The same exploratory analysis built twice, once in Python and once in Tableau, comparing what each tool makes easy when relating sugar content, price and popularity.
Explore what makes certain candies more popular: sugar against price, type-based patterns, top performers, and which ingredients move win percentage.
Architecture
toffee-data.csv -> Jupyter (pandas, seaborn) -> PNG and CSV outputs. Same data -> Tableau -> interactive workbook, dashboard PDF and deck.
What it does
Dual implementation: Jupyter notebook and Tableau workbook over one dataset
Scatter, pair, box and bar charts, a correlation heatmap, a bubble chart and an executive dashboard
CSV outputs for the analysis summary, correlation matrix and ingredient impact
Packaged Tableau workbook, dashboard PDF and presentation deck
Challenges & learnings
Keeping the Python and Tableau narratives aligned taught me to structure an EDA for two deliverables from the start rather than porting one to the other.
Designing an executive dashboard that answers the question in one view instead of showing every chart produced.
Visuals & outputs
Toffee data exploration project overview
Executive dashboard summarising the key findings
Pair plot of sugar, price, and win % relationships
