Holiday Ad Creative Fatigue: How Many New Ads to Have Ready for November and December 2026

How many new Meta ads do you need for November and December 2026?

About 37 new ads between 2 November 2026 and 3 January 2027 if your brand spends $10,000–$50,000 a month on Meta, at the average pace in Motion's holiday-window benchmark, or about 72 at that tier's top-quartile pace. Variants count, so many can be cut from footage you already own. At the average pace those nine weeks surface only about 0.4 to 2.4 new winning ads, so December runs on ads you found before Thanksgiving.

The holiday ad plan at a glance:

• New ads, 2 Nov – 3 Jan: about 25 under $10,000 a month, about 37 at $10,000–$50,000, and about 72 at that tier's top-quartile pace.
• New winners in those nine weeks: about 0.4 to 2.4 at the $10,000–$50,000 average pace.
• Black Friday ad sets live: by Thursday 19 November 2026.
• Post-cutoff ad sets live: by Thursday 10 December 2026, ahead of the USPS send-by dates of 17–19 December.
• Q4 price premium on Meta: 13% in 2022 and 17% in 2023 by Gupta Media's measure, and about 32% in 2024 by our arithmetic.

Both load dates are Dream Aspect's reading of Meta's learning-phase rule, not Meta's dates. The counts come from Motion, which sells creative analytics, and we sell production, so both of us benefit when you make more ads. The plan covers Meta only, using other advertisers' averages from one season; where your account disagrees, your numbers win.

How many new ads does a brand need between 2 November 2026 and 3 January 2027?

Motion's Creative Benchmarks 2026 counts new ads launched each week by monthly Meta spend, across 578,750 creatives, 6,015 accounts and $1.29 billion of spend from 1 September 2025 to 1 January 2026. Spread over the nine weeks from 2 November 2026 to 3 January 2027 (our arithmetic), Motion's average and top-quartile weekly rates give these totals:

• Under $10,000 a month: 2.8 a week, so about 25 on average and about 43 at top-quartile pace.
• $10,000–$50,000: 4.1 a week, so about 37 on average and about 72 at top-quartile pace.
• $50,000–$200,000: 6.6 a week, so about 59 on average and about 143 at top-quartile pace.
• $200,000–$1 million: 11.2 a week, so about 101 on average and about 280 at top-quartile pace.
• Over $1 million: 18.8 a week, so about 169 on average and about 491 at top-quartile pace.

Motion's window ran through Black Friday, and Motion warns that "patterns may look different in steady-state periods", so these are holiday-window averages, not peak-week rates; don't multiply them up again. Variants count, too: Motion counts "duplicates and variations" as distinct "when launched as separate ads", so 37 new ads means 37 hooks, cuts, captions and formats, not 37 productions.

Category moves the number a little. Motion's vertical sample puts the named consumer verticals at 4 to 5 new ads a week in the $10,000–$50,000 tier (other verticals at 3), and at 8 to 12 in the $50,000–$200,000 tier against its all-vertical 6.6. Motion says those values were not fully reconstructed, so treat them as an order of magnitude only. For October's count, see our Halloween production window piece; outside the holidays, see how many videos a month a brand needs.

How many new winning ads will the holiday weeks produce?

Between about 0.4 and 2.4 for a $10,000–$50,000-a-month account at the average pace, and Motion's own pages disagree about which end is right. Motion counts a winner as an ad spending at least 10 times the account's median creative spend and at least $500. Three of its figures give a nine-week yield, by our arithmetic at 4.1 new ads a week:

• Tier hit rate of 6.4%, meaning winners as a share of an account's new ads: about 2.4 winners.
• Motion's rule of thumb: "An advertiser testing 4 ads per week will surface roughly 0.2 winners per week on average." That gives about 1.8.
• Winners per month on Motion's top-quartile page: 0.2 for all accounts in the tier, which gives about 0.4.

