The quiet death of the seat-based licence
When software does the work instead of the person using it, charging per person stops describing the value delivered.
Per-seat pricing was a good proxy for value for twenty years. More employees using a tool generally meant more value extracted from it, and the metric was trivially easy to count and to audit.
Why the proxy broke
Software that completes tasks autonomously severs the link between headcount and consumption. A support team of five can resolve the volume that once required fifty. Under a seat model the vendor's revenue falls precisely as its product gets better, which is not a sustainable arrangement for either side.
The replacements, and their trade-offs
- Usage pricing aligns revenue with consumption but produces unpredictable bills, which procurement teams dislike more than they dislike high prices.
- Outcome pricing — per resolved ticket, per completed document — reads well in a sales meeting and requires an attribution model both sides trust.
- Hybrid platform fees with usage on top are becoming the default compromise: a predictable floor, elastic upside.
What breaks internally
Changing the pricing metric changes far more than a page on a website. Sales compensation, revenue forecasting, and the definition of net revenue retention all assume a stable unit. Companies that migrate successfully tend to run both models in parallel for at least a year and accept the reporting mess as the cost of the transition.
About the author
Senior Writer, Software & Startups
Daniel Okoye covers software business models, developer tooling and early-stage company building. He has spent a decade analysing SaaS pricing and go-to-market strategy.
