Pika View Dashboard

FMCG teams were working with large, fragmented datasets across brands, SKUs and regions. The data comes from informal retail: the Pika mobile app, used by 6,000+ traders, captures it in the field.FMCG teams were working with large, fragmented datasets across brands, SKUs and regions. The data comes from informal retail: the Pika mobile app, used by 6,000+ traders, captures it in the field.

Existing tools made it difficult to explore that data confidently, compare performance, or extract actionable insights quickly. The questions executives brought were specific: market share, pricing shifts, distribution gaps and brand concentration.Existing tools made it difficult to explore that data confidently, compare performance, or extract actionable insights quickly. The questions executives brought were specific: market share, pricing shifts, distribution gaps and brand concentration.

The constraints were just as real: large datasets with performance costs, several user roles with different needs, data sold and accessed by region, and engineering limits on data loading and state management.The constraints were just as real: large datasets with performance costs, several user roles with different needs, data sold and accessed by region, and engineering limits on data loading and state management.

Every business question needed a different slice of the same data, so I designed one flexible filtering system across brands, SKUs and regions rather than a fixed report per question. Because data is sold and accessed by region, the filters also had to respect what each customer could see.Every business question needed a different slice of the same data, so I designed one flexible filtering system across brands, SKUs and regions rather than a fixed report per question. Because data is sold and accessed by region, the filters also had to respect what each customer could see.

Executives needed the answer before the evidence, so I prioritized progressive disclosure: summary metrics up front, comparisons and performance breakdowns one level down. The overview stays readable without hiding depth.Executives needed the answer before the evidence, so I prioritized progressive disclosure: summary metrics up front, comparisons and performance breakdowns one level down. The overview stays readable without hiding depth.

Usability testing with internal stakeholders and external users shaped both decisions, with each round aimed at clarity and trust in the data.Usability testing with internal stakeholders and external users shaped both decisions, with each round aimed at clarity and trust in the data.

This project reinforced the importance of designing systems, not screens, especially in data-heavy environments where clarity directly affects decision quality.This project reinforced the importance of designing systems, not screens, especially in data-heavy environments where clarity directly affects decision quality.