The third is four to six times smaller than the first two, and Motion doesn't explain the gap. Plan December on the low end: assume the peak weeks find nothing new, and protect the ads proven before Thanksgiving rather than replace them. Don't size production from the low figure, though, because it would sell you the most production; the counts here come from Motion's volume rates.

Doubling the pace buys little. At about 72 ads, Motion's top-quartile column gives about 1.0 winner over the nine weeks and the hit rate about 4.6, because top-quartile accounts "aren't producing dramatically higher hit rates". About half of new ads won't last a month either: Motion's portfolio breakdown found 49.3% of the tier's new ads switched off before day 28, and 49.3–53.9% at every tier, or about 18 of 37. Winners are defined by spend, and Motion warns that "A winner by spend can still be ROAS-negative." If your October tests found nothing, we would rather you spent a December production budget in January.

What is the week-by-week refresh schedule for November and December 2026?

Nine weeks, three jobs: test from 2 to 22 November, protect what works from 23 November to 6 December, and change the message twice in December. At the $10,000–$50,000 average pace, Dream Aspect's split puts 16 new ads before Thanksgiving week, 6 across the two peak weeks, 10 in the two gifting weeks and 5 at the turn of the year, 37 in all. The weekly split is our planning choice; the total is Motion's.

• 2–8 Nov, 5 new ads: test new concepts in new ad sets, in what Meta's holiday guidance calls "The discovery phase (October into early November)".
• 9–15 Nov, 5 new ads: keep testing and shortlist Black Friday ads.
• 16–22 Nov, 6 new ads: load the Black Friday ad sets so they are live by Thursday 19 November, then leave them alone.
• 23–29 Nov, 2 new ads: Thanksgiving falls on Thursday 26 and Black Friday on Friday 27. Hold, replace tired ads only as new ad sets, and edit nothing carrying spend.
• 30 Nov – 6 Dec, 4 new ads: Cyber Monday on 30 November, then gifting ad sets.
• 7–13 Dec, 8 new ads: test gifting variants and load the post-cutoff ad sets so they are live by Thursday 10 December.
• 14–20 Dec, 2 new ads: the USPS send-by dates fall on 17, 18 and 19 December. On your cutoff, switch off delivery promises and add gift-card variants.
• 21–27 Dec, 1 new ad: Christmas Day falls on Friday 25. Hold.
• 28 Dec – 3 Jan, 4 new ads: New Year's Day falls on Friday 1 January 2027. Run self-purchase concepts and retarget add-to-cart browsers.

Black Friday and Cyber Monday are counted from the OPM federal holiday calendar, which puts Thanksgiving on 26 November 2026. At top-quartile pace every week roughly doubles, to about 72 in total. For a small account, Meta warns that "When you create many ads and ad sets, the delivery system learns less about each ad and ad set than when you create fewer ads and ad sets", so open one new ad set per job, not one per ad. Meta's help page also tells advertisers to "Avoid high ad volumes", the clearest published argument for fewer new ads than Motion's averages suggest.

When must Black Friday ads be live, and can you add ads during the peak?

Live by Thursday 19 November 2026, a week before Thanksgiving. Meta's learning-phase page says an ad set usually leaves learning "after about 50 results in the week after the ad set’s last significant edit", so 19 November is Dream Aspect's reading, not Meta's date: seven days to collect those 50 results before Thanksgiving. Meta's holiday page points the same way: "Advertisers who started in October have weeks of optimization behind them once Black Friday rolls around."

Once the peak starts, new creative goes in a new ad set. As of 5 October 2026, Meta's list of significant edits includes "Any change to ad creative" and "Adding a new ad to your ad set". Budget changes may or may not count, "depending on the magnitude of the change": $100 to $101 is not likely to, while $100 to $1,000 may. Both routes cost a learning phase: ad sets in learning "are less stable and usually have a higher CPA" (cost per acquisition). Adding an ad resets the ad set carrying your peak spend; a new ad set learns alone and, under Advantage+ campaign budget, "won't cause other ad sets within the same campaign to reenter the learning phase."

