Example Outputs
Explore real sample reports generated from different datasets, including the insight summary, chart choices, refinements, and downloadable PDF output.
Jump to an example
Tap or click to scroll directly to that showcase.
Healthy Foods Database
Stacked horizontal bar chart revealing how Snacks & Sweets dominate calorie and fat loads while Grains and Vegetables lead on health score.
Meteor Strikes
Dual-panel chart showing how modern detection finds far more meteorites per decade — but at dramatically smaller median masses than a century ago.
Iris Flower Clusters
Strip plot with species rows and two dashed decision boundaries showing how petal length alone separates all three iris species.
These outputs were generated and refined directly in the app. Replicate them with your own file.
Example 1: Healthy Foods Database
DecisionHealth score, calorie, and fat trade-offs across 8 food types
Focus question
“Which food_type gives the strongest protein_g while keeping sodium_mg relatively low, and which types look risky if I want leaner high-protein options?”
The result
Stacked horizontal bar chart comparing average health score, calories, and fat (×10) per food type — making Snacks & Sweets' calorie and fat dominance immediately visible against leaner alternatives.
Input CSV
Healthy foods database with nutritional attributes (protein, sodium, calories, fat, health score) across 8 food type categories.
Data adjustments (before generating)
- Grouped individual food items by food_type and computed mean values for health score, calories, and fat.
- Scaled fat values by ×10 to make them visible alongside calorie bars on the same axis.
- Sorted food types by descending total bar length to surface the highest-load categories first.
Final takeaway
Snacks & Sweets carry roughly 6× the fat load of Fruits and Vegetables despite similar health scores. Seafood emerges as the strongest lean, high-protein option with the lowest combined calorie and fat footprint.
Preview + Download
Health score, calorie, and fat trade-offs across 8 food types
Example 2: Meteor Strikes
Deep DiveDiscovery counts and median mass by decade, 1800–2009
Focus question
“Where do the heaviest meteorites appear geographically, and how much of that pattern reflects real impact distribution versus discovery and preservation bias?”
The result
Dual-panel chart — bar chart of discovery counts per decade on top, median mass trend line (log scale) on bottom — revealing the inverse relationship between detection volume and specimen size over two centuries.
Input CSV
200 meteorite records spanning 1576–2009, with mass ranging from 148 kg to 60,000 tonnes and discovery status (Fell vs. Found).
Data adjustments (before generating)
- Binned records into decades for both panels to smooth year-level noise.
- Applied a log scale to the mass axis to handle the 9-order-of-magnitude range between specimens.
- Checked Fell vs. Found status for correlation with mass — found that the largest specimens are predominantly Fell records, confirming observational bias.
Final takeaway
Median meteorite mass dropped roughly 94% since 1900. Modern surveying and Antarctic programs uncover 3× more specimens per decade, but the five largest meteorites were all discovered before 1920 — the era of landmark finds appears largely over.
Preview + Download
Discovery counts and median mass by decade, 1800–2009
Example 3: Iris Flower Clusters
StoryPetal length distribution across three iris species
Focus question
“Why does Iris-setosa look like it comes from a different world compared with versicolor and virginica?”
The result
Strip plot with jittered species rows and two red dashed vertical boundary lines at ~2 cm and ~4.9 cm, illustrating near-complete petal length separation between all three iris species.
Input CSV
Classic Fisher Iris dataset — 150 flowers across 3 species (50 each) with petal length, petal width, sepal length, sepal width, and species fields.
Data adjustments (before generating)
- Selected petal length as the sole x-axis variable after confirming it yields the clearest species separation.
- Added random jitter on the y-axis to prevent point overlap within each species row.
- Identified and marked two key decision boundaries: 2.0 cm (setosa cutoff) and 4.9 cm (virginica cutoff).
Final takeaway
Setosa sits in a fully isolated zone below 2 cm with zero overlap — a single measurement classifies it perfectly. Versicolor and virginica share only a small 4.5–5.0 cm overlap zone where a second feature like petal width is needed for reliable disambiguation.
Preview + Download
Petal length distribution across three iris species
Ready to turn your CSV into insights?
Start your free trial to upload your own file and export presentation-ready outputs.