Ad Health Index, November 2026: how many active e-commerce ads in Turkey lead to a page you can't buy from?
Issue 1 · November 2026 · Data: CSV · Türkçe
About one in three active e-commerce ads we checked in Turkey (22 of 65, 34%) sent people to a page where the product could not be bought: out of stock, a broken link (404), or a redirect to a category page. One week later, 21 of those 22 pages were still in the same state.
Numbers (65 unique ad destination pages, 50 brands)
| Result | Pages | Share | |
|---|---|---|---|
| Purchasable | 38 | 58% | |
| Partial stock (some sizes/colors gone) | 5 | 8% | |
| Out of stock | 10 | 15% | |
| Broken link (404 or redirect to category) | 12 | 18% | |
| Not purchasable (out of stock + broken) | 22 | 34% (95% CI 23–46%) |
- 18 of 50 brands (36%) had at least one ad pointing to a page that could not be bought.
- Persistence: 21 of 22 problem pages were unchanged a week after the first check.
- By category (small samples, directional only): fashion and footwear 16/32 (50%); all other categories 6/33 (18%).
Method
- Source: Meta Ad Library, Turkey, active ads. Keyword searches produced 197 destination addresses; app, streaming, tour, school and ticket links and marketplace, Instagram and WhatsApp addresses were excluded, leaving 183.
- Sample: the first 20 brands in order, then 30 more chosen systematically from the remaining 117 (50 brands, 65 unique pages after removing one duplicated page).
- Checks: each page was opened in a browser and checked for a 404 or redirect to a category page, the stock status in the product's structured data (InStock / OutOfStock), per-size and per-color stock on Shopify stores, total stock on Ticimax stores, and page text such as "sold out" or "notify me when available".
- Timing: first check 28–29 October 2026; second check 5 November 2026 on the same pages (both gave 22 of 65).
Limits
- "Active" in the Ad Library does not mean the ad is spending. Long-running ads may be over-represented, and keyword choice shapes the sample.
- Whether each ad was still active at the second check was not verified; the same pages were reopened.
- Stock can change within a day; the two checks are one week apart.
- Small sample: margin of error roughly ±11 points. Category splits are indicative only.
- One site could not be measured (redirect issue) and was excluded.
Data
Aggregate tables (no brand names, no page addresses): ad-health-index-2026-11.csv. Licence: CC BY 4.0 — free to reuse with attribution.
Cite as
AI Trade, “Ad Health Index — Turkey, November 2026 (issue 1)”, aitradeai.net/ad-health-index