace team wants to profile CDN paths containing 'api' that serve the 'us-east' region.

Calculate crash rate using conditional aggregation

Aggregate bids per listing with highest bidder.

The deployment team wants to find authors who have deployed to all three environments: dev, staging, and production.

The ML platform team wants to benchmark accuracy by framework, but only for deployed models with versions between 2 and

For the weekly platform review, the team needs the average latency per endpoint per year from the API calls table, sorte

HR wants the average total compensation by department and account status.

Avg trip length per category within a time window.

The logistics team wants to measure fulfillment speed.

The API performance team wants to benchmark first-contact latency.

The search relevance team is benchmarking click-through.

Engineering leadership is reviewing code quality by author.

Product analytics needs average session duration per user.

The API performance team measures latency for follow-up calls.

The people analytics team is studying alert response patterns by severity.

The engagement team wants to measure average daily active users per endpoint for January 2026.

The product analytics team wants the average session duration for users reviewed by each content creator, treating the c

The ops team is grading batch job performance.

The ML platform team wants to find the model with the best accuracy-to-training-time ratio.

The analytics team wants to find which day of the month is best for ad revenue.

The ops team wants to know which team had the biggest month-over-month drop in deployments from March to April 2026.

The feature management team needs a quick view of every flag's on/off state as binary columns.

The mobile team wants to identify endpoints with the least POST traffic.

The QA team wants to know how many builds each author triggered, broken down by branch.

The CI/CD team wants to measure build success rates by trigger type.

The CI/CD dashboard needs a build outcome report.

The data engineering team wants to know which month saw the most pipeline runs.

The engagement team wants Friday ad click counts.

The analytics team is comparing ad engagement across campaigns.

The marketing team needs cost effectiveness for each ad campaign between 2025 and 2026 inclusive.

The marketing team is evaluating campaign effectiveness across ad formats.

The ad platform team is reviewing campaign effectiveness.

The cross-sell team targets users who engage with multiple content types.

The ops analytics team wants to profile CDN traffic patterns.

The procurement team is looking for the cheapest product with a rating above 4.

The infrastructure team wants to classify each service into categories: 'api_service' if the name contains 'api', 'cache

The engagement team wants the count of clicked ad impressions belonging to campaigns with 'holiday' in the name, limited

The search quality team needs two metrics in a single row: the percentage of all search queries where a result was click

Finance needs the minimum, average, and maximum cloud cost per provider, combining data from both the cloud costs table

The procurement team tracks cloud cost trends across billing periods.

The finance team needs total cloud spending for each combination of region and service, merging data from both the cloud

The revenue team wants total transaction revenue for each user and product pair, but only for pairs where every single t

The content recommendations team wants to surface pages users haven't discovered yet.

The product analytics team needs session metrics per content item.

The cloud team needs to identify regions with the highest and lowest cost density, defined as total cost divided by the

The FinOps team wants to see how each cost entry contributes to its category's total.

The content team wants to find creators with the most top-rated items.

The compliance team needs to count how many distinct customers have placed orders in a region different from their home

The experimentation platform needs to pair users who participated in the same experiment under different variants but on

Finance needs a cumulative average of monthly revenue for 2026.

The ops team monitors feature flag state changes.

The CRM team needs a count of customers who have never placed an order.

The messaging team wants to count how many times each user received a message with a custom type (anything other than 't

The SRE team is tracking daily error trends.

The trust and safety team needs a daily ratio of removed content to reported content.

Calculate day over day percentage change in a metric per entity using LAG.

The engagement team wants daily session metrics.

Trust and safety wants the daily spam rate for ad impressions that had an associated page view.

The ops dashboard needs the top 3 endpoints by call count for each day.

The data engineering team needs to count commits mentioning 'fix' in repos with 'data' in the name (case-insensitive), e

The messaging analytics team needs all records from days when the number of distinct message recipients exceeded the num

ROW_NUMBER dedup keeping latest row per key.

Deduplicate transactions, then sum sales per category.

HR is building a headcount and compensation dashboard.

Compute cumulative metric values within each department using window operations.

The platform team wants to score deployment authors over the past 12 months using the formula (100 minus average duratio

The reliability team wants to measure failure impact by service.

The DevOps team needs a quick environment breakdown showing how many deployments went to each environment.

Engineering is running a deploy reliability competition among teams.

Product analytics needs a breakdown of device diversity among younger users.

The growth team wants to know which device serves the most unique users.

The platform team needs to calculate the ratio of disabled feature flags among all flags that were updated on January 10

The messaging analytics team needs total unique conversations in chat_msgs.

The data quality team is benchmarking check pass rates across tables.

The QA team wants to measure the spread in data quality scores.

