1B+ person profiles, delivered on your schedule.
The whole person graph as files you own — professional profiles, decision makers, developers and clinicians in one delivery, on the cadence you agree. For teams who want the data in their own warehouse rather than behind a call.
Who already builds on it.
Platforms shipping their own products on these records today.
Four segments. One delivery.
Most bulk person files are one undifferentiated pile of professional profiles. Three of these four are populations other providers do not hold at all.
People
Professional profiles with full work history — titles, employers, start and end dates, role descriptions and education.
Decision Makers
A curated, continuously monitored set. 10M+ of them sit across the top 250K companies.
Developers
GitHub profiles and real developer activity — the engineers behind a stack, not the ones who filled in a form.
Healthcare
Doctors, nurses and residents, with the specialty context a commercial team needs.
What one row contains.
The fields that usually go missing elsewhere — role dates, descriptions, and the link back to a company — are the ones that decide whether a file is usable.
Identity
Name, location, current title and employer, and the public profile URLs that anchor the record.
Work history
Every role with employer, title, start and end dates and the description — not just the current one.
Education
Institutions, degrees, fields of study and dates.
Social activity
Posts and comments carried on the person record, so you get what someone is saying, not only what they are.
Specialty attributes
Developer activity where the person is an engineer; specialty and credentials where they are a clinician.
Company link
Every person resolves to a company in the Company Dataset, so the two files join cleanly.
Four steps, then it repeats.
Bulk is a standing arrangement rather than a purchase — the point is the next drop, not the first one.
- 01ScopeChoose the segments and the regions you need. Anything more specific, we scope it with you.
- 02Your paceScheduleAgree the cadence — monthly, quarterly, or a one-off cut.
- 03DeliverFiles land in your warehouse or object storage in the format you use.
- 04Top upKeep the records that move current with the APIs, so the file does not age between drops.
What teams do with the file.
Each of these is a combination of segments plus a delivery cadence — nothing here needs a product we have not already listed.
Keep a recruiting product from going stale
People and Developers on a monthly drop, with Signals flagging who moved so you refresh those rows instead of the file. For ATS and sourcing platforms.
Find the companies whose engineers use your stack
The Developers segment joined to the Company Dataset — real activity, not a form fill. For dev-tools go-to-market.
Stop CRM decay without a cleanup project
The People segment in your warehouse, topped up through the APIs as records change. For RevOps and data teams.
Map the clinicians who shape a field
The Healthcare segment with its specialty context, joined to your own territory data. For pharma and medtech commercial teams.
Score accounts on who actually works there
Decision Makers joined to the Company Dataset, so a score reflects the buying committee rather than headcount. For sales-intelligence platforms.
Ground an agent without a live call per turn
People and Decision Makers held locally, so retrieval is a lookup in your own store. For AI and agent teams.
Know when a champion leaves
Join your closed-won contacts to the People and Decision Makers segments, and let Signals flag the moves — a departure is churn risk, an arrival somewhere new is a warm lead. For customer-success and revenue teams.
Build the tool instead of buying seats
The whole graph in your warehouse, so prospecting, routing and territory planning run on infrastructure you own rather than a per-seat licence. For teams replacing an off-the-shelf platform.
Own the file. Keep the movers fresh.
The two are not alternatives. Files give you scale and a cost you control; the APIs keep the records that actually change from going stale between drops.
The file
Everything you scoped, in your warehouse, queryable with the rest of your data.
The APIs
Resolve a record at the moment of decision rather than trusting the last drop.
Signals
Be told which records moved, so you refresh those rather than re-pulling everything.
Company Dataset
The company side of the same graph. Person rows join to it cleanly.
The graph is ours. The identifiers stay yours.
We resolve and validate our own graph rather than reselling someone else’s, and matching stays hashed — so the raw identifiers you send are never exposed.
Security & Trust Centre
EU & UK
California
Audited
Dataset questions,
answered
Can we take one segment rather than all four?
Yes. Scope is by segment and by region — take developers in North America alone if that is what your product needs. Anything more specific than that, we scope with you rather than list it here.
What format do files arrive in?
The one your stack already uses, delivered to your warehouse or object storage. Format and destination are part of the delivery agreement, not a fixed product.
How often is the file refreshed?
On the cadence you agree. 350M+ profiles are refreshed every month on our side, so a monthly drop is a genuinely different file rather than the same one redated.
How do we keep it current between drops?
Do person rows join to company rows?
Yes. Every person resolves to a company in the Company Dataset, so the two files join without a matching step of your own.
Is this resold from another provider?
No. We resolve and validate our own graph across 18+ data sources and reconcile them into a single answer.
We already pay for an off-the-shelf platform. Why take the file?
Because the file is infrastructure and the platform is a seat licence. With the data in your own warehouse you build the workflow your team actually runs, join it to your own CRM and product data, and stop paying per head for a tool you cannot change.
Tell us which segments you need.
Fifteen minutes to scope the segments, the attributes and the cadence — and to see what the rows look like on your own data.