We fact-checked 43 brands from “best Black Friday ads” roundups against Meta’s Ad Library. Only 22 check out.

Only 51% of “best Black Friday ads” survive a fact-check. We collected 48 candidate brands from five kinds of public sources — evergreen swipe-file roundups, post-Black-Friday editorial recaps, an email platform’s client case studies, and a listicle published three days before the sale — and checked 43 of them against Meta’s Ad Library, brand by brand, using each brand’s verified official Page rather than keyword search. 22 brands had verifiable Black Friday ads running in November 2025. 20 showed nothing. One was inconclusive. That is a 51% verification rate for the industry’s most-cited Black Friday examples. To be precise about what “nothing” means: a no-show is not proof a brand skipped Black Friday — several run the sale entirely over email and SMS, in-app, or under wording like “holiday sale” that a “black friday” search will never catch. It is proof that no Black Friday Meta ad could be verified where the roundups implied one existed.

Black Friday on Meta is a six-week war that starts in October. Of the 22 verified campaigns, 20 have clean flight windows. The earliest went live on October 29 — a full month before Black Friday, which fell on November 28, 2025 — and six of the twenty (30%) were already running by October 31. The median start date was November 6, three weeks and a day before the sale itself. At the other end, 16 of the 20 flights ran past Cyber Monday, and the longest — a jewelry brand’s gift-with-purchase campaign — ran from November 6 to December 27, a 51-day flight. Exactly one brand in the dataset waited until Black Friday week itself: Apple, whose flight ran November 28 to December 3. If your Black Friday media plan switches on the week of the sale, you are entering an auction where most of the verified winners have already been warming audiences for three to four weeks.

Verified discounts ran from a flat 23% to 70%, with a median peak of 48%. Nine of the 22 campaigns carried a clean percentage offer we could verify in the ad copy itself. Sorted, the peaks run 23, 25, 30, 40, 48, 50, 50, 70, 70 — a median of 48%. Both extremes are instructive. The 23% is The Ordinary’s deliberately modest “Slowvember”: a flat −23% held for six weeks with no escalation, positioned against Black Friday urgency itself. Both 70s were the loudest wave of a multi-wave structure, not a static offer. Three campaigns escalated the discount across waves — 25%→70%, 20%→25%, and one four-wave sequence that opened at 70% off — and at least five layered a gift-with-purchase on top of, or instead of, a percentage: from a free lip product at a £100 spend threshold to an Apple gift card worth up to €200.

Creative volume was polarized — and the strongest mechanics shipped the fewest ads. Black-Friday-tagged ad counts in the verified set run from roughly 21 units (Glossier) to roughly 5,600 (SHEIN), with True Classic near 960, Tropicfeel near 980, and Jet2holidays near 1,300 among the volume plays. But volume did not track with inventiveness — if anything, it tracked against it. The campaigns with the most distinctive mechanics were small: Glossier’s fake Notes-app checklist ran behind about 21 units, The Ordinary’s anti-rush Slowvember behind about 84, BabySam’s two-livestream funnel behind about 91, and Nordic Nest’s nine-language pun launch behind about 29 findable creatives. The volume plays, meanwhile, scaled one idea: True Classic duplicated a single creative into as many as 24–25 versions, and Jet2holidays templated one copy line across six-plus resort pairings, generating variants dynamically. A mechanic is what you run when you can’t outspend; duplication is what you run when you can.

The biggest retailers in the roundups are missing from Meta entirely. Amazon, Walmart, Target, and Samsung — all cited in post-Black-Friday editorial recaps — returned zero verifiable Black Friday ads on their official Pages, checked in both US and global views. Zara, Nike, and Sephora’s US Page came up empty too, even though Zara’s Page carries heavy always-on catalog spend. The pattern is consistent: brands whose Black Friday play was described as app-exclusive drops, livestreams, member early access, or UGC waves simply don’t show a matching paid Meta creative. The verified list skews hard toward DTC and performance-led brands — beauty, apparel, home and kitchen, the categories that acquire customers on Meta ads. The practical read: if you’re benchmarking your Black Friday Meta plan against what Amazon or Nike supposedly “did,” you’re benchmarking against campaigns that, on this channel, never existed.

Keyword-searching the Ad Library is where most “verification” quietly dies. Brand names that double as common phrases drown: searching “True Classic” or “Great Jones” returns more than 50,000 results, with car clubs, clinics, and politicians on page one and zero relevant ads. So every check in this study used the same three-step procedure:

1. 1.Lock the official advertiser, not the keyword. Type the brand name into the Ad Library search box and select the blue-check official Page from the Advertisers dropdown — never press enter on the raw keyword. This scopes all results to view_all_page_id=<official page>, cutting out fan pages, affiliates, and counterfeit stores (a search for one whitening brand surfaces pages of Arabic-language reseller clones before the real account).

2. 2.Filter within the locked Page by adding black friday (and where relevant cyber monday) to the query — this isolates genuine Black Friday copy inside catalogs of thousands of ads.

3. 3.Set active_status=all. The default view shows only currently active ads; the November 2025 flights are long inactive and invisible without this switch.

Two traps are worth naming because they generate false no-shows. Acquisitions rename Pages: Normatec shows nothing under “Normatec” — its ads live under parent company Hyperice, and we only confirmed it after checking the ownership trail. Multi-region brands split Pages: Sephora runs 20+ country Pages; we checked the US Page and found nothing, which says nothing about the other 20 — so it stays a no-show with an asterisk, not a verdict.

Hit rate by source type

* One further brand in the recap layer (REI) was deliberately skipped: its featured campaign was an anti-Black-Friday store closure, so there is nothing to verify. Gymshark and SHEIN each appear in two source layers and are counted once each in the deduplicated totals (43 checked / 22 confirmed / 20 no-show / 1 inconclusive).

Three honest annotations on that table:

• The swipe-file roundups verify well.Foreplay’s picks confirmed at 70% and Trendtrack’s at 89% — the caveat for both is vintage, not accuracy: these are evergreen “best Black Friday ads” pages that predate BFCM 2025, so confirming them meant confirming each brand re-ran a matching campaign in November 2025, which most did.
• The 0% row is not an indictment. The email-platform layer is a set of email/SMS case studies; five of the six write-ups describe email-only playbooks and never claim a Meta campaign. Zero Meta hits confirmsthose write-ups rather than contradicting them. The sixth — the only one that mentioned Meta ads, as lead-gen forms — is our single inconclusive, because lead-form copy rarely contains the phrase “black friday” at all.
• The weakest verification layer is the editorial recaps, and mostly because of who they cover: mega-retailers and brand-marketing giants whose Black Friday, per the recaps’ own descriptions, ran on apps, livestreams, and organic social rather than paid Meta creative.

First, an honest split before the highlight reel: on a close read, 11 of the 22 verified campaigns had no distinct mechanic at all — the copy was a discount number and a product shot, competently executed at scale. Of the remaining 11, six mechanics survived a second “would you actually steal this” review and became full teardowns (next section). Verification tells you a campaign was real; it doesn’t make it interesting.

Timing

Discounts

Creative volume

Mechanic taxonomy

At least five campaigns ran verified creator/branded-content ads for the same sale concurrently with house creative — properly tagged partnership posts, not disguised UGC.