The data quality team suspects duplicate check entries.

The stream ops team needs a duplicated report of messages from the 'alerts' topic.

The ML platform team is flagging duplicate model training runs.

The onboarding team wants to track early activation.

The SRE team is benchmarking pipeline efficiency.

Self join to compare rows within the same table

The SRE team needs to understand latency spread across API endpoints.

The platform analytics team wants the endpoint with the most users who have exclusively used the GET method.

The content team needs total engagement broken down by content type, using duration in seconds as the engagement proxy.

Error rate per day and region via conditional aggregation.

The product team measures platform exclusivity.

The A/B testing platform needs a pivot report of conversion outcomes by variant.

The release management team needs to extract the numeric version from deploy log entries for the staging environment.

The platform team needs to flag outlier API tokens.

Find groups with the highest and lowest sums.

The SRE team flags incidents where resolution took more than 20 minutes beyond the predicted time.

The ops team wants to know how many data quality checks with a constraint-related rule resulted in a failure.

The data team wants to find the build date of the fastest CI build ever recorded, showing the date and the duration.

Rank by speed within daily partitions

The performance team benchmarks API endpoints by region.

The platform team is auditing feature flag adoption.

The ML platform team is profiling feature quality across data sources.

The product team runs a feature voting system where each user splits a single vote equally among all the flags they vote

Join three tables with filters and aggregate.

The analytics team wants to find the fifth highest cloud cost amount.

The ML team wants to know when peak accuracy was first and last recorded.

The reliability team needs the first and last times each service experienced a timeout.

The growth team measures first-touch attribution by service.

The data team wants to retrieve the first half of all page view records, based on the total row count, ordered by view I

Find each day's brand new users

Find each user's earliest channel interaction.

The messaging team wants to identify spammy senders.

Logistics wants the percentage of orders that can be fulfilled.

Find users with concentrated or broad ad exposure across campaigns.

Finance wants to spotlight the extremes in cloud spending.

Finance needs the customers with the highest daily spending between March 1 and June 1, 2026.

The infrastructure capacity team wants to find the top 3 regions by node density, defined as distinct nodes divided by d

The data engineering team needs to find which pipelines produce the highest throughput for runs started in 2026.

The experimentation platform team is debugging a test cohort.

The growth team needs to identify users who had zero sessions between June 1 and July 1, 2026.

The product team compares engagement between premium and free-tier users.

Product analytics is measuring mobile adoption across age buckets.

The growth team wants to measure mobile engagement by region.

The data pipeline team tracks job state transitions.

Deduplicate rows keeping the latest per key

The search quality team needs to find users whose search terms contain the words 'deploy', 'release', 'build', or 'ship'

The experimentation platform team wants to find the test with the most unique participants.

The FinOps team needs to find the single largest cloud cost entry, showing the service name and the amount.

The SRE team wants to compare server 'web-prod-07' against the 10th-fastest server by average response time.

The data pipeline team needs each author's most recent applied migration balance.

The ML platform team wants to identify the most popular training frameworks.

The product team needs to find the least-viewed content for potential deprecation.

The people ops team tracks hiring and attrition gaps.

The platform team wants to find the single data pipeline that ran the longest, returning just that pipeline's name.

The platform team is auditing long-running feature flags.

The cost optimization team is investigating teams that spend more on compute than storage.

The SRE team wants the lowest latency recorded for each service in the us-east-1 region.

The compliance team needs the total number of 'low' severity data quality checks that ran in 2026.

The stream processing team is looking for low-volume topics.

The revenue team needs March 2026 transaction totals per customer.

For the data quality report, the team needs the median null percentage across all ML features that have a data type of '

The people ops team is building mentorship pairs.

The QA team is exploring metric correlations.

The FinOps team needs the lowest monthly cloud bill for each provider.

The growth team wants to compare user engagement by device type in 2026.

The ML platform team wants to find models whose accuracy has varied across versions.

The ML platform team reviews model evaluation results.

Compute month over month retention rates for user signup cohorts.

The growth team is tracking month-over-month revenue changes per product.

Cumulative sales per product across months.

The FinOps team needs a pivot table of monthly cloud spend by provider.

The product analytics team needs a monthly engagement summary for 2026.

The on-call team wants to find which calendar month (across all years) had the fewest deployments.

The community team wants to recognize power users on the chat platform.

The code velocity team wants to find the most active committers over the past 2 years relative to the latest commit in t

Infrastructure wants to identify the busiest servers by log volume in 2026.

Product wants to surface the code review that received the most comments.

The A/B testing team wants to find the most common experiment outcome among experiments created on a Monday.

The SRE team is evaluating which API endpoint delivers the best throughput efficiency.

The platform team wants to know which error types occur most frequently.

