How many AI requests does each Asana plan give you, and what are they for?

Summary

Two meters, not one

Asana counts AI usage in two different ways, and they behave very differently as a team grows.

Requests cover AI Teammates and Asana Dash. Each user brings 5 a month, but the account can't go past its cap: 50 on Starter and Advanced, 250 on Enterprise and Enterprise+. Here is how that plays out on Starter or Advanced:

Credits cover AI Studio and belong to the billing account, not to people. Starter includes 50,000 a month, Advanced 75,000, and Enterprise and Enterprise+ 200,000. A 5-person team and a 50-person team on Starter get the same 50,000. A credit allowance cannot be turned into a fixed number of runs. Input size, output, model and workflow frequency affect consumption.

The pricing page says more credits and requests can be purchased. AI Assists, the AI built into task, project, and goal screens, is listed separately as included on every paid plan.

What each piece does

AI Studio is a no-code builder. You choose what sets it off (a new task, a form submission, any project event) and write plain-language instructions telling the AI to check a request for missing information, flag duplicates, classify it, route it, or send an alert.

AI Teammates are prebuilt agents, more than 30 of them, covering jobs like writing a launch brief from a project, triaging the day's intake requests, or showing who is overloaded before a new initiative starts. You can also build your own.

Asana Dash is pitched as an AI chief of staff. It pulls decisions out of meetings, Slack threads, and email, maps them back to projects, and hands the next step to an AI Teammate. It draws from the same 5-per-user allowance as Teammates.

AI Connectors and MCP run the other direction. Asana publishes an MCP server (MCP is the protocol AI assistants use to read and act on outside data), and connectors for ChatGPT, Claude, Gemini, and Microsoft 365 Copilot let people search, create, and update Asana tasks from those apps. Enterprise+ and legacy Enterprise can manage access in the admin console. V2 clients are controlled through each associated app; the older beta clients share the Asana MCP app. For other tiers, the developer guide describes a super-admin support request to block that beta v1 app. It does not establish the same procedure for every v2 client.

Data handling

Asana states that its AI partners don't use customer data to train their models and must delete customer data after each query. Review the workflow’s permissions and required approval steps before assigning work.

Ordinary Rules are separate from these AI meters. Pricing lists unlimited automations, while the rule help sets a maximum of 50 rules per project. Our comparison of automation limits across vendors covers those.

A request is completed work, not every message

The pricing FAQ defines an AI request as successful work by a Teammate or Dash. Failed, timed-out and canceled work does not count. Dash’s administrative tasks, such as daily prioritization, are included without consuming paid requests; complex work or a handoff to a Teammate does count. Estimate completed jobs rather than counting every conversation message.

Before buying additional usage

Pricing advertises prepaid requests at $0.50 each and on-demand requests at $0.60 each, billed monthly in arrears. A 100-request prepaid example is $50 up front. Paid packs expire at the monthly billing cycle’s end for monthly customers; annual customers can use them over their contract term. These pack rules differ from the monthly included allowance.

The on-demand card also displays $5.99 per user per month billed annually, while its bullets and FAQ describe per-request billing. The page does not explain how that line relates to the usage charge. Confirm the actual order and any fixed charge before treating $0.60 as the whole price.

If ordinary rules can route the request using fields you already collect, start there. AI Studio is worth testing for text that needs classification or interpretation; Teammates and Dash are options for work that changes from one assignment to another. Measure a representative workflow, inspect the result and assign a reviewer before paying for a larger allowance.

Sources

• Plan prices and AI allowances
• AI overview and data handling
• AI Studio
• AI Teammates
• Asana Dash
• AI connectors
• MCP access controls

Rule limits; AI Studio credit accounting. Checked 10 October 2026.

← Back to the comparison