Data Analysis
Prompts.

Data Exploration Fundamentals

Prompting for data analysis turns AI into an analytical partner, extracting trends, anomalies, and recommendations from complex datasets via structured prompts. Data analysts, BI professionals, and business leaders use these prompts to interpret CSVs, reports, and visualizations.

Analyze this dataset [UPLOAD CSV or PASTE SUMMARY]:

1. **Structure**: Rows/cols/types/missing values (%)
2. **Distributions**: Skewness/outliers per numeric col
3. **Correlations**: Top 5 pairs (strength/direction)
4. **Segments**: Natural groupings/clusters
5. **Quality Issues**: Priorities + fixes

Provide: Summary table + 3 priority questions to ask next

Exploratory Data Analysis (EDA) Prompts

EDA for [DATA DESCRIPTION]:

- Key stats: mean/median/mode/IQR per column
- Visual suggestions: 3 chart types
- First insights: Hypotheses to test
- Cleaning plan: Step-by-step
- Missing data strategy

Analyze time series: [PASTE DATA or "Monthly sales 2020-2025"]

1. Trends: Growth/seasonality/cycles
2. Anomalies: Dates + magnitudes
3. Forecasts: Next 6 periods (method/confidence)
4. Drivers: Likely correlations
5. Interventions: When/impact?
Provide: Line chart description + forecast table

Trend and Pattern Recognition

Find strongest relationships in [DATA]:

Create table: | Var1 | Var2 | Corr | P-value | Causal plausibility |

Analysis:
- Visualize top 3 as scatter + regression
- Business implications ranked
- Confounders to consider

Statistical Inference Prompts

Test hypothesis: "[HYPOTHESIS]" on [DATA]

1. Statistical test recommended
2. Null/alternative clearly stated
3. P-value + interpretation
4. Effect size/practical significance
5. Conclusion + confidence level

Analyze A/B test results:
Control: [METRIC=N, MEAN=VAL, SD=VAL]
Variant: [SAME]

Determine: Winner? CI? Power? Subgroups? Recommendations

Customer Segmentation Prompts

Cluster analysis for [DATA: customer metrics]:

1. Natural segments (3-5)
2. Profiles: Traits + size (%)
3. Value differences (LTV/churn)
4. Strategies per segment
5. Validation metrics (silhouette score)

Output: Segment comparison matrix

Actionable Insight Generation

From analysis [PASTE RESULTS], create executive brief:

1. The story (headline insight)
2. Key evidence (3 bullets)
3. Actions (prioritized table: Impact/Effort/Owner)
4. Risks/watch items
5. Next dataset to acquire

Quantify business impact:

Finding: [INSIGHT]
Metrics affected: Revenue/churn/LTV
Magnitude: $$ range
Leverage points: Experiments to run
ROI estimate: Year 1/2/3

Data Analysis FAQ