Deck vs Data Audit
Every numerical claim from the submitted FoodHub deck, recomputed against the bundled CSV.
Are the numbers in the submitted presentation accurate?
Demonstrates analytical rigor by validating reported figures against the source data before submission.
| Claim | Deck | Actual | Status |
|---|
Cuisine count. The dataset contains 14 cuisines, not 6 — update Slide 6 and the Data Overview narrative.
Average rating. The mean of rated orders is ≈ 4.34 / 5, not 3.4. The 70.7% "4 or 5 star" figure should also be ~84%.
Delivery-time → rating bullet. The claim that orders >45 min average 3.2 and <30 min average 4.5 is not supported by the data — the actual average rating for >45 min orders is ≈ 4.34, and no orders have a total time under 30 minutes (minimum prep + delivery is 35).
Weekend spend uplift. The real weekend-vs-weekday average-cost gap is roughly +1.6%, not +14%.
Delivery time std dev. Actual std is ≈ 4.97 minutes; the deck reports 6.8.
Premium $25+ share. Using "cost > $25" gives ≈ 16%, not 9.4%. If the deck intended a different threshold, state it explicitly.