Dream Aspect's reading is that the new ad set is the smaller risk. In the peak weeks, add only replacements, and set the campaign budget during the load week, because adjusting it later "might cause multiple ad sets within the campaign to reenter the learning phase". One limit applies however early you load: an ad set that cannot reach about 50 results a week shows a Learning limited status. Footage timing for the same campaign is in our Black Friday content calendar.

How much do Meta CPMs go up in Q4, and does that change how many ads you need?

Meta CPMs (cost per thousand impressions) rose 13% to 17% across the quarter in Gupta Media's 2022 and 2023 figures, and much more around Black Friday. The price changes when you test, not how many ads you need. A test needs roughly the same number of results in any week, so reaching a verdict in Thanksgiving week costs more. That is why the schedule puts 16 of the 37 new ads before 23 November and only 2 in Thanksgiving week: test in the cheaper weeks, and spend the expensive ones on proven ads.

Gupta Media measures the premium by "comparing the CPMs in Q4 to CPMs in Q1 through Q3": on Meta, "that premium was only 13%" in 2022 and "the 2023 premium was 17%." Gupta gives no 2024 premium. Its monthly table puts October–December 2024 about 32% above January–September by our unweighted arithmetic, though the same method gives 14% for 2023 against Gupta's 17%, so it runs low. Gupta's figures come from its own clients' accounts, and its 2024 weeks stand in for 2026.

In Gupta Media's 2024 data, Meta's CPM averaged $9.61 in November and $10.83 in December, against $8.30 in September and $8.27 in January 2025. Single weeks and days were steeper:

• ISO week 48 (25 November – 1 December, Thanksgiving and Black Friday): $13.42.
• ISO week 49 (Cyber Monday): $12.53.
• ISO week 50 (9–15 December): $11.03.
• Most expensive days: Cyber Monday at $17.70 and Black Friday at $16.85.

Thanksgiving and Cyber Monday fall in the same ISO weeks in 2026 (our check), and Gupta found that "the weeks directly before and after BFCM saw CPMs between 12% and 27% cheaper". Prices fell on average in 2025, but not for everyone. Tinuiti reported Meta CPMs down 7% year over year in Q4 2025, yet "44% of advertisers studied saw CPM rise" on Facebook, and from the Saturday after Thanksgiving to Cyber Monday "CPM rose for 37% of the Meta advertisers studied" (Tinuiti's recap). Skai put paid-social "CPMs down 6%".

How do you know a holiday ad is fatigued, and why might Meta not tell you?

Meta flags fatigue on cost. Its creative fatigue page shows Creative limited when cost per result rises above your past ads, and Creative fatigue once it reaches twice as much. But Meta also says "This feature is only available for ad sets with one creative", with exceptions for Advantage+ catalog ads, dynamic creative and Advantage+ app campaigns. Meta's holiday page says "Adding 3 or more creative formats can make the difference", and a holiday ad set running three or four ads is outside the feature, so run Meta's check by hand each week: compare each ad set's cost per result with your past ads.

Meta's fix is a new image or video "materially different from the original creative"; a new caption on the same footage may not qualify. Meta's analytics team ran "a simple two cell split test" across "roughly 26 thousand cases" and wrote in 2023 that it "suggests that adding in new creative into fatigued ad sets has an overall causal improvement in conversion rates." The post states no season and measured conversion rate, not cost or learning status. Outside the peak, follow it. From 23 November to 6 December, Dream Aspect's reading favours a new ad set, which starts its own learning but leaves the one carrying peak spend undisturbed.

New creative isn't the only remedy. Meta's fatigue page lists "Expand your audience", and the same 2023 analytics post says "Alternatively to refreshing creative, targeting new audiences could also resolve this problem." If the tired ad was a winner, try a new audience first.

What should change in your ads after the shipping cutoff?

