1 sentence covers it: unlimited reviews, and your credits are spent only when the product acts for you.
Free, always, however much you use it
- Collecting reviews from every connected source, at any volume.
- Reading them, searching them, filtering them and exporting them.
- Every action you take by hand: replying, copying a reply and marking it replied, assigning, archiving, approving and discarding a draft.
- Connecting, reconnecting and disconnecting sources.
- Inviting people, creating locations, writing rules and agents, and reading insights.
There is no cap on reviews or team members. A busy month costs the same as a quiet one as long as people are doing the work.
The credit meter showing how much of the month's credits have been used.
Metered, when the product acts for you
Each action a rule takes spends 1 credit, counted per action rather than per rule: a rule that flags and assigns spends 2. Translation costs 1 credit per review or reply it translates.
A reply an AI agent writes costs more, because the AI writes it, and a reply to a low rating costs more than a reply to a happy one. The cost shows under every draft before it goes out. A draft you discard costs the same as one you send, and approving it costs nothing. The full list is in What uses your credits, under Billing.
What happens at zero
Automation pauses. Rules keep matching every review and record what they would have done, so you can see exactly what was missed. Agents stop.
What never pauses: reviews still arrive, the feed still opens, search and filters still work, and every manual action is still there. You are never locked out of your own reviews because a meter ran out.
Top up your credit pool or wait for the monthly reset, and the automation resumes on the next review. The backlog is not replayed, because a thank-you that lands 2 days late reads worse than one written by hand.
Reading the meter
Your credit pool and what is left of it sit on the billing screen in Settings, with the date it resets. You get a warning at 90%.
Insights shows the same story from the other end: how many actions your automation took, which rules and agents took them, and how much people handled by hand. If your credits go faster than you expected, that screen usually names the rule responsible.
Why it is built this way
Metering the reviews would mean paying more the more customers talk about you, which is a strange thing to charge for. Metering the automation keeps the price attached to work that is done for you, and keeps the part that matters most, reading and answering your own customers, permanently free.