When you A/B test a proof point, you end up with two versions of it live at once. Meta's library makes that visible. In the 150 active monday.com ads I pulled on November 9, 2026, the first 30 texts contain the same sentence twice with different numbers: "180K+ customers" in 2 texts and "225K+ customers" in 2 others.
This post covers that pair, the rest of the copy, and what the sample cannot say.
Sample frame
| Source | Snapshot date | Filters | Ads collected | Texts analysed |
|---|---|---|---|---|
| Meta Ad Library | 2026-11-09 | page facebook.com/mondaydotcom, activeStatus: active, cap 150, no country or date filter |
150, all active | first 30 bodies in library order; 15 distinct |
| Google Ads Transparency Center | 2026-11-09 | domain monday.com, cap 150, no country or date filter |
150 | headline field read for 30 ads; copy of the 8 longest-running not available |
| Google Trends | 2026-11-09 | term "monday.com", past 12 months | n/a | top 10 and rising 10 related queries |
Both ad sources hit the cap of 150, so each is likely a slice of a bigger set. Meta here is a sign-up push (the button is "Sign up" on 149 of 150 ads); Google mixes text and video ads for the domain. So I compare those campaigns, not the two networks.
Two counts, one sentence
The template is identical and only the number changes:
"There's a reason why 180K+ customers use monday․com to efficiently manage their teams, work and processes. See why."
"There's a reason why 225K+ customers use monday․com to efficiently manage their teams, work and processes. See why."
Each version appears in 2 of the 30 texts. The library shows both as active in the same snapshot. It does not show which went live first, so I can't say one is stale. It could be an old creative nobody switched off, or a deliberate test. Either way, a reader of the two ads would see two different claims about the same company.
The other 28 texts
Of the 30 texts, only 15 are different. The mix by language, using a few marker phrases checked in a fixed order: English 14 texts, Spanish 10, Portuguese 5, French 1.
The most repeated text appears 8 times and is a Spanish line that names no feature at all:
"Todo tu trabajo en un solo lugar"
The second is English and appears 5 times:
"Finally, a work platform your team will actually love to use so they can focus on what matters most."
Four texts name an AI assistant, "monday sidekick", and they take a lighter tone than the platform ones:
"Still dodging AI implementation like it’s another budget meeting? Time to meet monday sidekick, the AI you’ll love to use."
The Meta formats are 147 DCO (dynamic creative, where Meta mixes assets), 2 video and 1 image, and the page's variants reach 11 per ad at most (variantsMax). With so many dynamic ads, repeated texts are expected: the same body is paired with many assets.
Google: the same domain, other advertisers
For monday.com, 146 of the 150 Google ads sit under "monday.com LTD". The other 4 are under two names with no obvious link to the brand: "Sponsors for Educational Opportunity Africa" (2) and "RICHFIELD ELECTRONICS LIMITED" (2). The library returned them for this domain query. I don't know why, and I'm not claiming they are impersonators; it could be a data quirk. If you track a competitor on Google, group by advertiser and look at the small rows too.
The format split is 102 text, 45 video and 3 image, and 45 ads carry a YouTube link.
Of the 8 longest-running Google ads, all are text ads, the oldest first shown on October 25, 2021 (1,840 days), and all were still shown on the snapshot date. The library left their headline and description empty, so I can't say what they say. Their first-shown dates run from 2021 to 2023 rather than clustering on one day, so I don't see the batch-date artifact here.
What people search for
Trends gives "monday.com" an average interest of 64 over the year, a peak of 100 on March 1, 2026, and 36 now, a change of -30.5% over the window in the collector's output. The top related queries are mostly price and product questions: "monday.com price", "what is monday.com", "monday.com ai", "monday.com review". Among rising queries are "monday.com layoffs" (+300%), "jira" (+130%) and "asana" (+120%). Trends gives no volumes, so this tells you what people type, not how many.
What this data can't tell you
- A capped slice and a small text sample. Both ad sources stopped at 150, and every count of copy comes from 30 Meta texts, 15 of them distinct.
- No dates for the two counts. The library doesn't say when each version started. The eight longest-running Meta ads all started on May 12, 2026, so I don't read their order as a ranking either.
- No spend, reach or results. Nothing here says which customer count converts, or how monday.com splits budget across languages. Google's copy for the oldest ads is empty, and there are no LinkedIn ads in this snapshot (the LinkedIn run returned 0).
3 things to test this week
- List every number in your live ads. Search your own active copy for customer counts, ratings and discounts, and check that no two ads disagree.
- Test the proof point on its own. If you run a "180K+" and a "225K+" version, give each the same image and audience so the number is the only difference.
- Check the small rows in a competitor's Google list. Group a domain's ads by advertiser before you read anything into the totals.
Repeating this for a competitor
I built these scrapers. The Facebook Ads Library Scraper takes a Facebook page URL and returns body text, CTA, format, variants and start date per ad, and the Google Ads Transparency Scraper returns format, advertiser and dates for a domain. Both are no-code forms with CSV or Excel export, free to try on the Apify free plan, and a run of a few hundred ads costs well under a dollar.
Not affiliated with monday.com; all data is from public ad libraries.
If two versions of your best proof point were live right now, which one would you switch off first, and how would you decide?
Top comments (0)