Introducing the Similar People Endpoint
Recently, a customer reached out to ask if we could help with the following use case:
we would provide a small input set (say, 5 email addresses) and expect the system to return a larger set (e.g., 50) of similar profiles with enriched data.
I'm happy to introduce the Similar People Endpoint today, which takes an input of identifying factors of a person and returns similar people. By similar, I mean people in the same role at competing companies of your target.
For example, these would be my peers, as defined by the Similar People Endpoint:
Target: Steven Goh — 33 attempted searches, 27 similar people, 175 deci credits (17.5)
Similar people (CEOs/founders at competing data/proxy/scraping companies):
- Kabiru Mosadoluwa Audullahi — Executive Director, Swiftproxy
- Ugnius Zasimauskas — CEO, Coresignal
- Yoni Tserruya — Co-founder & CEO, Lusha
- Will Cannon — Founder & CEO, UpLead
- Karolis Toleikis — Founder & CEO, IPRoyal
- Ben Eisenberg — CEO, People Data Labs
- Or Lenchner — CEO, Bright Data
- Henry L. Schuck — CEO & Chairman, ZoomInfo
- Matt Sornson — GM Operations Hub & Breeze Intelligence, HubSpot
- Matt Curl — CEO, Apollo.io
- Scott Kim — CEO, RocketReach
- Yauheni Stsiapnou — CEO, Floppydata
- Liam Xavier — Founder & CEO, Scrapeless
- Jessica Thompson — CEO & Founder, Thordata
- Thibeau Maerevoet — CEO & Founder, Buy Proxy Servers
- Ryan Huber — CEO & Co-founder, Defined Networking
- Dinesh Gabriel — CEO & Founder, Nebula
- Shanelle Roman — Co-Founder & CEO, Nebula Proxy
- Denis Mars — Founder & CEO, GoProxy
- Aleksandr Sadovskij — CEO & Founder, Proxy-Cheap
- Shachar Daniel — CEO & Co-Founder, NetNut
- Wolfgang Udo von Kindberg — CEO, Von Kindberg LLC
- Sajib Hossain — CEO, ipmela.com
- Ignacio Martín Llorente — CEO/MD, OpenNebula
- Raymond — Head of R&D, IPcook
- Cahyo Subroto — Founder & CEO, MrScraper
- Charles Cao — CEO, Nebula Block
This API call took 85 seconds to complete.
You can think of the Similar People Endpoint as "AI deep research" in the space of B2B data, and honestly, you wouldn't be very far off as to how it works in the background. It is a long-running query. It takes on average 1–2 minutes to complete, and the JSON response is streamed, not unlike LLM tokens.
Use-case
The obvious use case for the Similar People Endpoint is expanding prospect lists. For example, what I'll be doing is sending cold emails to similar people for every new user sign-up, as they are immediate good fits.
How well does it work?
In live tests against real CEOs, the endpoint returned a full set of same-role peers at competing companies with very high accuracy:
- Apple — Tim Cook: 18 competitors attempted, 18 similar people returned (100%)
- Tesla — Elon Musk: 11 attempted, 11 returned (100%)
- Stripe — Patrick Collison: 19 attempted, 16 returned (84%)
Beyond executives, we also went down the organization chart and did further tests.
- Middle manager — Bryan Irace, Engineering Manager (Link team) @ Stripe
- Time: 81s · Size: 25.7 KB · Credits: 105
- Attempted: 19 · Found: 12 (63%)
- Sample matches: Engineering Managers / Heads of Engineering at PayPal, Square, Cash App, Rippling, Tipalti, Finix, Helcim, Stax, FastSpring, Flock — exactly the right archetype (payments-company EMs).
- Rank-and-file — Robert Heaton, Member of Technical Staff @ Stripe
- Time: ~180s · Size: 65 KB · Credits: 335
- Attempted: 65 · Found: 36 (55%)
- Matches split into two clusters because Heaton holds multiple current roles in his profile (Stripe + several effective-altruism nonprofits + academia), so the algorithm fanned out across all of them:
- Engineers at payments competitors: Wise, PayPal, Checkout.com, Square, Cash App, Rippling, Tipalti, Helcim, FastSpring, Paddle, Stax, PaymentCloud
- "Member of Technical Staff" peers at AI labs: Anthropic, OpenAI, DeepSeek, Cohere, Meta — picked up via the MTS title alias
- Engineers at EA/charity orgs: Giving What We Can, Charity Navigator, Kiva, ACE, CEA, Coefficient Giving
- Academics at RCA, Royal Holloway, Surrey — from his secondary teaching/research affiliations
Accuracy table
| Tier | Target | Attempted | Found | Yield |
|---|---|---|---|---|
| Executive | Tim Cook (Apple) | 18 | 18 | 100% |
| Executive | Elon Musk (Tesla) | 11 | 11 | 100% |
| Executive | Patrick Collison (Stripe) | 19 | 16 | 84% |
| Middle manager | Bryan Irace (Stripe) | 19 | 12 | 63% |
| Rank-and-file | Robert Heaton (Stripe) | 65 | 36 | 55% |
Observations
- Yield drops as you go down the org chart. CEOs are public figures listed on every company's leadership page → near-100% hit rate. Mid-level and IC roles depend on per-employee enrichment, which is noisier — yield falls to 55–63%.
- Result count goes up the lower you go because the role is held by many people per company. Heaton's 36 matches vs. Cook's 18 reflects that.
- Multi-role profiles fan out the search. Heaton's secondary affiliations (EA orgs, academia) generated additional competitor sets and pulled in cross-domain peers — useful if you want full coverage, surprising if you expect a tightly focused list.
Limitations
Unfortunately, what we are not able to do as of now is return similar employees within the same company. That would require maintaining an employee database, which we are currently not doing because—short of scraping LinkedIn, which we will strictly never do—how else can we build up such a dataset? I have an idea, but it'll take time to flesh out. You'll be the first to find out when I have something in the pipeline.
Try Similar People Endpoint
What are you waiting for? Give the Similar People Endpoint a try right now!