Ad IntelligenceJewelry & Watches
Updated September 2026
We track one trade and nothing else. Every Google and Meta ad running for 142 jewelry and watch retailers, captured on September 20, 2026 and compared with the same panel in April 2026. Every creative was read one at a time against a fixed vocabulary. No business is named, there is no PDF to download, and there is nothing to sign up for.
Almost nobody in this data asks for the appointment.See where your own ads sit
The quarter in four figures.
Counts of live ads, not spend — neither ad library reports what anyone spent. Movement figures cover the 93 retailers present in both the April and September captures.
What each retailer did with its own advertising.
Ranked by how far a retailer’s own advertising moved relative to where it started, not by how many ads it runs, because nothing-to-a-dozen is a decision and 400-to-430 is noise.
- Turned it on 2
- Doubled or more 4
- Grew 24
- Held steady 44
- Still dark 1
- Cut back 8
- Cut by half or more 9
- Turned it off 1
Find your category.
Pick the size and the kind of business you run. Categories with fewer than 4 retailers are not published, so no business can be identified by elimination.
Single store
66 retailers, 50 of them advertising
- Own ads, typical
- 27
- On Google
- 24
- On Meta
- 3
- From a maker’s account
- 0
- Running both platforms
- 66%
- Creative that makes a claim
- 70%
What the ads sell
The claim they lead with
Google and Meta do different jobs.
Every live creative from a retailer’s own account, read one at a time against a fixed vocabulary. Pick what you want to compare.
Meta
What the top tenth of your class does.
Each kind of business, and the same figures for the tenth of it running the most of its own advertising. A ranking of volume, not of results — nothing in either ad library says whether an ad sold anything. No business is named, and a class too small to show a top tenth without identifying its members is left out.
Rolex dealer
64 on the panel · 52 advertising
Measure The class Top 5 Own ads, typical 35 312 On Google 27 285 On Meta 12 18 Started since April 22 116 Runs both platforms 92% 100% Creative that makes a claim 67% 75% Its top tenth leads with authenticity, inspected and expertise, service.
Watch dealer, no Rolex
35 on the panel · 31 advertising
Measure The class Top 4 Own ads, typical 33 145 On Google 27 133 On Meta 4 24 Started since April 7 55 Runs both platforms 71% 75% Creative that makes a claim 70% 36% Its top tenth leads with heritage, long established and selection, in stock.
Jeweler, no watch brands
26 on the panel · 18 advertising
Measure The class Top 4 Own ads, typical 37 139 On Google 29 129 On Meta 4 11 Started since April 16 97 Runs both platforms 72% 100% Creative that makes a claim 57% 57% Its top tenth leads with selection, in stock and price or value.
Pre-owned specialist
6 on the panel · 4 advertising
class too small to show a top tenth without identifying its members
Online-first
5 on the panel · 4 advertising
class too small to show a top tenth without identifying its members
How to read these numbers.
- “Live” is a recency rule, not a snapshot: an ad whose last recorded impression falls within 14 days of the day we looked at that retailer, or 40 days on Meta. At 7, 21 and 30 days the direction holds but the size moves.
- Neither ad library reports what anyone spent. Every figure on this page is a count of live ads, so a retailer can add creative while cutting its budget and look like growth.
- Google covers the full quarter. Meta was captured once, in September, so a campaign that ran only in July or August is invisible here.
- The two captures are not equally fresh: April was collected over two weeks, September in a single day.
- Google search ads carry copy but no picture, so anything about imagery covers image and video ads only.
- Populations differ by question: 155 retailers targeted, 142 observed this quarter, 136 in the category figures, and 93 present in both captures — the only ones in any movement figure.
- A retailer who funds an ad and is later reimbursed by a brand looks identical to one who is not. That is invisible in both libraries.
- Five accounts hit the ad library’s per-search ceiling, so their totals are floors. The same five hit it in April, so the comparison holds.
Google captured as 2026-Q3, Meta as 2026-09, both compared with April 2026. Categories holding fewer than 4 retailers are not published.
Get told when this changes
We refresh this page when a new capture lands. Leave your email and we will tell you what moved, in a short note. No other mail, and you can unsubscribe at any time.