Blog

The Claude Team plan, which plan to pick, and why your usage runs out

By Chad Maxwell, Prompt Works Studio, Toowoomba · Published 21 October 2026 · Updated 6 November 2026
Drafted with Claude. Checked and edited by Chad Maxwell.

It's 2 pm on a Wednesday. Your office manager has spent the morning in Claude, pulling three months of supplier emails into one summary for a meeting at 3. Then a message pops up. She's reached her limit, and it won't reset for another two hours. The meeting goes ahead with half a summary.

That's a made-up Wednesday, but it's a common one. Between 3 and 11 October, 93 posts in one of Reddit's Claude forums had "limit" or "usage" in the title. Those are people's own posts, and nobody has checked them, but they show where the pain is. The plans keep changing too. Anthropic released three new models between 22 September and 7 October, and it keeps updating its page on how the limits work.

My short answer: don't buy the biggest plan first. Put one person on Pro, with the privacy setting checked. If your office handles client files, start on Team instead, which needs at least two people. Show them what eats usage: long chats, the heavier models and tools left switched on. Look at the usage page after a fortnight. Move up when the work calls for it.

What is the Claude Team plan?

It's Claude's plan for a business with more than one person. Anthropic's pricing page says it's for teams of 2 to 150. You get one bill and one place to run everyone's accounts, and you can mix seat types.

What does the Claude Team plan cost?

Those are Anthropic's US prices, read on 6 November 2026. Anthropic says they don't include tax and can change, and that pricing, currency and tax vary by region. So check what it shows you for Australia at claude.ai/upgrade before you pay.

The smallest Team plan is two standard seats. Billed monthly, that's US$50 a month before tax.

Who sets up the Claude Team plan?

You can do it yourself. Or I'll do it with you, as part of my Claude Set-up, which also covers your privacy settings, an AI use policy and training for your people. The plan itself is paid to Anthropic, on top of what I charge.

Is Claude free for a small business?

You can start for free. Anthropic says the Free plan "covers everyday questions". For paid plans it doesn't promise a number of messages. It tells you how much more you get than another plan:

• Pro gives you at least 5 times the usage of Free in each 5-hour session.
• Max gives you 5 times the usage of Pro, every 5 hours and every week.
• Ultra gives you 10 times the usage of Pro every week.
• On Team, a standard seat gives more than Pro, and a premium seat gives 5 times a standard seat.

How far any of that stretches depends on you. Anthropic says it depends on the length and complexity of your conversations, the model you choose and the features you use. In its words, there's "no fixed message count". That's why two people on the same plan can have very different weeks.

Does Claude train on my chats?

Look at privacy first. Free, Pro, Max and Ultra are personal plans. On those, you decide in your privacy settings whether Anthropic can use your chats to improve its models. Team and Enterprise come under Anthropic's business terms, and Anthropic doesn't train on those chats by default. If your office handles client files, tax records or patient notes, check that setting on day one, or start on Team. Team needs at least two people. And some jobs don't belong in any AI chat. I can help you work out where that line sits for your office.

Where did my usage go?

Paid plans have two limits. One resets every five hours. The other resets once a week. Verified Ultra accounts don't have the five-hour limit. Hit either, and you wait for the reset, move up a plan, or keep going on usage credits if you've turned them on. Usage credits are extra usage you pay for as you go.

Everything you do on your account counts towards the same pool: a chat on the website, the desktop app and Claude Code, the version programmers use. If you also use it to build things, your everyday chats run out sooner.

Then there are long chats. On some accounts, Claude tidies up the earlier messages when a conversation runs long, so it can keep going. Anthropic says those long conversations use more of your limit, and suggests starting a new chat when you're getting close. In that made-up Wednesday, the morning-long chat is what ran the limit down.

Does it matter which model I pick?

Yes. Claude comes as a few models, and you can change which one you're using. Anthropic says the model you're chatting with, and the effort level you've chosen, both change how fast your usage goes. The effort level decides how thorough its answers are, and higher effort uses more.

Fable, one of Claude's models, is the clearest case. On Max, Ultra and a premium Team seat, you can spend up to half your weekly limit on it at no extra cost. It also uses that limit faster than the other models. On Pro, and on a standard Team seat, Fable isn't part of the plan's limits at all. It runs on usage credits instead. If someone on Pro is using Fable for everyday emails, that's the first thing I'd change.

What should I check before paying for a bigger plan?

Here's what I'd go through with whoever hits the limit most, before anyone pays for an upgrade.

Remember when the office internet had a data cap?

You learned fast which jobs ate the data and when it reset. You didn't buy a bigger plan until you knew that. Claude's limits work the same way. Find out what's eating it and fix the easy things first. Then pick the plan that fits the work you do.

Want to know where AI fits in your business? Book a call.