30 free prompts · all verified
Data & BI prompts
Prompts for the analyst layer above the query itself — turning a raw dataset or spreadsheet formula into a defensible finding, choosing the right chart and KPI for a dashboard, and writing the executive narrative that explains what the numbers actually mean without overclaiming causation.
Every prompt lists the exact tool and version it was tested against.
Get a trustworthy first read on a new dataset before you build anything on top of itTurns a raw data dump into a structured first-look summary — shape, likely grain, suspicious columns, and the three questions you should answer before trusting any number from it.ChatGPT (GPT-5.1)2026-08-08Turn a messy spreadsheet column into a defensible cleaning plan you can hand to someone elseProduces a step-by-step data cleaning plan for a specific messy column — with the exact transformation rule for each pattern of mess and a log of what changed, so the cleaning is reproducible instead of a one-off manual fix.ChatGPT (GPT-5.1)2026-08-08Get a SQL query for an analytics question you can actually explain and defend, not just paste and runConverts a plain-English analytics question into a SQL query written for your actual schema, plus a plain-English explanation of every join and filter, so you're never running a query you couldn't defend if someone asked how the number was computed.ChatGPT (GPT-5.1)2026-08-09Get an Excel formula that survives contact with your actual messy spreadsheet, not just the happy pathBuilds an Excel formula for a specific calculation and explicitly stress-tests it against blanks, text-in-number-columns, and merged cells before handing it back, so it doesn't break the moment it hits real data.ChatGPT (GPT-5.1)2026-08-09Write a Google Sheets formula that won't quietly break when three other people are editing the same sheetBuilds a Google Sheets formula for a specific calculation with explicit safeguards against the failure modes unique to a shared, live-edited sheet — inserted rows, someone typing over a reference cell, and stale cached values.ChatGPT (GPT-5.1)2026-08-10Write a Power BI DAX measure that gets filter context right, not just one that compilesBuilds a DAX measure for a specific requirement and explicitly checks it against row-context-versus-filter-context mistakes and slicer interactions, since a DAX measure that compiles cleanly can still return a silently wrong number.ChatGPT (GPT-5.1)2026-08-10Turn a Power BI report full of numbers into three insights an executive will actually act onReads a description of your Power BI report's visuals and numbers and produces a short, ranked insight narrative — what actually matters, what's noise, and what decision each insight should trigger.ChatGPT (GPT-5.1)2026-08-11Write the dashboard requirements doc that stops you from rebuilding it three timesInterviews the request behind a dashboard ask and produces a requirements doc — audience, decisions it needs to support, grain, and refresh cadence — pinned down before a single visual gets built.ChatGPT (GPT-5.1)2026-08-11Pick KPIs that are hard to game and actually track the thing your team is supposed to improveSelects and stress-tests a small KPI set for a specific team goal — checking each candidate for how easily it can be gamed or how it might drive the wrong behavior before it goes on a scorecard.ChatGPT (GPT-5.1)2026-08-12Pick the chart type that actually proves your point instead of just displaying your dataRecommends the right chart type for the specific claim a visualization needs to support, rejecting whichever default chart type would technically show the data but bury or misrepresent the actual point.ChatGPT (GPT-5.1)2026-08-12Find out if a trend in your numbers is real or just noise before you tell anyone about itAnalyzes a described time-series trend for whether it's a genuine signal worth acting on or normal noise/seasonality dressed up as a story, before it gets repeated in a meeting as fact.ChatGPT (GPT-5.1)2026-08-13Triage a data anomaly before you page anyone about itWorks through a spotted anomaly in a metric or dataset to separate a real operational problem from a data-pipeline glitch or reporting artifact, before escalating it to a team that will drop what they're doing.ChatGPT (GPT-5.1)2026-08-13Design a cohort retention analysis that answers the actual question instead of just producing a heatmapDesigns a cohort analysis structure matched to a specific retention question, then reads a described cohort result for the pattern that actually matters, rather than just describing the shape of the heatmap.ChatGPT (GPT-5.1)2026-08-14Diagnose why a funnel is leaking, not just draw the funnel chartTakes a described conversion funnel and its drop-off numbers and produces a ranked, specific diagnosis of the most likely cause at the worst step, instead of stopping at describing where the drop-off happens.ChatGPT (GPT-5.1)2026-08-14Build a customer segmentation that tells