How Many Ads Do DTC Brands Run on Meta? Creative Volume and Spend Concentration Benchmarks (2026)

By Kamal Razzak, founder of MHI Media · Published 31 October 2026 · Updated 31 October 2026

Every figure here is a count or a share, read on 29 October 2026 from the Meta ad accounts of 73 DTC brands in MHI Media's warehouse of client ad accounts, data settled to 28 October 2026. Facebook and Instagram only. The 90-day window (31 July to 28 October 2026) includes a brand with at least £3,000 of ad-level spend and 30 purchases; the monthly tables take a brand in any month with at least £1,000 of spend and 10 purchases, 40 to 69 brands a month. Cells are medians across brands with the interquartile range (25th to 75th percentile) and the 10th and 90th percentiles. No money appears on this page; the hub quotes USD at £1 = $1.3289 (multiply by 0.7525 for GBP), and the thresholds above are in GBP. Method: the methodology on the benchmarks hub.

An ad is one distinct ad name within an ad account, so copies in several ad sets count once; account-level rows are excluded, and ad history starts 1 September 2025. Elsewhere in the dataset, CPA, ROAS, spend and purchases run to 28 October 2026; CTR, CPM, CPC and conversion rate cover 8 January to 7 September 2026 only; frequency to 3 August 2026 only; placement 8 June to 13 July 2026 only. This page uses ad-level spend alone. It does not see blended or lifetime revenue, profit, other channels, or what each ad looked like, so an account that names every variant separately counts higher.

How many ads do DTC brands run on Meta in 90 days?

The median DTC brand had 1,392 distinct ads with spend in the 90 days from 31 July to 28 October 2026, and 19 of those ads took 1% or more of its 90-day spend (73 brands).

A tenth of brands ran under 206 ads and a tenth ran over 5,908. The second row barely moves: 13 to 24 ads cleared the 1% line whether the brand ran 200 ads or 6,000, and that is the number I would pin up. Over a full year it is starker still: the top 1% of ads at 59 brands took 61% of the spend, in who gets the money on Meta. Whatever the volume, the budget settles on about 20 ads; the rest are tests, catalogue variants and ads Meta stopped serving after a few pounds.

How concentrated is Meta ad spend on the top ads?

The median brand put 8.0% of its 90-day spend on its single top ad, 26.6% on its top 5 ads and 39.0% on its top 10 (73 brands, 31 July to 28 October 2026).

A quarter of brands put over 55.8% of spend on ten ads and over 13.3% on one; a quarter had a top ad under 5.1%. Heavy concentration means the account is riding a winner until it fades; a thin spread means near-equal ads or no winner yet. I read the top-ad share with the age of that ad: over 13.3% on an ad from the spring needs a replacement in testing now.

How much Meta spend goes on ads under 90 days old?

The median brand put 58.9% of its 90-day spend on ads that first spent inside the window, so on ads under 90 days old, with the middle half between 42.3% and 70.4% (last row of Table M1-1b; 73 brands, 31 July to 28 October 2026).

One brand in ten put all of its spend on ads that new; the bottom tenth put under 27.2% there. The top group may be launching a lot or may have cleared out old winners; the data cannot separate the two. Under 27.2% means most of your budget is on creative from before August, and I would check whether that is because it still wins or because nothing new has beaten it. The post on reducing creative fatigue on Meta covers how to tell.

How many new ads do DTC brands launch a month?

Averaged over July to September 2026, the median DTC brand had 782 active ads a month and launched 353 new ones (61 brands).

The median brand launched about 350 ads a month over the summer, and about 19 ads took a meaningful share of its budget. That is the honest answer to how many ads to test: volume buys the shortlist. Launching 10 a month puts you under the 10th percentile here (54), so each of the ten has to be a distinct concept. How many creatives to test per week covers cadence by budget; the creative testing framework covers how we fill and read each batch.

Active and new ads by month, October 2025 to October 2026

Median active ads a month rose from 224 in October 2025 to 711 in September 2026, and median new ads from 102 (November 2025) to 389, as qualifying brands rose from 40 to 69.

Part of that climb is roster mix: 45 brands qualified in January and 69 in September. Part is behaviour, as accounts leaned on Advantage+ and catalogue formats that multiply ad names; the table cannot split the two. October covers 1 to 28 October, and accounts moved to a second Business Manager on 9 October report at account level, so they drop out and n falls to 51. Wait for the full month before reading the 876.

What this means for your account

Over your last 90 days at ad level, count ads with spend, ads taking 1% or more, and the share on your top ad. Fewer than 13 ads at the 1% line puts you under the 10th percentile. A top ad over 13.3% of spend puts you in the most concentrated quarter; over 21.6% is the top tenth. I treat a single ad over a fifth of spend as a health check rather than a win: work out what replaces it when it fades. If your top 10 take under 26.0%, check whether they are distinct concepts or one concept under several names; if distinct, there is no winner yet and the concepts come first.

Behind these tables sit three campaigns: a testing campaign where new ads launch in small batches, a winners campaign that scales the ads that keep taking spend, and a graveyard campaign where faded or near-miss ads sit on a bid cap; the testing, winners and graveyard post sets it out. If the bottleneck is making enough ads to keep testing fed, that is what our creative testing agency does, and the Icy Bear case study shows what one fresh batch of statics did. MHI Media's studio makes tens of thousands of ads a month across its clients; that is our production, separate from this dataset.

FAQ

How many ads should I run on Facebook and Instagram?

Across 73 DTC brands in the 90 days to 28 October 2026, the median brand had 1,392 distinct ads with spend (middle half 446 to 2,841), and only 19 of them took 1% or more of its spend. Run enough ads to find about 20 that deserve the budget; the dataset counts ads per account, so it cannot size an ad set.

How many new creatives should I test on Meta ads each month?

Averaged over July to September 2026, the median DTC brand launched 353 new ads a month, with the middle half of 61 brands between 94 and 924 and the bottom tenth under 54. These are distinct ad names, so catalogue and placement variants inflate the count, and the dataset cannot set the right number for you.

What share of Meta ad spend goes to a brand's top ad?

Across 73 DTC brands in the 90 days to 28 October 2026, the median brand put 8.0% of its spend on its single top ad, with the middle half between 5.1% and 13.3% and the top tenth above 21.6%. The top 5 ads took 26.6% at the median and the top 10 took 39.0%.

How often do DTC brands refresh their Meta ad creative?

The median DTC brand put 58.9% of its 90-day spend on ads that first spent inside the window, so on ads under 90 days old, across 73 brands in the 90 days to 28 October 2026. The middle half sat between 42.3% and 70.4%, and one brand in ten put 100% of its spend on ads that new.

Sources

• MHI Meta Benchmarks, DTC, October 2026 edition, MHI Media; hub post with methodology, 22 October 2026
• MHI Media warehouse of client Meta ad accounts, benchmark cuts edition, read-only extract 29 October 2026, data settled to 28 October 2026