Today’s Marketing

Per seat vs usage based pricing

Per seat pricing charges for each person who uses the product, so the bill is easy to predict. Usage based pricing charges for what gets consumed, such as emails, API calls or compute, so the bill tracks value. Many AI and infrastructure products mix both: a seat or plan fee plus metered usage.

Updated · By Blake Emal · 6 real examples, each linked to its source

What are the steps

  1. Find your value metricThe unit that grows when customers get more value.
  2. Check who paysIf one person gets value for a whole team, seats undercharge.
  3. Make the bill predictableSupabase calls its pricing predictable and Vercel promises control over cost.
  4. Give a calculatorResend lets people slide to their own volume.
  5. Cap the scary caseSpend limits and alerts calm the fear of a surprise bill.

How do the options compare

Per seatUsage based
Bill is easy to predictYesHarder
Price tracks valueOnly if value scales with peopleYes
Fits AI and APIsPoorlyWell
Main buyer worryPaying for idle seatsA surprise bill
Real examplesLinear, AttioResend, Vercel, Supabase

Which real examples show it

Get three ideas for your own product

Pick a channel and a goal. Get real examples and a plan to fill in. Free, no login.

Open the idea generator

What else do people ask

What is hybrid pricing

A base plan or seat fee plus metered usage above an included amount.

Is usage based pricing better for AI products

Often, because the cost to serve grows with use. Pair it with clear limits so buyers can predict the bill.

How do I move from seats to usage

Protect current customers and explain what happens to their plan first, as Cursor did when it changed its plans.

Related answers

One email every Monday