Skip to main content
FHAVI

Research · Merchant Center

How long a Merchant Center review takes — and what actually determines it

Google publishes no per-category review duration, so any table of fixed timelines is invented. What genuinely affects how long a review takes, and the three things that make it slower.

3 min read

In brief

  • There is no published duration by suspension type. Anyone quoting 'three to five business days' for a specific category is guessing.
  • What you control is the completeness and legibility of the evidence, and whether you resubmit while the account is unchanged.
  • Resubmitting during an open review, or twice in the same state, is the most reliable way to make a review take longer.

The honest answer

Merchant Center reviews are queued work performed by people against policy categories whose investigation depth varies enormously. A missing policy page can be confirmed in minutes. A misrepresentation finding across a catalogue requires sampling product pages, comparing them with the feed, checking the business details and reading the account history.

That is why no duration table is trustworthy: the same label can mean a fifteen-minute check or a multi-day investigation, and the reviewer cannot know which until they have started. What can be described — accurately — is what shifts a case in either direction.

Not a timeline, a queue position

Treat a review as a queue with an unknown service time. You cannot shorten the service time, but you can avoid being sent to the back of the queue, and you can avoid being re-queued after a rejection.

What makes it slower

  • Requesting review while the finding is unresolved — the reviewer checks, finds it, and the case is closed with a rejection. The clock restarts from your next request.
  • Resubmitting a second request while one is open. Two open requests do not double your chances; they add noise.
  • A store that changed materially during the review, so the evidence the reviewer is looking at no longer matches what you described.
  • Account history with unresolved warnings or repeated review requests. This is read as a pattern and invites the longer investigation.
  • Evidence that requires the reviewer to infer a fix — 'we improved our policies' rather than a link to the page that now exists.

What shifts it the other way

  1. 1

    Fix everything, then request once

    One request against a store where every finding is demonstrably resolved is the fastest path available, and it is the only one you control.

  2. 2

    Make the fix visible without logging in

    Policy pages in the footer, business details on the contact page, prices visible in the target currency. Reviewers look at what a customer sees.

  3. 3

    Reduce scope

    If ten items caused a catalogue-wide finding, removing those ten and re-syncing is a cleaner state than a hundred partial edits.

  4. 4

    Keep the account quiet

    No theme deploys, no feed experiments, no new accounts during a review. Every change during review is a variable you have introduced.

You cannot know how long a review will take, and anyone who tells you otherwise is guessing. You can know that you requested once, against a store where every finding is fixed and every fix is visible to someone who is not logged in — which is the only lever this process gives you.

Questions

Asked about this directly

Is there an average?
We do not publish one, because the sample any agency sees is small and biased toward hard cases. Google does not publish durations either, and a number offered without a source is a guess presented as a fact.
Should I contact support while waiting?
Contacting support to ask for a status does not move a review, and repeated contacts do not create priority. It is reasonable to contact support when the interface is broken or a finding is unclear — not to accelerate a queue.
What if it has been weeks?
Check that the request is genuinely open and that the store has not changed since you made it. If the state has moved, the most useful action is usually a fresh, single request against the current state rather than a chase.

Keep reading

Other pieces in the library

All research

Numbers first, promises second

Send your store and we will tell you which of these failure modes is already happening — before you pay for a fix.

Prefer email? Write to md@fhavi.com