Data & Analysis
SQL, spreadsheet formulas, dashboards, summaries.
Prompts that translate fuzzy questions into runnable queries: SQL drafting, spreadsheet-formula generation, dashboard scoping, exec-summary distillation.
Generate a Google Sheets / Excel formula from a description
Spreadsheet formulas have many ways to be wrong. This prompt produces the formula plus a verification step.
Write a one-paragraph exec summary from a data table
Most data summaries are 'here's what the chart shows.' This prompt produces an exec summary that names what changed, why it might matter, and what to do.
Scope a dashboard from a stakeholder ask
Stakeholders ask for 'a dashboard'. This prompt extracts what they actually need, so you build it once.
Generate plausible explanations for an unexpected metric movement
When a number jumps unexpectedly, the answer matters less than the structured search. This prompt enumerates causes by category so you can rule them out fast.
Draft a SQL query from a plain-English question
Translating a business question into SQL is where most non-engineers get stuck. This prompt turns a natural-language ask into a ready-to-run query with inline comments explaining each clause.
Interpret a cohort retention table and surface key insights
Cohort retention tables are easy to generate but hard to read quickly. This prompt extracts the most important signals and frames them as actionable observations for a product or growth team.
Write step-by-step data cleaning instructions for a messy spreadsheet
Raw exports from CRMs, survey tools, or ERPs are almost always dirty. This prompt produces a concrete, ordered cleaning checklist tailored to the specific problems in your dataset.
Translate A/B test results into a plain-English decision memo
Stat-sig outputs from an A/B test rarely translate directly into a shipping decision. This prompt converts raw test results into a clear recommendation that includes confidence level, practical significance, and the right caveats.
Build a KPI tree from a top-level business goal
Before you build a dashboard, you need to know which metrics actually connect to the goal you care about. This prompt constructs a structured KPI tree that shows drivers, lag and lead indicators, and data-source requirements.
Design a pivot table structure from an analysis question
Most people open a pivot table tool before they know what they're trying to answer, which leads to trial-and-error. This prompt works backwards from your question to specify exactly which fields go in rows, columns, values, and filters.
Convert a vague data request into a precise analyst ticket
Vague requests like 'can you pull something on churn?' waste analyst time because they bounce back for clarification. This prompt turns an ambiguous ask into a scoped, specific ticket an analyst can act on immediately.
Stress-test the assumptions behind a numeric forecast
Most forecasts fail not because the math is wrong but because one or two key assumptions are never questioned. This prompt systematically surfaces those assumptions and shows what happens when they break.
Identify and explain statistical outliers in a time series
When a number spikes or drops in a report, stakeholders immediately ask why. This prompt helps you quickly flag which data points are genuine outliers and draft candidate explanations before your next stand-up.
Draft a multi-step funnel SQL query from event data
Funnel queries are notoriously tricky to write from scratch—ordering events per user, handling repeated steps, and choosing the right window all introduce bugs. Use this prompt to get a solid draft with logic you can review and adapt.
Build a weighted scoring matrix to rank competing options
When a team debates options with gut feel, decisions stall or get made by the loudest voice. A weighted scoring matrix makes the criteria explicit and the trade-offs visible—this prompt builds one you can drop straight into a spreadsheet.
Write and explain a regex pattern for extracting structured data
Regex is fast once written but time-consuming to construct correctly from scratch, especially for messy real-world strings. This prompt produces a working pattern, explains each component, and warns you about common failure cases.
Generate a plain-English data dictionary from a table schema
Undocumented tables slow down every analyst who touches them. This prompt turns a raw schema dump into a readable data dictionary that teams can paste into a wiki or Notion page immediately.
Quantify and rank themes from open-ended survey responses
Open-ended survey responses are rich but hard to summarize for leadership, who want numbers. This prompt structures the synthesis so you can report 'X% of respondents mentioned Y' rather than offering vague impressions.
Set defensible alerting thresholds for a business metric
Alert fatigue kills monitoring programs—too many false positives and the team ignores pings, too few and real problems go unnoticed. This prompt helps you derive thresholds grounded in actual data behavior rather than arbitrary round numbers.
Reconstruct the calculation logic behind a reported number
Inherited dashboards often show metrics with no documentation of how they are calculated. Before you can trust or replicate a number, you need to reverse-engineer the logic—this prompt walks through that systematically.