# Let Users Change Their Plan Without a Ticket

> How Peeve lets SaaS users change their plan in-app: it confirms the exact effect and makes the change with a signed audit log, or shows the steps instead.

Canonical: https://peeve.ai/use-cases/change-plan/

Peeve lets your users change their plan without opening a ticket. It confirms the exact effect of the change, then completes the backend action with a signed audit entry, or simply shows the user the step on their live screen. Changing a plan is a Do action, available on the Growth plan and up; Show is on every plan.

## How do users ask for this?

- "How do I upgrade to the annual plan?"
- "Can I downgrade before my next bill?"
- "I need to switch plans, where do I do that?"

## What does Peeve do: show or do?

Peeve can show the user the exact control on the billing screen and let them click it, or, with your Do connector, complete the plan change itself after confirming what it will do. You choose which, and it never changes a plan without the user's go-ahead.

## Guardrails: confirmation and audit log

Peeve runs only the actions you allow, within the logged-in user's permissions, and it confirms the exact effect before anything changes. Every action is written to a signed audit log, and the model never sees a password or token: it requests a capability by name.

## Which plans include it?

Showing the user the step is on every plan, including Starter (Answer and Show, read-only). Completing the change as a backend action is a Do feature, on the Growth plan ($199 a month) and above.

## Questions

### Can an AI support agent change a user's plan?

Yes, if you connect it to your APIs and set rules. A good agent runs only actions you have allowed, within the logged-in user's permissions, asks the user to confirm before any change, and logs every step. Peeve's Do feature changes a plan this way on the Growth plan and up; the model never sees passwords or tokens.

### Is it safe to let an agent change plans?

Yes, because the change is gated. Peeve confirms the exact effect before it runs, acts only within the user's own permissions and your allow-list, and writes a signed audit entry for every action. If you prefer, it shows the user the step instead of acting, which is available on every plan.
