Why Did My Meta Ads Stop Spending?

By Kamal Razzak, founder of MHI Media · Published 1 November 2026 · Updated 1 November 2026

Meta ads stop spending for three reasons: something switched them off (a spending limit, a rejection or a pause), the ad wore out, or a newer ad took the budget. Across 67 DTC brands, 55% of ads that reached £100 a day had stopped within 7 days and 83% within 60. Budget rises are rarely the cause. In 98 rises the median spent 99% of the extra budget, and about one in four spent under half of it.

I pulled this from the Meta ad accounts of the DTC brands in MHI Media's warehouse of client ad accounts, read on 1 November 2026 with data settled to 31 October 2026. Facebook and Instagram only. A stop means an ad averaged at least £100 (about $133) a day for 7 days, then spent under 20% of that level over the next 7. Every cell covers 8 or more brands. The method is at the bottom of the page.

Check these first

Rule out anything that has switched the ad off before you start reading the numbers. Work down this list in Ads Manager. It takes about five minutes.

  1. Delivery column. Check it at campaign, ad set and ad level. Each can show a different status, so an active campaign can still hold a paused ad set (Meta, delivery status).
  2. Rejected or flagged ads. A rejected ad will not run. Request a review or make a new version.
  3. Account spending limit. When you hit it, your ads pause and stop spending until you raise, remove or reset it (Meta, account spending limit).
  4. Campaign spending limit. When you hit it, every ad set and ad in the campaign stops (Meta, campaign spending limits).
  5. Payment method. Look for a failed payment in Billing.
  6. End dates and schedules. Check the ad set hasn't reached its end date.

If all six are clear, the ad is on and Meta has stopped buying impressions for it. The rest of this page is about that case.

How long do Meta ads keep spending?

Most ads that get to £100 a day don't stay there long. We followed 9,596 ads across 67 brands from the day they first averaged £100 a day. More than half had stopped within a week, and about one in six was still going after 60 days.

Share of ads that had stopped, by days since first averaging £100 a day. 9,596 ads, 67 DTC brands, first reached between 8 May and 24 August 2026, followed to 24 October 2026.
Days since reaching £100 a dayShare stopped
754.6%
1466.0%
3076.8%
6083.0%

Source: MHI Media warehouse of client Meta ad accounts, read 1 November 2026.

Among the 12,784 ad stops we counted between 13 July and 24 October 2026, the median ad stopped 11 days after its first spend, and the middle half stopped between day 6 and day 24 (76 brands). So an ad that spends for a fortnight and fades has had a normal life. The account needs the next winner ready before that happens. The ads that don't fade carry the account: over 12 months the top 1% at 59 brands ran for a median of 92 days and took 61% of the spend (who gets the money).

What had happened to the ads that stopped?

About one in five stopped ads is still switched on, which means Meta simply stopped delivering it. We matched 11,728 of the 12,784 stops to the ad's status today.

Status today of ads that stopped spending between 13 July and 24 October 2026. 11,728 stops with a status record. Status was read on 1 November 2026, so a pause may have come after the drop.
Status todayAdsShareBrands
Ad set switched off4,39737.5%57
Ad switched off2,93025.0%61
Still switched on2,25219.2%60
Campaign switched off1,39811.9%49
Archived or deleted3312.8%19+
Flagged with issues3072.6%32
Disapproved1121.0%20

Source: MHI Media warehouse of client Meta ad accounts, read 1 November 2026. One further ad was pending review.

Rejections are a small part of it: 3.6% of stops were flagged or disapproved. The big group is ads and ad sets that a person has switched off. Buyers often pause an ad once it has already faded, so read that row as the clean-up after a drop as much as the cause of one.

Did the ad wear out?

Rising CPA is the best warning we found in the data. Of the stopped ads with enough purchases to compare, 56.7% had seen CPA rise by 10% or more in their last week, measured against the week before (4,934 ads). For ads that kept spending, the same rise showed up in 40.9% of weeks (65,949 ad weeks).

That's a useful signal and a noisy one. Plenty of healthy ads have a bad week. What matters is the pattern: spend is flat or falling on your best ad, and CPA has moved up for a week or two. That's when you want new concepts going in, ahead of the drop. Our guide to creative fatigue on Meta covers what to change first, and frequency management covers how to read the frequency column alongside it. An iteration on the tired ad buys you a little time. A new angle usually buys you more.

Why does one ad take all the budget?

Meta puts money behind the ads it expects to bring in purchases, so spend inside an ad set piles up fast. In October 2026 we looked at 5,975 ad sets with three or more ads and at least £700 of spend, across 74 brands. The top ad took 75.3% of the median ad set's spend, and the top two took 95.1%. The median ad set had five ads, so the other three shared what was left.

This explains a lot of ads that look broken. An ad getting a few pounds a day has usually lost the auction inside its own ad set. It hasn't been penalised.

It also explains how winners end. In 11.6% of stops the ad set kept spending at 70% or more of its old level, which means the money moved to other ads. In 68.4% of those cases a new ad had started spending in the same ad set within the week before or the three days after. So when your best ad drops and the ad set holds steady, look at what you launched last week before you panic.

What happened to the ad set when an ad stopped?

In two thirds of stops the whole ad set went quiet with the ad. That's the most common picture in the data, and it points at an account with too few ads able to take spend.

