Newsletter Sponsor Finder — Substack

Scan any list of Substack newsletters and extract who sponsors them: advertiser name and domain, sponsorship format, how often they appear, in which newsletters, and when they were last seen.

Newsletter scrapers give you posts. Nobody gives you the sponsors — yet that is the data two groups pay for:

• Newsletter operators selling ad space: every brand already sponsoring a newsletter in your niche is a warm prospect for yours — they have budget and proven intent.
• Brands & media buyers: see where competitors spend, which advertisers are active in a niche, and which newsletters actually run sponsorships before you pitch or buy.

What it does

1. Scrapes the recent posts (full content) of the newsletters you list — works with *.substack.com and custom domains.
2. Detects sponsorship sections via editorial markers: brought to you by, sponsored by, today's sponsor, presented by, in partnership with, together with...
3. Extracts the advertiser behind each sponsorship (link + brand anchor), ignoring CDN assets, social links, self-links and reference sites, and bounding the sponsor section so editorial links don't pollute the data.
4. Aggregates everything into one item per sponsor across all scanned newsletters, sorted by activity.

Input example

{
  "newsletterUrls": [
    "https://newsletter.pragmaticengineer.com",
    "https://www.lennysnewsletter.com"
  ],
  "maxPostsPerNewsletter": 20
}

The defaults work as-is: just list the newsletters of your niche.

Output example

{
  "type": "sponsor",
  "sponsorName": "WorkOS",
  "sponsorDomain": "workos.com",
  "sponsorUrl": "https://workos.com/",
  "totalMentions": 7,
  "newslettersSponsored": 2,
  "newsletters": [
    { "newsletterUrl": "https://newsletter.pragmaticengineer.com", "publicationName": "The Pragmatic Engineer", "subscriberCount": "1,100,000", "mentions": 5, "lastSeenAt": "2026-07-01T16:57:46.499Z" },
    { "newsletterUrl": "https://www.lennysnewsletter.com", "publicationName": "Lenny's Newsletter", "subscriberCount": "1,000,000", "mentions": 2, "lastSeenAt": "2026-06-20T12:00:00.000Z" }
  ],
  "markers": ["brought to you by"],
  "firstSeenAt": "2026-04-02T16:00:00.000Z",
  "lastSeenAt": "2026-07-01T16:57:46.499Z",
  "sampleContext": "Antithesis is autonomous testing... WorkOS — auth for B2B SaaS..."
}

A SUMMARY record in the run's key-value store gives the counts: newsletters scanned, posts scanned, posts with sponsors, sponsors found.

How to use it

• Selling ad space: scan the 5-10 newsletters closest to yours, sort sponsors by lastSeenAt — active advertisers first — and you have your outreach list, with proof of where they already spend.
• Competitive intel: scan the newsletters of your category monthly (schedule + postedAfter) and diff which sponsors appear and disappear.
• Frequency = budget signal: a sponsor with totalMentions 6+ across recent posts is running a campaign, not a one-off test.

Fair use & data

• Uses only public newsletter posts (no login, paywalled content is not accessed).
• Outputs company-level advertiser data only — no author or reader personal data.
• Inputs are hard-capped (10 newsletters, 100 posts each) so runs stay bounded and predictable.

FAQ

Does it work on beehiiv/Ghost newsletters? v1 covers Substack (including custom domains). Other platforms are on the roadmap.

Can detection be wrong? Sponsorship is editorial text, so heuristics apply. Each sponsor item includes sampleContext and per-post provenance so you can verify in one click.

How far back can it look? Up to 100 posts per newsletter, optionally bounded with postedAfter.