Dick's Sporting Goods Campaign Performance Dashboard

The brief

Dick’s Sporting Goods analytics and media planning teams and agency stakeholders across strategy, investment, and audience and platform services need real-time access to omnichannel media performance. We’re talking data and optimizations spanning CM360, Programmatic DSPs, Social APIs, direct partner site-served data, search, etc. Without a central performance hub, data is fragmented, discrepancies slow decisions, and insights are blocked by noise.

The brief was deceptively simple: one place to do the work — for a $300M+ media program across five-plus channels and ten-plus data sources.

The approach

I’ve owned the Tableau dashboard end-to-end. Discovery with strategy, investment, and client stakeholders; requirements writing; sprint planning with engineering on the Redshift pipelines; managing the feature/fix backlog; UAT; and the post-launch roadmap.

Key moves:

• Tiered reporting framework — the dashboard supports pacing, pulse, and wrap views with defined scope, SLAs, and quality bars. Data storytelling operationalized at scale.
• Unified data model — Leverage Publicis datalake joining Meta, TikTok, Snapchat, Pinterest, DCM, Reddit, programmatic, third-party verification, and lagging outcomes in Redshift via daily automated ETL.
• Benchmarks + scatter logic — primary efficiency and performance KPIs coexist and add context. eCPM vs. CTR scatter bubble-sized by spend, color-coded over/under-performance vs. channel benchmarks.
• Agile delivery with a lean four-person team — sprint planning, requirements tickets, cross-functional facilitation; teams stayed engaged and the product keeps shipping.
• AI as a force multiplier — ChatGPT, Claude and Copilot assist on mapping, spec drafting, requirements synthesis, taxonomy fixes, and prototyping. Privacy-safe handling of client data was non-negotiable.

The outcome

The dashboard largely replaced platform-hopping with a single unified view used by exec, strategy, and analyst audiences. The AI-assisted workflow added 1–2 FTE of analytical and data-engineering capacity without growing headcount — capacity that got reinvested in benchmark automation, modeling, and cross-channel reporting that previously stalled in analysis paralysis.