A 14-station network across five Nigerian states came to us with strong audiences and almost no monetization technology. Programmatic fill sat at roughly 12%, and most of what did clear was sold manually, station by station, on relationships rather than systems. Twenty months later, programmatic is their single largest revenue line.
Where it started
The network's ad ops process, before the rebuild, ran almost entirely on spreadsheets and phone calls. Each station negotiated its own rates. There was no shared reporting standard, no consistent way to verify a spot actually aired, and no programmatic demand at all — every deal was sold direct, one advertiser at a time.
What changed
- A single ad-serving layer across all 14 stations. Instead of each station running its own ad insertion (or none at all), the whole network moved onto shared infrastructure with consistent tagging and verification.
- Programmatic demand switched on gradually. Rather than flipping every station to open programmatic at once, the network opened inventory market by market, watching fill rates and yield before expanding.
- Direct sales stayed local, but got tooling. The existing direct sales relationships didn't go away — they got a proper CRM and inventory calendar, so direct and programmatic demand stopped colliding on the same spots.
- Monthly reconciliation became automatic. What used to take a finance team days each month — matching what aired against what was invoiced — became a same-day reconciliation.
The result
Fill rates moved from 12% to 82% over 20 months. More importantly, the network now has four distinct revenue streams instead of one, and direct sales grew alongside programmatic rather than being cannibalized by it. The lesson that generalizes: the constraint usually isn't demand. It's whether the inventory is structured in a way that demand can actually reach it.