Why are my CPMs so high around Black Friday?
Why do CPMs go up before Black Friday?
Each time someone scrolls Facebook or Instagram, Meta runs an auction for that ad slot. Every advertiser whose audience includes that person competes, and the winner is the ad with the highest total value: the bid, the estimated chance the person engages or buys, and ad quality (Meta Business Help Center).
In late November, advertisers raise their budgets at the same time and for the same shoppers. In Supermetrics' data on more than 5,000 retail advertisers, Meta spend in Black Friday week 2025 was 1.8 to 2.8 times its September level, depending on the sector. More money chasing the same people pushes the price of each impression up, and a small store bids in the same auctions as national retailers. Meta also notes that its delivery starts with lower-cost opportunities and may move on to more expensive ones when those run out (Meta Business Help Center), so a bigger Black Friday budget usually buys pricier impressions too.
How much do CPMs rise, and when?
We have no CPM data of our own for Black Friday 2025, because our daily ad data starts in 2026. The clearest public timeline is from Lebesgue, a Shopify analytics app, published in October 2026. It covers $22.6 million of Meta spend by its biggest advertisers, compared with each brand's own September.
| Date | Day | CPM vs September |
|---|---|---|
| 10 Nov | First clear rise | +13% |
| 20 Nov | A week before Thanksgiving | +27% |
| 27 Nov | Thanksgiving | +53% |
| 28 Nov | Black Friday | +84% |
| 1 Dec | Cyber Monday | +71% |
| 3 Dec | Wednesday after | +20% |
Until the first week of November 2025, the median brand stayed within 7% of September. The timing moves from year to year: for the same brands in 2024, CPMs first reached 10% above September in the week of 14 October, and Black Friday week was up 80%, against 59% in 2025.
Supermetrics found smaller rises: typical CPMs 16% to 41% above September by Black Friday week 2025, depending on the sector, with the top of the apparel range up 57%. The two samples differ (biggest spenders vs a broad set of retailers), so read both as a range, not one number.
Does a higher CPM mean a higher cost per order?
Not by itself. Your cost per order is the CPM divided by 1,000, divided by your click rate, divided by your conversion rate. If more of the people who click also buy, a higher CPM can still give a cheaper order. In Lebesgue's sample, the median brand's conversion rate rose 48% in Black Friday week and its cost per purchase fell 20%, on purchases as Meta reported them.
Demand really does jump. Across the 118 stores in our data, orders per day in Black Friday week 2025 (24 November to 1 December) were a median 1.94x October, and average order value was 1.00x. People buy more often, not bigger baskets. But the middle half of stores ranged from 1.13x to 3.26x, so for some stores demand barely moved while CPMs climbed.
| October | Black Friday, buyers respond | Black Friday, they do not | |
|---|---|---|---|
| CPM | $12 | $19 | $19 |
| Click rate | 1.2% | 1.5% | 1.2% |
| Conversion rate | 2.0% | 3.0% | 2.2% |
| Cost per order | $50 | $42 | $72 |
Same CPM, very different results. Whether Black Friday is cheap or expensive for you depends on how strongly your buyers respond to the offer, and your own last year is the best guide to that.
Is Ads Manager's cost per purchase the real one?
Often it is lower than the real one. By default, Meta counts a purchase if the buyer clicked an ad in the last 7 days or saw one in the last day (Meta attribution settings), and in Black Friday week many people who see an ad were going to buy anyway. Even in a normal month, October 2026, Meta reported a median 1.12x the sales we could match to real orders, across 91 stores. Why that gap tends to grow in sale weeks is covered in why store orders do not match Meta and Google during Black Friday.
What should I do when CPMs climb?
- Judge cost per real order against your break-even, not CPM. Here is how to calculate break-even ROAS after the discount.
- Do not cut a working campaign on CPM alone. Meta says CPM may not be a good indicator when you optimize for conversions, because its system may buy pricier impressions to get cheaper results (Meta).
- Have new creative live before mid-November. In Meta's auction, a more relevant ad can beat ads with higher bids.
- Compare with your own September and last year, not with a benchmark. A "$20 CPM" means little without your country, audience and conversion rate.
- Size your budget to demand, as covered in how much to increase your Black Friday budget.
- Check again after Cyber Monday. In Lebesgue's data, CPMs fell from 71% above September on Cyber Monday to 20% above two days later.