The promise. After your last shipping day, ads promising delivery by Christmas are wrong, and the job turns to gift cards and self-purchase. As of 5 October 2026, USPS recommends sending by Thursday 17 December (Ground Advantage, First-Class Mail), Friday 18 December (Priority Mail) or Saturday 19 December (Priority Mail Express) for expected delivery before 25 December in the Lower 48. Those are recommendations, not guarantees, and your cutoff is your own: take dispatch time off the USPS dates, while gift cards, digital products and store pickup run later.

Load the replacement ad sets by Thursday 10 December, then switch off the delivery promises on your cutoff. Rewriting a live gifting ad is a significant edit that sends it back into learning, and by Dream Aspect's reading of the 50-results rule, a new ad set launched on 17–19 December could still be learning on Christmas Eve. Loaded by 10 December, it learns while gifting ads still sell, so its messages must hold on both sides of the cutoff: gift cards, self-purchase, and last-chance offers that promise no delivery date.

Meta's holiday phases describe the same switch. After December's gifting phase comes "The fresh start phase (Late December / New Year)", which Meta says "is all about the self-purchase", with "aspirational reels and retargeting the pool of holiday browsers who added to cart but didn't convert." The demand is still there. In a 2025 survey of 8,005 adults, the National Retail Federation found that "As of early December, consumers had completed just over half (51%) of their holiday shopping on average", and that "70% of consumers plan to shop the week immediately after Dec. 25", with gift cards (26%) among the top reasons. Those are stated intentions from 2025, not a 2026 forecast.

Do the extra holiday ads need a new shoot?

Mostly no, if your library has the angles. The extra ads can be new hooks, captions, cuts and formats from footage you already have, and Meta's holiday guidance puts the order plainly: "Repurpose first, produce second, refresh throughout". Motion's data agrees, which cuts against a production company like Dream Aspect. Its asset types by hit rate, "In approximate rank order", begin "Text only, Product image with text, Lifestyle-product image, UGC, High production", and Motion notes that "A high-production video takes days or weeks." Its reading: "High production is sometimes more useful as a second stage than as a first test."

Meta goes further, without citing a source: "A Reel shot on your phone outperforms most polished brand videos". It is the strongest published line against what we sell. A new shoot earns its place when fatigue calls for materially different creative and your library is spent, when October found no winning concept, or for new concepts such as fresh-start creative. Meta's line "One shoot day with a real customer gives you testimonial video cuts for the entire season" describes the shoot worth having.

If that customer, or a UGC creator, is paid or given free product, disclose it in the ad. Under the FTC's Endorsement Guides (16 CFR 255.5), a connection the audience would not expect "must be disclosed clearly and conspicuously".

If you do shoot with Dream Aspect, our video marketing page says: "Full content delivery typically takes 7-10 days after the content shoot day." It doesn't say whether paid cutdowns or your approval round are included, so ask before counting back from it. As of 5 October 2026, our UGC page starts pricing at $350 per video with a minimum of five videos per batch, so $1,750 at the floor. The same page publishes both "The production timeline is 14 to 30 days" and "delivered as short as 7 days" without reconciling them, so we give no UGC order date; if you are quoted one, get it in writing.

Planning holiday ad creative with a Los Angeles team?

Dream Aspect is a Los Angeles video production company that makes ad creative for Meta and TikTok. Our ads page sets out a "Test First. Pay Only If You Scale" model: "Each edit is packaged into an ad bundle designed for A/B testing", you "collect real performance data license-free", and "Once winning ads are validated and you decide to scale", "we get paid per licensed ad". It publishes no turnaround for ad bundles, so ask for a date in writing. Tell us which week you are planning for, and we will tell you what can still be made in time.

Want to know what a batch of new holiday ad creative would cost? The Dream Aspect pricing calculator shows the price on the page: choose a one-off shoot or a monthly plan, then add or remove deliverables such as a shoot brief, priority delivery, CGI or UGC.

Build your quote on the pricing page →

Prefer to talk it through? Talk to us → or book a 15-minute call from the pricing page.