The KPI looked simple on the slide: full-price sell-through by region. Underneath was a treaty between finance, merchandising, and data engineering.
Lineage workshops surfaced duplicate SKU mappings, returns counted twice, and outlet sales leaking into full-price views.
The team published the definition, the exclusions, and the refresh cadence in plain language. Executives stopped arguing about the number and started arguing about the business — which was the point.
Clean KPIs are editorial choices. Someone has to decide what the story is allowed to say.
The KPI looked harmless: full-price sell-through by region.
Then the questions started. Should outlet sales be excluded? What about online returns posted after the week closed? Should wholesale be in the same view? Were cancelled orders counted before or after allocation? Which region owned cross-border e-commerce orders? Every answer changed the number.
The workshop felt less like a technical meeting and more like a negotiation. Finance wanted consistency. Merchandising wanted usefulness. Regional teams wanted fairness. Data engineering wanted rules that could survive automation. The number on the slide was simple because all the complexity had been hidden underneath it.
One duplicate SKU mapping explained part of the gap. A returns timing issue explained another. A legacy channel code carried old business logic nobody had questioned for years. None of this was glamorous. But every fix removed one reason for executives to argue about the number instead of the business.
The team eventually published the KPI as a short definition: what it includes, what it excludes, how often it refreshes, who owns it, and when it should not be used. That last part mattered. A clean KPI also knows its limits.
The lesson is that a KPI is not only a calculation. It is a social contract. When people agree on the number, they are agreeing on what kind of story the business is allowed to tell about itself.
