Tag
Raw numbers don't explain themselves. These prompts help you turn a dataset into patterns, patterns into findings, and findings into a summary someone outside the spreadsheet can actually use.
22 prompts · page 1 of 3
Breaks a funnel of four named stages and their visitor counts into per-stage conversion rates and drop-off points, compares them against industry benchmarks, and identifies the single bottleneck holding back the goal metric. Closes with optimization ideas split into quick wins and longer-term fixes, each with an expected-impact estimate, so growth and product teams know exactly which stage to fix first.
Reads a two-variant test — descriptions, results, sample size, and duration — and delivers a fast verdict: statistical significance, which variant won, confidence level, key learnings, and a rollout recommendation with suggested follow-up tests. A lighter-weight companion to the rigorous version, meant for everyday test readouts rather than high-stakes ship decisions.
Feeds churn data, customer base size, churn rate, and revenue impact into an analysis of what's actually driving customers to leave. Surfaces churn triggers, the segments most at risk, retention patterns, and lifetime-value impact, then turns those into prevention strategies, win-back campaign ideas, and the success metrics to track — useful for retention teams building a response plan, not just a diagnosis.
Goes past a raw correlation matrix to flag the relationships in a dataset that actually matter before modeling starts. - Inputs: key variables of interest plus dataset columns, types, and row count - Analysis: correlation matrix, non-obvious pairings, multicollinearity risk, categorical-to-numeric associations, and time-lagged correlations - Warns: spurious correlations that look causal but aren't - Output: plain-language findings plus runnable Python code
Turns raw survey responses — respondent count, key questions, and demographic breakdown — into a findings summary right after a survey closes. Produces the top five findings, sentiment analysis, demographic differences, and unexpected insights, plus actionable recommendations, quote-worthy responses to pull into a deck, and suggested chart types for presenting the results.
Walks from a target variable, candidate predictors, dataset size, and analysis purpose through a full regression build: exploratory relationships, feature selection, model choice with reasoning, assumption checks, and performance metrics like R², RMSE, or AUC depending on the target type. Delivers plain-language coefficient interpretation, residual analysis, and sample predictions with confidence intervals, plus complete scikit-learn or statsmodels code.
Forecasts a chosen metric over a set horizon from historical data, granularity, known patterns, and external factors, with a justified choice of model — moving average, exponential smoothing, ARIMA/SARIMA, or Prophet. Delivers 80% and 95% confidence intervals, a trend/seasonality/residual decomposition, holdout accuracy metrics like MAPE and RMSE, and optimistic, baseline, and pessimistic scenarios, so the forecast carries honest uncertainty bounds instead of a single line.
Calculates TAM, SAM, and SOM for a product, geography, and customer segment, showing the methodology, data sources, and explicit assumptions behind each figure rather than just handing over a number. Adds growth projections and a suggested penetration strategy, making it useful for pitch decks and planning docs where reviewers will challenge the assumptions before the market-size figures themselves.
Handles the type-conversion and encoding grunt work needed before a dataset is ready for analysis or modeling. - Input: a list of columns with their current and target types - Converts types with edge-case handling (stray text in numeric columns, mixed-type columns) - Encodes categoricals appropriately: ordinal, one-hot/target, or frequency encoding for high-cardinality fields - Output: Python code plus a before/after conversion report flagging failed conversions