DEV Community

Kata Omel
Kata Omel

Posted on Edited on

We sell B2B lead lists where every row has an evidence link — here's the free open-source pipeline that builds them

15–25% of emails on a bought B2B list bounce. Nobody can tell you where a record came from, and "verified" usually means "we ran a regex". If you've ever bought leads, you know the drill.

We flipped it: every row carries its source URL, verification method, and check date. Dead row = replacement or cash back within 24h — real money, not "account credits" (UpLead-style).

The collection pipeline is open source: jd-play-scraper — stdlib Python, no API keys, ~4 min for a full run. Last run: 2,662 outbound-sales roles across 214 companies, re-scanned daily.

The signal: companies hiring SDRs are buying pipeline now

A company posting an SDR/BDR role has (a) approved outbound budget and (b) a pipeline gap for the ~60–90 days until the hire ramps. The scanner reads public ATS feeds (Greenhouse / Lever / Ashby) — the same JSON their careers widgets use. Job pages, not scraped profiles.

# the whole method, simplified
def greenhouse_roles(slug):
    url = f"https://boards-api.greenhouse.io/v1/boards/{slug}/jobs"
    return [j for j in fetch(url)["jobs"] if OUTBOUND_RE.search(j["title"])]
Enter fullscreen mode Exit fullscreen mode

How each row gets verified (no black box)

Every row gets a method tag instead of a vague "verified":

  • mx-domain — the mailbox domain actually accepts mail (stdlib socket + DNS MX, no third-party)
  • pattern-site — email pattern extracted from the company's own site
  • source-page — person/role confirmed on a dated public page
  • pattern-only — guessed by pattern, flagged in-sheet as lower confidence so you know

For UK records, decision-maker names come from Companies House officer pages — free, public, and every name ships with a link to the exact register page. Here's a real 10-row sample built for a hospitality brief: 7 of 10 venues enriched with named directors + register links; the 3 we couldn't verify honestly say "enriched on delivery" instead of guessing.

What we'd tell you if you're buying lists anywhere

  1. Ask for the evidence column. If a vendor can't show source URL + method + date per row, "verified" is marketing.
  2. Generic inboxes (sales@, info@) are worthless for outreach — replies don't come. Named-person contacts only.
  3. Watch the refund currency. "Credits back" ≠ money back. Our per-row guarantee pays cash.
  4. Freshness decays fast. A list older than 30 days is 10–20% dead on arrival. Our scanner re-runs daily; rows ship with the check date.

Honest numbers

Cold email reply rates in 2026 run 0.45–5% (Belkins' public data). No list quality fixes a dead channel — leads are only half the job, and we say so on the landing page instead of promising "meetings guaranteed".

Pricing is public: $299 / 100 records one-off, $799 with a done-for-you campaign, $2,500/mo retainer. First orders can run through escrow — money releases only when you accept the sheet.

Questions about the ATS-scanning approach, MX verification, or the Companies House trick? The interesting parts are all in the repo — happy to dig in the comments.

Top comments (0)