Independent public data project · Python / Pandas
When does a restaurant check look different?
A reproducible analysis of the public Plotly restaurant bills sample. The question is how lunch and dinner differ by bill size and bill per cover, and what more an operator needs before changing staffing or offers.
Descriptive differences can suggest what to investigate; they cannot establish profitability or justify a rota change.
Explore the evidence ↓01 · Business question
Which service period differences warrant a closer look?
With limited resources, a restaurant would want to understand how the value of a typical check and party size vary across service periods. This dataset supports an exploratory comparison, not an actual staffing recommendation.
02 · Runnable method
Validate, derive, compare and export.
I used Plotly’s public educational tips dataset; the Python script checks its 244 records for completeness, expected columns and plausible values, calculates bill per cover and tip rate, and exports period level results and a chart.
Open the runnable code and reproduction steps ↗03 · Evidence
A descriptive comparison with the sample sizes visible.
Lunch includes 68 bills, with a mean bill of 17.17; dinner includes 176 bills, with a mean bill of 20.80. These results are descriptive and rounded from the script outputs.
Source: Plotly’s public 244 bill restaurant example, not transactions collected by Daniel. Currency, margin and business outcomes are not established.
04 · Decision implications
What an operator would need next.
Before adjusting staffing or promotional spend, measure representative hourly orders, cover counts, contribution margin and labour costs over multiple weeks. The sample is small and unbalanced across service periods; the data have no known timestamps, staff allocation or profit information.
Back to selected work