AI Pulse · Business Models & Winners

AI pricing is all over the map right now, and that’s worth noticing if you’re buying or building anything in this space. The same industry podcast that championed usage-based billing one week turned around and made the case for per-seat subscriptions the next. No one has figured out the “right” answer yet.

Usage-based pricing sounds logical. You pay for what you use, costs scale with value, everyone’s happy. Except when the bill swings wildly month to month and finance teams lose their minds trying to forecast. Per-seat feels safer and predictable, but it penalizes teams that want to experiment or run agents that don’t map neatly to headcount. Both models have real problems, and the market hasn’t settled on which trade-off matters less.

This unsettled state is actually useful information when you’re evaluating AI tools. If a vendor insists their pricing model is the only sensible one, they’re either guessing or selling. The honest answer right now is that no one knows. What matters more is whether the pricing structure aligns with how your business actually works. Can you predict the cost? Does it reward the behaviour you want, or does it create weird incentives? Will you get a surprise bill that derails a project?

When we scope the Omni Command Centre as a retainer, it’s not because we’ve cracked some secret code. It’s because fixed monthly costs let you plan, test, and iterate without watching a meter. That might not be the right model forever, but it’s the right model for teams that need to move without second-guessing every API call. The broader lesson is simpler: if someone tells you they’ve figured out AI pricing, they haven’t. Ask how it works for your specific situation, then decide if the trade-offs make sense.