In Review is the status a ChatGPT ad or campaign enters after you submit it, while OpenAI checks the title, copy, media, and landing page against its advertising policies before the ad can serve. It gates whether an ad serves and sits apart from billing and account verification, so an ad can remain In Review even when your account is fully approved and funded.
If you treat In Review as a formality, you will schedule launches and spend commitments around a date the platform has not agreed to. Ads stuck or rejected in review never reach a single ChatGPT conversation, so flight dates slip and planned reach quietly disappears.
An ad marked In Review has been submitted but not yet cleared to appear in ChatGPT answers, because OpenAI's review of the creative and its destination is still open. The status applies at the ad and campaign level, and you read it in Ads Manager Beta, where a non-serving ad exposes the reason on hover.
Review is triggered by submission and by later edits, and it runs across separate checks that OpenAI documents as account verification, account approval, ad review, and ad delivery. Clearing one does not clear the others, so an ad can pass content review and still fail to serve because business verification or billing is incomplete. The review itself looks at the title, copy, media, and the landing page, which has to be reachable for the system to evaluate it.
In Review sits upstream of serving and is distinct from a Not serving result, which can mean review is still open or that something was rejected. Keeping your landing page accurate and consistent with your ad claims is what gets you through destination-integrity checks, and AirOps helps you keep that content aligned.
Resources: see how to prepare ChatGPT ads and landing pages that pass review
A ChatGPT ad moves through In Review in a fixed order, and each stage can send it forward or back.
Submit. You finish the campaign and submit it; there is no pre-clearance, so the ad enters review only once it is submitted.
Policy check. OpenAI reviews the title, copy, media, and landing page against its advertising policies, and loads the destination to confirm it is reachable.
Decision. The ad is approved to serve, flagged for a fixable edit, or rejected on policy; the status tooltip in Ads Manager carries the reason.
Resubmit or appeal. For a fixable issue you edit and resubmit, which restarts review; for a policy rejection you appeal or rework the ad.
Ongoing review. Approval is not final, and OpenAI can pull a live ad back into review or limit its delivery later.
The status tells you whether an ad is cleared to serve right now. It does not tell you how an approved ad will perform once it reaches conversations.
Media plans commit budget to dates. In Review decides whether those dates are real, because an ad that has not cleared review cannot spend a cent or reach a single person. Treating review as a checkbox is how a launch date becomes a guess.
It controls your launch date. A campaign you promised stakeholders for Monday does not exist until review clears it, so build review time into every flight. Resubmitting after an edit restarts the clock, so a last-minute change can cost you the launch window.
It is where silent failures hide. An ad can read as active in the API while review still gates serving, so a campaign looks live in a dashboard yet delivers nothing.
It shapes creative and compliance work. Because review inspects copy, media, and the landing page together, a mismatch between ad claims and the destination sends you back into rework before you can optimize. Fixing that loop early keeps spend flowing instead of stalling.
SEO managers use In Review to confirm a landing page is reachable and matches ad claims before a campaign can clear policy checks.
Content strategists use In Review as a prompt to align ad copy and destination messaging so the creative survives review the first time.
Growth marketers use In Review status to gate launch timelines, holding spend commitments until an ad is actually cleared to serve.
Account verification, account approval, ad review, and ad delivery are independent gates in ChatGPT Ads, so clearing one of them tells you nothing about the status of the others, and an ad can pass content review while a billing or identity hold quietly stops it from serving.
There is no pre-clearance step in ChatGPT Ads, so an ad only enters review once you submit it, every later edit sends it back through the same gate, and resubmitting restarts the clock on how long clearance takes.
Passing review once does not lock an ad in place, because OpenAI can pull a live ad back for additional review, limit its delivery, or remove it at any time if a later signal raises a policy concern.
Confirm an ad is cleared to serve in ChatGPT before you commit spend to a flight.
Catch landing-page and policy problems while they are still cheap edits.
Protect launch dates by building review time into the media plan.
Avoid wasted budget on a campaign that reads active but is gated from serving.
Keep ad claims and destination pages aligned so review passes the first time.
Submit early, because review takes time you do not control and resubmits restart it.
Make your landing page reachable and fast, since an unreachable destination cannot be evaluated and will not clear.
Match ad copy to the destination, so a claim in the ad appears on the page it links to.
Check the status tooltip before editing, because the reason often points to verification or billing instead of the creative itself.
Finish account gates first, including business details, identity verification, and billing, so content approval can actually convert into serving.
Keep a resubmission log, so you learn which edits trigger review and how long clearance usually takes for your account.
