cold-iq Meta Ad Library Search
Find brands actively running video ads on Meta — a strong signal of marketing budget + digital maturity — and turn them into net-new accounts for DTC/e-commerce clients.
This replaces any local Playwright/SQLite Meta scraper with ColdIQ's async Meta Ads endpoint — no local browser, no database to maintain. The valuable part — the 135-term search strategy — is preserved below as the query set you submit.
ColdIQ Marketplace Endpoints
| Task | Method | Path | Credits | Endpoint ID | Notes |
|---|---|---|---|---|---|
| Start a Meta Ads search | POST | /v1/meta-ads/search |
flat | meta_ads.search.create |
Async; returns jobId |
| Get search results | GET | /v1/meta-ads/search/{jobId} |
free | meta_ads.search.result |
202 running, 200 done |
How to run
For each search term, submit a job and poll for results (see resources/async-job-pattern.md):
- Submit one keyword/term (or URL).
→ POST
/v1/meta-ads/search· flat ·meta_ads.search.create{ "query": "shop now free shipping", "country": "US" } - Poll until done, then read the ads.
→ GET
/v1/meta-ads/search/{jobId}· free ·meta_ads.search.result
Iterate over the term set below (one job per term), accumulate brands, then de-dup.
Search strategy — 135 terms across 4 phases
Key insight: universal DTC ad language beats niche product keywords. A jewelry, baby, and supplements brand all run "shop now free shipping". A brand with 10+ video ads = real ad operation (budget + creative team + performance-marketing leadership).
- Phase 1 — Universal DTC video hooks (35, US): "shop now free shipping", "use code save", "subscribe and save", "limited time offer", "before and after results", "real customers real results", "try risk free", "as seen on shark tank", "best selling", "clinically proven results", "watch what happens", "I tried this", "honest review", "game changer product", "why everyone is switching", "claim your free sample", "exclusive offer today only", "save 20 percent", "first order free", "bundle and save", "selling out fast" …
- Phase 2 — Gap verticals (47, US): Baby & Parenting (baby essentials, nursery must haves, toddler snacks, diaper subscription, postpartum, maternity wear, baby food organic, kids vitamin gummies…); Jewelry & Accessories (fine jewelry, engagement ring, personalized jewelry, dainty jewelry, birthstone, mens jewelry…); Wine & Spirits (wine delivery, craft spirits, tequila, wine club, natural wine, mezcal, hard seltzer, non alcoholic spirits…); Dairy & Alt-Dairy (oat milk, plant based milk, greek yogurt, cheese subscription, grass-fed butter, kefir probiotic…).
- Phase 3 — Brand-adjacent signals (33, US): subscription box, curated subscription, auto ship, member exclusive; just launched new brand, DTC brand launch, founder story, small batch artisan; free returns, BOGO, flash sale, new arrivals, trending now; valentine/mothers day/fathers day/ holiday gift guide/summer essentials.
- Phase 4 — Canada (20, CA): rerun the top Phase 1–3 terms targeting Canada.
After scraping — next steps
- De-dup discovered brands (see list-dedup).
- Resolve & enrich each brand: domain → company LinkedIn → firmographics (see coldiq-search-enrich, contact-enrichment).
- Score against the client ICP (see tam-scoring).
- Pull contacts for qualified companies, then load to a campaign (see instantly-api).
Quality notes
- Filter to brands with 10+ video ads for serious ad operations; lower the threshold (5+) for smaller verticals.
- Advertiser names from Meta need cleanup before enrichment (strip "LLC", emoji, store suffixes).
- A brand with many ads but no resolvable domain is usually a marketplace seller — deprioritize.