your team what to actually do differently for each groupDesigns a customer segmentation scheme tied to a specific business action, rejecting descriptive-only segments that are interesting to look at but don't change what anyone does for any given group.ChatGPT (GPT-5.1)2026-08-14Turn a revenue variance into a driver-by-driver breakdown instead of one vague explanationTakes a period-over-period revenue number and your raw line items and produces a ranked breakdown of what actually moved — price, volume, mix, or new/lost accounts — so you're not stuck writing "revenue was up due to strong performance."ChatGPT (GPT-5.1)2026-08-08Pressure-test a sales forecast against pipeline coverage before you present it, not after it missesFeeds your current pipeline and historical conversion rates through a coverage-ratio check so an optimistic forecast gets flagged before it goes to leadership, instead of the gap surfacing at quarter-end.ChatGPT (GPT-5.1)2026-08-09Get a straight verdict on an A/B test instead of a hedge that avoids saying whether it wonRuns your test results through a significance and practical-impact check and forces a plain ship/hold/kill call, instead of the noncommittal "trending positive" summary that leaves the actual decision to you.ChatGPT (GPT-5.1)2026-08-10Code hundreds of open-text survey responses into themes without losing the outlier that mattersTurns a pile of raw open-ended survey answers into a ranked theme list with verbatim quotes as evidence, while explicitly holding back rare-but-important comments from getting averaged away.ChatGPT (GPT-5.1)2026-08-08Build the narrative arc around a set of metrics before you build a single slideTakes a pile of disconnected numbers and finds the actual story connecting them — tension, turning point, resolution — so a reporting deck doesn't end up as a wall of charts with no throughline.ChatGPT (GPT-5.1)2026-08-11Compress a full analysis into a one-page brief an executive will actually read before the meetingDistills a longer analysis or dashboard into a single page built for a five-minute pre-read, with the decision needed stated up top instead of buried under methodology.ChatGPT (GPT-5.1)2026-08-12Generate a real data dictionary from a raw schema dump instead of column names with guessed descriptionsTurns a raw table schema or column list into a data dictionary with definitions, types, and known ambiguities flagged for confirmation, rather than confidently inventing what an unclear field means.ChatGPT (GPT-5.1)2026-08-09Draft a source-to-target ETL mapping that names every transformation instead of leaving it implicitProduces a field-by-field mapping between a source system and a target schema, including the exact transformation logic and the edge cases that would silently break a naive one-to-one copy.ChatGPT (GPT-5.1)2026-08-13Turn a messy CSV export into a precise cleanup spec before you touch the actual dataReads a sample of a messy CSV and produces an exact, ordered list of cleanup transformations to apply, so the fix is a repeatable spec you can hand to a script rather than a one-off manual scrub.ChatGPT (GPT-5.1)2026-08-10Reshape a nested JSON payload into a flat schema without losing fields that only appear sometimesDesigns the exact flattening logic for a nested JSON API response into a tabular schema, explicitly accounting for fields that are optional, inconsistently typed, or only present in some records.ChatGPT (GPT-5.1)2026-08-11Turn a description of business logic into a checkable list of data validation rulesConverts a plain-language description of how a business process should work into explicit, testable validation rules for a dataset, with each rule stated so a failing row can actually be identified.ChatGPT (GPT-5.1)2026-08-12Write a KPI definition precise enough that two teams calculate it the same wayProduces a full KPI spec — exact formula, inclusion/exclusion rules, edge cases, and calculation cadence — so the metric doesn't quietly mean something different in two different dashboards.ChatGPT (GPT-5.1)2026-08-13Trace a sudden metric drop back to its actual cause instead of the first correlated event you noticeWorks through a structured root-cause process for an unexplained metric drop, ruling out data artifacts and coincidental timing before committing to a causal story.ChatGPT (GPT-5.1)2026-08-14Build a what-if scenario model that shows its assumptions instead of one confident-looking numberStructures a scenario/what-if analysis around explicit, adjustable assumptions and a spread of outcomes, so the output is a decision tool rather than a single fabricated-precision forecast.ChatGPT (GPT-5.1)2026-08-14Audit a dashboard for the specific mistakes that make numbers look right until someone checksRuns a structured QA pass over a dashboard's metric definitions, filters, and calculations, targeting the exact class of silent errors that pass a casual glance but break under a real audit.ChatGPT (GPT-5.1)2026-08-14