Avoid treating a rejection as a creative problem by default. Competent teams burn days rewriting copy when the real block is an unverified business or an unreachable landing page, so read the status reason before you touch the ad.
AirOps: keeps your landing pages and ad claims accurate and consistent, so destination-integrity checks clear the ad instead of flagging a mismatch.
ChatGPT Ads Manager: shows the live status of every ad and campaign and exposes the review reason on hover so you know why an ad is not serving.
Screaming Frog: crawls your destination pages to catch broken links, slow responses, or unreachable URLs before they stall an ad in review.
Audit your landing page. This week, open the destination URL and confirm it loads, matches the ad's claims, and names the same offer. No budget approval needed.
Clear account gates. Complete business details, identity verification, and a billing method so that content approval can convert into live serving. You cannot serve on content alone.
Build the creative to policy. Write titles, copy, and media that fit the character limits and advertising policies for your category before you submit. A mismatch here is the most common reason an ad stalls.
Submit and watch the status. Publish the campaign, then check Ads Manager for the serving status and hover any non-serving ad to read the reason.
Fix and resubmit deliberately. For a fixable flag, make the one change the reason points to and resubmit; for a policy rejection, appeal or rework instead of guessing at edits.
In Review means a ChatGPT ad has been submitted but is not yet cleared to serve to users.
You read the result in Ads Manager, where a non-serving ad reveals its reason on hover.
Account verification, approval, review, and delivery are separate gates, and all must clear before an ad serves.
An ad can look active while review still blocks it, so a dashboard can hide a campaign that delivers nothing.
Aligning the landing page with ad claims is the fastest way to pass review and protect the launch date.
In Review describes an ad that OpenAI is still evaluating after submission, while Not serving is the broader status you see in Ads Manager when an ad will not show to users. An ad can read Not serving because review is still open, because it was rejected on policy, or because an account gate like verification or billing has not cleared. So every In Review ad that is not yet approved shows as Not serving, but plenty of Not serving ads have already finished review and failed a different check. The practical move is to hover the status and read the reason before you act. If the reason is a fixable creative flag, you edit and resubmit. If it is an account hold, no amount of creative editing will change the status, and you need to clear the underlying gate first.
It depends, because OpenAI does not publish a fixed review time, and the clock is affected by your category, whether your account has cleared verification, and whether an ad needs manual review. Standard categories in approved accounts can clear quickly, while restricted categories such as healthcare, financial services, and legal services are reviewed manually on a case-by-case basis and take longer. Every edit you make after submission restarts review, so a last-minute change can reset whatever progress you had. The safe planning habit is to submit well ahead of your launch date, avoid editing once the ad is in review, and give a newly launched ad time to begin delivering before you treat zero delivery as a problem. If an ad sits far longer than your other ads, hover the status for a reason and check your account gates instead of assuming the review itself is stuck.
Variation comes mostly from the risk profile of what you are advertising and the state of your account. Ads in restricted or regulated categories trigger extra safeguards, including enhanced advertiser verification and manual, case-by-case review, so they move slower than a standard product ad in an already-approved account. The landing page matters too: if OpenAI cannot load or evaluate your destination, the ad cannot clear, and a slow or intermittently reachable page drags out the check. Account readiness is the other big factor, because an unverified business or a missing billing method holds the whole account regardless of how clean the creative is. Finally, edits reset the clock, so an ad you keep tweaking will always look slower than one you submitted once and left alone. If one ad lags your others, compare its category, its destination, and its edit history before assuming the delay is random.
You cannot force OpenAI to review faster, but you can remove most of the reasons review stalls, which has the same effect. Clear your account gates first, so business verification and billing are done before you ever submit an ad; a surprising share of slow or failed reviews are account holds wearing a creative disguise. Build the creative inside the stated character limits and advertising policies for your category, and make sure the landing page loads quickly and states the same offer your ad promises. Then submit once and leave it alone, because every edit restarts the clock and undoes any waiting you have already done. If your category is restricted, expect manual review and plan the timeline around it instead of fighting it. The lever you actually control is preparation, and ads that arrive clean clear faster than ads that need a round of fixes.
A healthy outcome is simple to describe: the ad clears review on the first submission and moves to a serving status without a round of edits. In practice, good looks like a campaign that passes because you did the preparation, your account was verified, your billing was live, your category was eligible, and your landing page matched your ad. A single fixable flag is still fine if the reason is specific and you resolve it in one edit. The warning signs are an ad that bounces in and out of review after repeated edits, a rejection you cannot map to a clear policy reason, or a campaign that reads serving yet delivers nothing after a full day. Track how many of your ads clear on the first try, because a high first-pass rate is the most honest measure that your review process is working.