PromptVaultData & Research
prompts.
50 data & research prompts for Perplexity, Gemini and DeepSeek, written and edited in-house. 30 of them are free. Each one leaves blanks only for what it cannot know — fill them in, copy, paste.
Hidden denominator shifts inside a dashboard exportFor dashboard readers who need a quick judgment on whether changing totals distort the reported trend.- Does this chart support the story attached to it?For beginners checking a visual claim through scale, encoding, and omitted context.
- A source pack with provenance checks already attachedFor researchers collecting current evidence that can survive citation and origin checks.
- Data-sharing requests arrive completeFor data custodians turning a request brief into a checklist that exposes missing safeguards.
- Reuse existing evidence or commission fresh research?For research leads making a defensible call when current evidence may be sufficient but imperfect.
- Weighted rates without the average-of-averages trapBeginners learn when percentages need denominator weights through one worked example and targeted drills.
- Where does this source packet genuinely agree?For analysts separating corroboration from repeated claims and shared dependencies.
- From field notes to an audit-ready observation recordField researchers receive a neutral, traceable record without blurred inference.
- Research questions that expose the decision behind a briefFor beginners who need answerable questions rather than a broad topic list.
- Survey responses checked for low-effort patternsFor survey owners judging whether inattentive responding could distort their findings.
- Bring an outdated research brief currentResearchers receive a rewritten brief whose dated claims have been verified.
- Scattered analysis notes, one reproducible handoffNew analysts turn fragments into a bounded record another person can rerun.
- A reproducibility run ordered by dependencyFor analysts converting a paper and repository notes into a schedulable rerun.
- Rescue an irrelevant literature searchResearchers get a ranked diagnosis before changing a weak database query.
- Response-rate collapse, traced before relaunchSurvey teams get a bounded diagnosis before changing incentives or outreach.
- Reproducible two-arm allocation from a fixed seedFor beginners needing a balanced assignment sequence that can be recreated exactly.
- Computational appendix, rerunnable by a second analystPractitioners turn private execution logs into a bounded reproducibility artefact.
- A survey launch plan built through four answersFor first-time researchers scheduling a small survey without missing fieldwork gates.
- SQL logic scored against the result it claims to returnFor analysts checking query semantics, grain, null handling, and scale risks.
- Is this experiment claim supported by its design?For analysts checking whether an A/B result can bear the conclusion attached to it.
- Causal claims tested against identification assumptionsFor practitioners stress-testing whether an observational design identifies its stated estimand.
- Which dataset acquisition route deserves the budget?For research leads choosing among buy, build, partner, or reuse options after a bounded interview.
- Sample versus population without guessworkResearch beginners learn which group a claim can legitimately describe.
- Six current sources with their weaknesses exposedFor beginners who need a compact evidence pack, not an unfiltered link dump.
- How defensible are these qualitative findings?For researchers stress-testing thematic claims through a short evidence interview.
- Decision briefing grounded in current evidenceFor beginners who need a concise, cited research brief tied to a real choice.
- Missingness patterns classified without false certaintyFor analysts judging whether absent values threaten a stated analytical goal.
- Search strings that survive systematic-review scrutinyFor review teams needing reproducible database syntax rather than a loose list of keywords.
- Dataset readiness exposed before analysis beginsFor beginners checking whether a dataset can safely answer its intended question.
- SQL assertions that catch silent metric driftFor analysts turning schema and business rules into executable data checks.
- Confidential research notes become a reproducible methods noteFor analysts converting sensitive working notes locally without exposing raw material.
- SQL window functions, learned by dependencySQL users build row-context intuition through a compact, cumulative lesson.
- Why does the dashboard disagree with the source?For analysts diagnosing a metric mismatch before changing queries or business definitions.
- Statistical results become a publication-ready results passageFor analysts tightening reported findings without changing estimates or certainty.
- Can you resist the base-rate trap?Analysts learn to check prevalence before accepting an impressive statistic.
- A controlled release plan for a research datasetFor research teams coordinating disclosure review, documentation, validation, and release.
- Neutral survey invitations without demand cuesSurvey teams receive send-ready invitations without exposing the expected answer.
- Synthetic data that tests failure paths safelyFor analysts needing a realistic fixture with edge cases but no copied personal records.
- Raw interviews become a publication-safe quote bankResearchers using local models get reusable excerpts without exposing participant identity.
- Which survey version should go to field?For researchers choosing between two instruments against a real decision context.
- Which statistical model best matches this decision?For analysts choosing among common models without hiding assumption costs.
- A fixed route from question to evidenceFor first-time researchers who need a dated checklist with assumptions exposed.
- Replication work sequenced to a fixed deadlineFor research teams turning a study brief into owned, dependency-aware replication work.
- Codebook readiness scored before dual codingFor qualitative teams checking whether two coders can apply the same rules independently.
- Confidence intervals under pressureWorking analysts uncover and correct their own interpretation errors through a short interview.
- Survey items scored for measurement riskFor researchers checking whether questions can support the construct they target.
- A fieldwork stop-rule card for consistent collectionFor first-time data collectors who need clear boundaries on when to pause or exclude.
- Has this research claim aged out?For analysts checking whether newer evidence has narrowed, overturned, or strengthened an older claim.
- Citation lists checked for correction and retraction riskFor analysts verifying whether a source pack still supports the claim attached to it.
- Sensitive interview text reveals re-identification riskFor researchers screening confidential qualitative material locally before controlled release.
Other categories.
Coding
Marketing
Business
Sales
Career
Productivity
Finance
Writing
Design
Art & Image
Video
Music
Social Media
Education
Language Learning
Study & Exams
Fitness
Cooking
Travel
Health & Wellness
Relationships
Parenting
Personal GrowthEvery data & research prompt, in your pocket
The whole library ships inside the app, so it works offline.
Coming to theApp Store
Scan with your iPhone