Grouping all 50,000 users by occupation and computing mean productivity scores reveals a clear but compressed ranking. Professionals lead at 5.53, followed by Freelancers (5.51), Business Owners (5.50), and Students (5.48). The rank order is intuitive, structured, accountability-driven work environments correlate with marginally higher scores, while Students, operating with more flexible and self-directed schedules, sit at the bottom.

However, the total spread across all four occupations is just 0.05 points, a gap so narrow it is unlikely to carry practical significance. The y-axis was appropriately constrained (5.45–5.54) to make the differences visible without visual exaggeration, which reflects good visualisation discipline. Across 50,000 users, this near-uniformity suggests that occupation type alone explains very little of the variation in how productively people work.

Key insight: Professionals are marginally the most productive group, but a 0.05-point range across all four occupations confirms that occupation is a weak standalone predictor of work productivity.

A Pearson correlation matrix across all nine numerical variables, including Daily_Phone_Hours, Social_Media_Hours, Sleep_Hours, Stress_Level, App_Usage_Count, Caffeine_Intake_Cups, Age, and Weekend_Screen_Time_Hours, produced the most definitive finding of the entire analysis.

Every correlation between Work_Productivity_Score and the lifestyle variables is effectively zero, with the maximum absolute value reaching just 0.01. This means that none of the measured behavioural variables; how many hours a person spends on their phone, how much they sleep, how stressed they are, how much caffeine they consume, or how many apps they use have a meaningful linear relationship with their work productivity score. The heatmap is dominated by deep blue across the productivity row, with no warm tones visible outside the diagonal.

This is a significant finding at scale. With 50,000 data points, even a weak true relationship would likely surface as a non-zero correlation. The near-total absence of correlation suggests either that productivity is genuinely independent of these variables in this population, or that the relationships are non-linear and would require more advanced modelling to detect.

Key insight: No numerical lifestyle variable, including phone usage, sleep, stress, or caffeine, shows meaningful correlation with work productivity. This challenges common assumptions about what drives productivity in smartphone users.

The pie chart breakdown of average productivity scores by gender shows a distribution that is essentially perfectly equal: Female and Other both at 33.4%, Male at 33.2%. The 0.2 percentage point difference between the highest and lowest gender groups is negligible, and the near-identical three-way split confirms that gender has no meaningful association with productivity in this dataset.

This finding is relevant beyond the dataset itself. It provides data-backed evidence against assumptions that productivity is gendered, at least within the context of smartphone-using populations represented here. The inclusion of a third gender category (‘Other’) also reflects thoughtful dataset design and inclusive analysis practice.

Key insight: Gender shows no association with work productivity. Male, Female, and Other groups contribute virtually equal shares, making gender a non-factor in productivity prediction.

Comparing productivity scores between Android and iOS users produces the largest gap in the entire analysis, and yet it remains modest. Android users average 5.53 versus iOS users at 5.47, a difference of 0.06 points. Android users are marginally more productive on this measure, making device type the variable with the strongest, though still limited, association with productivity across all four analyses.

Whether this difference reflects genuine platform effects, demographic differences between Android and iOS user bases, or simply sampling variation is a question the descriptive analysis cannot resolve. It is, however, the finding most worth investigating further,through subgroup analysis or statistical significance testing in subsequent work.

Key insight: Android users score 0.06 points higher than iOS users, the largest gap in the analysis, though still narrow. Device type is the most differentiated variable examined and the strongest candidate for further investigation.