Claude Enterprise and Team seats are sold by the count, often with a minimum, and the count rarely matches who logs in. We measure real usage, right size the seat plan, negotiate the minimum down, and put true up protections in the contract so a growing headcount does not become a surprise invoice.
When an organization buys Claude Enterprise or Claude for Work, the seat number is usually picked early, before anyone knows who will use the product day to day. Procurement rounds up to be safe, a minimum is attached to unlock the plan, and the count gets locked into the term. Six months in, a large share of those seats have never been activated or are used once a month. The invoice does not care. You are paying for provisioned seats, not active ones.
Seat optimization is the unglamorous work of matching the seat plan to reality. It is where a procurement leader and an admin together can take real money out of the bill without touching a line of code, and it is one of the first places we look in any engagement.
We pull activation and active use data so the conversation starts from who logs in, not who was provisioned.
We set the seat number to active users plus a defensible buffer, and we cut the dead seats out of the renewal.
Seat minimums are a starting position, not a law. We negotiate them down to a level your real adoption supports.
We replace automatic mid term true ups with terms that let you add seats at the agreed rate without a penalty reset.
We check whether Enterprise, Team, or a mix fits each group, so you are not paying Enterprise rates for users who need Team features.
We phase the seat ramp to match hiring, so you commit to the seats you will fill, not the org chart you imagine.
Three leaks show up again and again. The first is the dead seat, provisioned and paid for but never activated. The second is the mid term true up, where adding people during the term triggers a recount at a worse rate than your original deal. The third is the tier mismatch, where a whole population sits on the most expensive plan because it was simpler to buy one plan for everyone.
The fixes are commercial, not technical. You measure activation honestly, you negotiate the minimum to your real floor, and you write the true up terms so growth is rewarded rather than penalized. The difference between a seat plan negotiated from data and one negotiated from a spreadsheet guess is often a double digit percentage of the seat line. To understand which plan each group actually needs, start with our comparison of Claude Enterprise vs Team.
Anthropic often bundles seats and API commitment into a single enterprise agreement. That is leverage if you use it. We negotiate the seat plan and the API commit together, so a concession on one side strengthens your position on the other, rather than letting the vendor optimize each line in isolation against you.
"We were paying for 400 seats and 230 people had ever logged in. They right sized the count, took the minimum down, and the renewal came in well below what we expected."
Fixed fee or gainshare. We sit between you and Anthropic and we never take vendor money.
Get a QuoteWeekly intelligence on Anthropic pricing moves and the buyer side counters that work.