What the ad set's spend did in the 7 days after an ad stopped. 12,686 stops matched to an ad set, 13 July to 24 October 2026.
Ad set spend after the stopStopsShareBrandsNew ad started in the ad set
Held at 70% or more1,47411.6%5768.4%
Fell by 30% to 80%2,65020.9%7049.6%
Fell by more than 80%8,56267.5%7135.6%

Source: MHI Media warehouse of client Meta ad accounts, read 1 November 2026. New ad means first spend from 7 days before the stop to 3 days after.

Where the ad set collapsed, a replacement had started in only about a third of cases. If one ad is carrying the ad set and it fades with nothing behind it, the spend goes too. Get new concepts into the account while the winner is still working. Our Meta ads team runs testing, scaling and resting campaigns for this reason: there's always something coming through to take the spend.

Why do Meta ads stop delivering after I scale the budget?

Mostly they don't. Spend followed the budget in most of the 98 increases we could measure cleanly, across 25 brands. The median rise was 75%, spend rose 68.9%, and 99% of the extra budget got spent. The problem cases are real, though. In about one rise in four, under half of the new money was spent, and in 13.3% spend fell in the week after.

Budget rises of 20% or more on budgets that were being spent, 13 July to 17 October 2026, with spend compared for the 7 days either side. Only groups of 8 or more brands are shown.
GroupRisesBrandsMedian budget riseMedian spend riseSpent under half of the extra budgetSpend fell
All rises982575%68.9%23.5%13.3%
Campaign budgets281350%53.6%25.0%17.9%
Ad set budgets701890.1%73.9%22.9%11.4%
Ad set budgets up 50% to 99%241253.8%48.5%33.3%12.5%
Ad set budgets up 100% or more3210100%114.5%21.9%15.6%

Source: MHI Media warehouse of client Meta ad accounts, read 1 November 2026. Small sample, so read it as a guide.

The bigger risk comes a week later. In 25.5% of rises, spend in the second week had fallen back below where it started. Cost also moved: across 25 rises at 9 brands with enough purchases to compare, median CPA went up 9.5%, it rose by 20% or more in 40% of cases, and it fell in 24%.

Why spend stalls after a budget rise

When a rise doesn't get spent, look at these four things in order.

  1. Your bid or cost control. A cost cap or bid cap tells Meta to walk away from impressions above your limit, so a bigger budget can sit unspent. Our bidding strategies guide explains the trade-off.
  2. A reset learning phase. Meta says a budget change may count as a significant edit, depending on how large it is, and that a significant edit may restart learning (Meta, last significant edit). Meta publishes no fixed percentage. Check the Last significant edit date after you make a change. More on this in our learning phase guide.
  3. One tired ad. If a single ad holds most of the ad set's spend, the extra budget has only that ad to run through. Scaling a fading ad speeds up the fade.
  4. Too little behind it. The ad set needs a second and third ad that can take spend, or a winners campaign built for scale.

What to do on Monday

If this keeps happening, creative is usually what's holding the account back. That's the work our Facebook ads and performance creative teams do.

How we built these numbers

The data is ad-level daily spend and purchases from the Meta ad accounts of the DTC brands in MHI Media's warehouse of client ad accounts, with money converted to pounds so one threshold applies to every account. A brand is one client with all its mapped ad accounts added together, and accounts mapped to two clients are left out. We publish aggregates only, and every cell covers 8 or more brands.

Budget rises and ad set status changes come from Meta's activity data, which we hold from 6 July 2026. A rise counts when the budget went up by 20% or more, the campaign or ad set had spent 50% to 150% of the old budget a day over the week before, it had at least £30 of daily spend, and no further budget change or switch-off followed within 7 days. Those filters keep the sample small and clean.

The data shows what happened to spend. It can't show what Meta's system decided, and it can't always show whether a person paused an ad before or after it faded. CPA is Meta-attributed.

FAQ

Why did my Meta ads suddenly stop spending?

Check the Delivery column at campaign, ad set and ad level first, then rejected ads, your account spending limit, your campaign spending limit and your payment method. If all of those are clear, the ad has probably worn out or a newer ad has taken the budget. Across 67 DTC brands, 55% of ads that reached £100 a day had stopped within 7 days.

How long does a Meta ad usually last?

Shorter than most people expect. Of 9,596 ads that reached £100 a day across 67 DTC brands, 66.0% had stopped within 14 days and 83.0% within 60 days. The median ad that stopped did so 11 days after its first spend. About one in six was still going after 60 days, and those are the ads worth building on.

Why do Meta ads stop delivering after I scale the budget?

Usually they keep delivering. In 98 budget rises across 25 DTC brands, the median spent 99% of the extra budget. About one in four spent under half of it. When that happens, check your cost cap or bid cap, whether the change restarted learning, and whether one tired ad is holding most of the ad set's spend.

Does increasing the budget reset the learning phase on Meta?

It can. Meta says a budget change may count as a significant edit depending on how large it is, and that a significant edit may restart learning. It publishes no fixed percentage, so check the Last significant edit date in Ads Manager after you make a change. Smaller steps are safer when an ad set is working.

Why is Meta only spending on one ad in my ad set?

Meta puts budget behind the ads it expects to bring in purchases. In 5,975 ad sets with three or more ads across 74 DTC brands in October 2026, the top ad took 75.3% of the median ad set's spend. That's normal. Move on from the low-spend ads, and put your effort into the next concept.

Sources