The international expansion team needs to identify popular products in specific markets.

The product team wants to identify the most popular content type.

The product team wants to know which day of the week sees the most user signups.

Finance wants to find the single most profitable region-month combination in 2026.

The growth team is analyzing regional infrastructure density.

Join three tables into a summary report.

The social features team wants to find mutual connections between two specific users.

Calculate the ratio of problematic events for users within their first 14 days.

Count users whose first order falls on each date.

The infra team needs a headcount of nodes broken down by region and node type, sorted alphabetically by region.

The infra team is auditing node distribution across six key regions: us-east-1, us-west-2, eu-west-1, eu-central-1, ap-s

The data quality team wants to surface the noisiest tables.

The analytics team categorizes error messages by length: 'short' (under 50 chars), 'mid' (50 to 150), and 'long' (over 1

The messaging reliability team needs the delivery ratio: what fraction of all sent push notifications were actually open

The ops team tracks how often push notifications are opened.

The product team wants notification volume pivoted by platform and day of week.

Find the 3rd-ranked value within each partition.

The marketing team needs to know how many push notifications were successfully opened during January and February combin

The PMO team needs to flag over-budget data pipelines.

Self join to detect overlapping date ranges per user.

The capacity planning team considers a node overloaded if its CPU exceeds 90% or its memory exceeds 85%.

The engagement team wants to understand which session durations drive the most page views.

The analytics team needs all pairwise latency comparisons.

The capacity planning team needs to identify the peak traffic hour of the day.

The support analytics team wants to count how many distinct users made 3 or more API calls during the afternoon rush (15

The platform team is reviewing peak API load.

The revenue team needs a monthly breakdown for ad campaigns containing 'retarget' in 2026.

The onboarding team tracks how far users get through multi-step flows.

The data pipeline team compares processing overhead between production and staging.

The data platform team is benchmarking pipeline recovery times.

Reshape rows into columns by event type.

The capacity planning team is analyzing resource efficiency.

The growth team defines a power user as someone with more than 3 sessions and over 100 total pages viewed across all ses

The growth team wants to find power users who have logged more than 3 sessions.

The product team wants to surface the highest-priced item per category for a pricing review.

The marketplace team wants to identify high-performing budget products.

The people analytics team wants to identify senior leads in the largest service teams.

The FinOps team wants to compare regional cloud spend between March 11, 2026 and April 11, 2026.

The growth team tracks push notification engagement.

The notification analytics team needs a pivot showing push notification counts by status for each user.

The push notification team is measuring opt-in engagement.

The finance team needs to consolidate international cloud costs into a single currency, using regional conversion factor

The platform team wants to rank users by total search queries.

Detect retried API calls within 5 minutes of failure.

The rate limiting team wants to document threshold rules per endpoint.

Find items with a price decrease or sale event recently.

The regional ops team needs order metrics by region: the number of orders, the number of customers, and the total profit

The ops team wants to rank regions by incident volume.

The reliability team needs the single region with the highest average service uptime.

The infrastructure team wants to know which region hosts the most nodes.

The retention team wants to find power users who made at least 3 transactions in both 2025 and 2026.

Detect same amount transactions within 10 minutes.

Find users with orders less than 7 days apart.

The retention squad is looking for repeat purchasers.

The data engineering team wants to rank repositories by total lines added across all commits.

The CI/CD team wants to compare build volume to deploy volume.

The retention team wants to identify returning buyers.

Finance is calculating total revenue per product, including products that have never sold.

The code review analytics team wants performance metrics for each reviewer who has reviewed at least one merged pull req

The code review team wants to know which repos had the most reviewers each year.

The trust and safety team counts blocked API tokens by scope for December 2026.

Rolling 7 day sum per user via window frame.

The data engineering team is cleaning err_tracks.

The cost team wants to identify the second highest cloud cost amount on record.

The capacity planning team wants to analyze rush-hour API behavior in the 'us-east' region.

Percentage of transactions on the signup date.

The messaging platform team wants to find users whose first and last message in a channel on the same day were sent to t

The product team is benchmarking user satisfaction across platforms.

The compensation team wants the second highest distinct cloud cost amount on record.

Engineering leadership wants the second-highest latency API endpoint in each HTTP method group.

HR analytics is studying workforce age distribution per department.

The infra team tracks server migrations between regions.

The SRE team needs to find the single server with the most logged errors.

The PMO wants to know budget-per-head for each data pipeline.

The observability team is classifying services by naming pattern.

Per endpoint stats: count, avg latency, high latency count.

The product team wants to classify services by reliability tier based on uptime: 99.

The SRE team needs to identify services performing exactly at the median uptime.

Calculate uptime from status change events.