Ads
Meta ads stopped converting: what changed?
When Meta ads stop converting, check causes in order from cheapest to confirm: tracking and events, attribution setting changes, the learning phase after edits, creative fatigue and frequency, audience and auction costs, the landing page and offer, then seasonality. Most sudden drops come from the first three, and each check needs evidence before anything is changed.
2.13× Meta ROAS for Moolwan on ₹28,064 spend, against a 4× benchmark, a why still being investigatedai-dm · Paid ads · 30 days to 26 Sep 2026 · real data
What you'll see in 10 minutes
- Your first brand's report, built from its domain alone.
- Its Meta ad account connected with one secure sign-in.
- A Meta ads audit with spend, results and the likely causes of a change, each linked to its record.
What does "Meta stopped converting" usually look like?
The message arrives before the report. The brand's founder writes on a Tuesday morning: purchases from Meta have halved since last week, spend is the same, and what happened? The ads look the same as they did when they were working. Nobody on the team made a big change, or nobody remembers making one.
Guessing is the expensive part. The tempting fix is to pause the campaign, swap the creative and raise the budget, all at once. If the real cause was a broken purchase event, the team has just reset the account's learning and spent another week on ads that were working fine. If the cause was the season, nothing they change will bring last month back. The order you check things in decides how fast you find the real answer.
In what order should you check a Meta account that stopped converting?
From the cheapest to confirm to the most expensive. Each check below takes minutes and either clears a cause or confirms it. Work down the list and stop when the evidence explains the drop.
Tracking and events
Open Events Manager and check that the Purchase event is still arriving at its usual volume, from both the browser pixel and the Conversions API if the brand uses both. Compare Meta's purchases with the brand's own store orders for the same days. If orders held steady and Meta's purchases fell, the ads may be fine and the tracking is not. A site update, a new checkout app or a changed thank-you page are the usual suspects.
Attribution setting changes
Check whether anyone changed the attribution setting on the ad sets or in the report columns. Moving from a 7-day click and 1-day view setting to a shorter one cuts the number Meta reports without a single sale being lost. If the reporting changed, compare like with like before you call it a drop. For why Meta's number differs from GA4 in the first place, read how to reconcile GA4, Meta and Google Ads numbers.
Learning phase and recent edits
Look at the account's change history for the days before the drop. A new ad set, a budget jump, a new optimisation event or a big edit to targeting or creative can send an ad set back into Meta's learning phase, when delivery and results are less stable. If the drop started the day after an edit, the edit is the first explanation to test, and more edits will usually prolong it.
Creative fatigue and frequency
Check frequency and click-through rate for the ads that carry most of the spend. If frequency has climbed and click-through rate has fallen on the same ads while CPM held steady, the audience has seen them too often. That pattern is creative fatigue, and it builds over weeks rather than overnight, so look at the trend, not one day.
Audience and auction costs
Check CPM across the account. If it rose across most ads at once, including new ones, other advertisers are bidding for the same people, which is common before festivals and sale events. Look at the ads rivals are running for signs of a competitor's sale. A smaller audience, such as a tight lookalike or a narrow interest set, makes this worse.
Landing page and offer
If clicks held and purchases fell, the problem is after the click. Open the landing page on a phone. Check speed, stock, price, the discount code in the ad and the checkout. A product that went out of stock or an offer that ended will stop conversions on ads that are otherwise working.
Seasonality
Finally, compare the same weeks last year, if the brand has the history. Some categories slow down after a festival or at month-end, and a drop that matches last year's pattern needs a different conversation with the brand than a drop that doesn't.
What does this look like on a real account?
Our own brand is below benchmark too. Moolwan, the Bengaluru D2C home décor brand the ai-dm founders run, shows a Meta ROAS of 2.13× on ₹28,064 of spend against a 4× benchmark, with 171,489 impressions and 3,895 clicks in the 30 days to 26 September 2026 (ai-dm · Paid ads). That is not a win, and we don't present it as one.
We are working through the same order. The why is still under investigation, and we're checking each cause against its evidence rather than changing several things at once. The numbers, including the weak ones, are published with their sources, so you can inspect Moolwan's real numbers for yourself.
How does ai-dm help with a Meta drop?
It does the first checks before anyone asks. ai-dm watches each connected Meta account every day and sends a change alert when results move outside their normal range. Its Meta ads audit shows spend, results, frequency and costs by ad set and ad, and lists the likely causes of a change, graded by confidence, each with an evidence trail to the source record.
People stay in charge. ai-dm recommends the next move, such as refreshing a fatigued ad or fixing an event, and drafts it. Your team decides, and ai-dm publishes ads or changes only after the team approves them.
What should you do today?
Answer the brand with a plan, not a guess. Reply that you're checking tracking, settings and recent edits first, and that you'll come back with the cause and the evidence. Then work down the list, record what you find and change one thing at a time. If the brand's numbers drop often, set up daily checks so the next drop reaches you before it reaches the founder.
Related pages
More posts
Google Ads vs GA4 ROAS: why don't they match?
Attribution, counting, dates, import lag and modelling make Google Ads and GA4 ROAS differ. Which to use for which decision, and what to tell the brand.
Read the post →A brand dropped in the Google Maps results: what should you check?
A check order for a Google Maps ranking drop, from geo-grid and profile edits to categories, reviews, competitors, NAP and website signals.
Read the post →
Questions agencies ask
What should we check first when Meta ads stop converting?
Check tracking first. Confirm the purchase event is still firing in Events Manager and that the site's orders match what it did before. A broken pixel or Conversions API setup makes good ads look like they stopped working.
How do we tell creative fatigue from a costlier auction?
With creative fatigue, frequency climbs and click-through rate falls on the same ads while CPM stays steady. With a costlier auction, CPM rises across most ads at once, including new ones, while click-through rate holds.
Should we pause the campaign when conversions drop?
Not before you know why. Pausing or heavily editing an ad set can restart Meta's learning phase and make delivery less stable. Work through the checks in order, record what you find and change one thing at a time.
Can ai-dm make the changes in the Meta account for us?
ai-dm recommends the next move with the evidence behind it, and your team decides. It publishes ads or changes only after your team approves them.
No plan to pick. Billed in USD and INR. A brand's data is deletable on request.