The result, in the first month.
from kickoff to the first leads delivered. Fully handed over on day 24.
of delivered leads with a named decision-maker and a direct email verified as valid in the last 7 days.
new verified leads in about 36 hours when the client raised the target from 300 to 500. Volume follows the goal the client sets.
data cost per lead: providers' monthly list prices divided by leads accepted. Hosting and the build are excluded.
The client.
A US fintech that helps independent medical and dental practices lower their card-processing costs. Founder-led, with no sales team to find prospects. It needed named decision-makers with direct, verified emails, in volume, from a system the founder could run on his own.
Why this market is hard.
Buyers tune out
Practice owners hear card-processing pitches every month. More volume on its own only adds noise.
Timing decides
Practices rethink how they take payments when they open, change owners, move or switch practice software.
Databases fall short
Single-location practices are thinly covered, and a front-desk inbox is not a decision-maker.
What I built.
One pipeline from public registries to a lead the founder can load straight into a sequence. Every step is automated, and every lead carries the reason it was accepted.
- 01 · WatchRegistry changes
- 02 · ScreenReal prospects only
- 03 · IdentifyNamed owner
- 04 · VerifyDirect email, ≤7 days
- 05 · RankExplainable score
- Screened outhospital systems, groups, duplicates, catch-all mail servers
- Monthly target, set by the clientruns each morning, stops when the target is met
Every accepted lead keeps a record of the rules and evidence behind it.
Watch for moments of change
New and changed practices come from the federal provider registry (NPPES) and state business filings as they are published. More than 29,000 practices are tracked for a new practice, a new owner, a new location or a move, for this client's market; coverage is set by the ICP, not the system.
Keep only real prospects, reach the person who decides
Each practice is matched to its website; hospital systems, large groups, multi-site brands, existing customers and duplicates are removed. The engine then names an owner, partner, executive, clinician or the registered official. Office managers and shared inboxes don't count.
Rank, then deliver
An explainable score (fit, timing, payment context, decision route) orders the list and flags recent buying moments. Leads arrive as files ready for the client's sequencing tool, with bulk tracking of what is already in a sequence.
Volume is a setting, not a ceiling.
The engine runs every morning until the monthly target is met, then stops spending. The founder changes the target in the app himself, with no engineer involved. He set it at 200, then 300, and currently around 500. The system scales to whatever volume is chosen.
Lime: leads accepted. Dashed: the monthly target the client set in the app. Each flat stretch is the engine stopping on purpose because the target was met. It went live on Oct 9.
Daily numbers
| Day | New | Cumulative | Target |
|---|---|---|---|
| Oct 4 | 16 | 16 | – |
| Oct 5 | 1 | 17 | – |
| Oct 6 | 41 | 58 | – |
| Oct 7 | 120 | 178 | – |
| Oct 8 | 0 | 178 | – |
| Oct 9 | 16 | 194 | 200 |
| Oct 10 | 23 | 217 | 200 |
| Oct 11 | 0 | 217 | 200 |
| Oct 12 | 0 | 217 | 200 |
| Oct 13 | 0 | 217 | 200 |
| Oct 14 | 0 | 217 | 200 |
| Oct 15 | 0 | 217 | 300 |
| Oct 16 | 90 | 307 | 300 |
| Oct 17 | 0 | 307 | 300 |
| Oct 18 | 0 | 307 | 300 |
| Oct 19 | 1 | 308 | 500 |
| Oct 20 | 60 | 368 | 500 |
| Oct 21 | 144 | 512 | 500 |
| Oct 22 | 0 | 512 | 500 |
| Oct 23 | 0 | 512 | 500 |
| Oct 24 | 0 | 512 | 500 |
What a delivered lead looks like.
Sample lead record · illustrative, not a real practice
| Entity | Registered practice (organization) |
| Specialty | General dentistry |
| Footprint | 1 location · independent |
| Why now | Change of ownership · 41 days ago |
| Decision-maker | Owner · named |
| Direct · valid · verified 3 days ago | |
| Mail server | Not catch-all |
| Score | 80 / 100 · fit 35 · timing 30 · route 15 · payment context 0 of 20 |
The first month, measured.
emails found were risky or invalid and never reached the list, protecting the sender's reputation.
to fill the gap after the founder raised the target from 300 to 500, with no code changes.
of manual research replaced in month one, about three weeks of full-time work. Estimate: 15 minutes per qualified lead; time on practices that didn't qualify isn't counted.
All figures come from the system's own records as of October 24, 2026.
Timeline.
Kickoff
Market, decision-maker rules and verification standard agreed.
First leads
The founder reviewed the first batch and approved it.
Handover
Fully handed over; the founder runs it and sets the target himself.
Target raised twice
200 to 300 to about 500, and he asked for bulk tools to work leads faster.
What this means for you.
If your market is large and looks identical from the outside, more list volume won't fix it. What works is spotting the change that comes before a buying decision, public or from licensed data sources, refusing any contact that can't be verified, and letting you set the volume. The signals and rules are built for each market; the pattern carries over.
Want to know which signals precede a purchase in your market? The Pipeline Leak Teardown includes a read on your signal options.
Get the teardown