ChatGPT Team Plan: Seats, Billing, and Shared Work

The ChatGPT Team plan is the former name of ChatGPT Business, a workspace subscription for organizations with shared administration and billing. It provides a place for colleagues to use ChatGPT, but membership does not automatically make everyone’s private conversations visible to the team. OpenAI’s Business FAQ↗ (opens in a new tab) explains the name and plan structure.

The buying question and the working question are different: who needs access, and how will everyone know what was agreed? Tools like fenbs keep projects, decisions, rules, and progress outside any one AI subscription, so a team can maintain a shared work record while choosing the AI apps each person needs.

Is ChatGPT Team still the plan name?

Look for ChatGPT Business in current settings and purchase information. OpenAI says existing Team workspaces became Business workspaces; the rename did not require a new workspace or a data migration. An old internal purchasing checklist may still say Team even when your account uses the Business label.

Before purchasing anything, ask whether your organization already has a workspace and who owns it. A colleague who cannot find it may need an invitation or the right account, rather than another subscription. Check the exact workspace name with its owner so similarly named organizations do not cause confusion.

For the layout and everyday navigation inside that environment, use the separate ChatGPT team workspace guide. Here, the focus is what the subscription means for buying access and maintaining shared work.

How does the subscription charge for a team?

Business billing uses paid seats, with monthly or annual billing options. A seat is purchased access capacity that can be assigned to a person. Check the current seat options, minimum commitment, and renewal details in OpenAI’s billing and seats guide↗ (opens in a new tab) rather than using an old cost estimate.

Removing a member and reducing the paid seat count are separate actions. OpenAI says a removed member’s seat becomes available for reassignment; reducing paid capacity must be scheduled separately. Added seats and seat reductions also take effect on different billing schedules, so the owner should review the confirmation before changing capacity.

• List the people who need ongoing access and the work each will do.
• Separate the initial commitment from the process for adding people later.
• Choose a person responsible for reviewing renewals and unused capacity.
• Record who may approve a subscription change before someone makes it.

Keep a proposed purchase separate from an approved one. For example, a studio can agree to evaluate a workspace without authorizing every requested seat change. A short written approval with its owner and review point makes the purchasing boundary understandable to the people doing the work.

What does a seat let a teammate do?

A seat and a role answer different questions. Seat type determines product access and included usage; a role determines what someone can manage. OpenAI’s member, seat, and role documentation↗ (opens in a new tab) describes these separately, including owner control over paid seats.

Test the ordinary member experience instead of relying on the owner’s screen. Ask a teammate to open the intended source and find the current task. An owner seeing a setting does not prove that every member can use it, and a successful invitation does not prove that onboarding is complete.

Does paying together make all work shared?

No. OpenAI explains that Business users have their own chat histories and other members do not automatically see private conversations. Shared conversation links provide access to selected chats within the workspace; continuing a shared chat creates a separate private conversation. See the Business sharing and privacy guide↗ (opens in a new tab).

That distinction matters even when everyone works on the same assignment. Say one teammate explores two proposal options and a manager approves one. Sharing an exploratory conversation is useful background, but the team still needs a clear statement of the approved choice and its consequences.

A shared resource can carry context without becoming the authoritative record of every agreement. Decide which result people should rely on, who checks it, and how changes reach the next person. The subscription provides access; your working routine makes that access useful.

What should you check before choosing it?

Start with a real task rather than a list of attractive features. Choose a small piece of work with known sources and a person who can judge the result. Ask whether the workspace helps colleagues complete it together, including the review and handoff, rather than only generating a convincing first draft.

If your decision includes Anthropic’s offering, the Claude Team plan guide covers that separate subscription. For differences in how the assistants support everyday work, read Claude versus ChatGPT. Neither comparison replaces testing the task your own team needs to finish.

Also ask what stays available when a person leaves, a project changes owner, or someone uses another AI app. Name the project record people should consult in each case. This is a continuity check you can perform before committing the team to a new way of working.

Where should decisions live across AI apps?

Keep each approved decision with the project it affects. Include what a person chose, why, which work it changes, and what still needs review. A rule that continues to apply should be recognizable as a rule, while an unanswered suggestion should remain an open question.

For the proposal example, save the approved direction and the remaining fact check. The next assistant needs that concise record more than a transcript of every rejected draft. fenbs puts Decisions and rules beside current tasks and progress; how it works explains the shared context connected assistants read.

How to do this with fenbs

1. Create a free account. One person can start free, with no credit card; use a team kanban board when colleagues need shared access.
2. Add one project and its current tasks. Write the goal, the source references, and the next piece of work that needs a person’s review.
3. Record approved choices in Decisions and rules. Keep suggestions open until a person decides, and give continuing rules a clear scope.
4. Connect a supported assistant using the relevant guide on Integrations. MCP, a standard way for AI apps to use outside tools, lets it read the work through an approved connection.
5. Ask the assistant to read project context before starting and save Where we left off before stopping. Include completed work, checks, and the next step.

fenbs does not manage your ChatGPT subscription or import its private chats. It is not the right fit if your only need is paying for ChatGPT seats. Use it when the project record needs to remain useful across people, sessions, and supported AI apps.

How can you tell whether the setup is working?

Give a teammate a fresh session and ask three questions: what was approved, what is unfinished, and who reviews the next result? Compare the answers with the actual project record. Fix missing sources or unclear ownership before adding more material.

A useful setup lets people find the same agreement without pretending every conversation is shared. Keep access checks and record upkeep part of finishing work. The team brain library covers other parts of that ongoing routine.