Paraform is the company the Contingency recruiting marketplaces vertical is built around, and the reason it is interesting is a pricing decision rather than a technology one: it pays independent recruiters roughly 70% of the placement fee against the 25–35% they would keep inside an agency (HeroHunt; Recruiterflow), and removes the standard 60-day payment hold. Doubling the supply side's income is a proposition that requires no product to be believed. Everything else is downstream of it.
Founded around 2022–2023 by John Kim and Jeffrey Li. Pre-seed $1.4M in March 2023; $3.6M seed in April 2024 led by A* (TechCrunch); $20M Series A in June 2025 led by Felicis, with A*, BOND, DST Global and Liquid 2 (BusinessWire); $40M Series B announced 18 March 2026 led by Scale Venture Partners, taking the total to $65M (Paraform; Axios Pro). No valuation has been disclosed at any round. [UNVERIFIED]
The arithmetic nobody does
$50M cumulative paid to recruiters (Paraform), at a ~70/30 split, implies roughly $71M of lifetime gross placement value and ~$21M of lifetime net revenue. [UNVERIFIED] derived, but it follows directly from two sourced figures. Against $65M raised.
Per placement: average placed comp is ~$260,000 cash (Paraform); at a 22.5% fee that is a ~$58.5K invoice, of which Paraform keeps roughly $15–18K — about 5–7.5% of first-year salary.
That is a thin slice by marketplace standards, and it is the right way to read the business. Press coverage will report the gross number; the company keeps the small one. See GMV is not revenue and What a rake can actually be.
What is disclosed
- 1,000+ companies served as of March 2026; "thousands" of recruiters; the stated ambition is "$1B+ in coming years" (Paraform).
- Year-one revenue >$1M on a three-person team, and 40x revenue growth from seed to Series A — but the metric is undefined, so it is unknowable whether that is gross placement value or net take (TechCrunch; BusinessWire).
- ~12 days average from role open to meeting the eventual hire — self-reported, and not the same metric as time-to-fill, which runs 32 days for agency perm and 44–45 days in-house (Paraform; Pin).
- Logos: Palantir, Rippling, Cursor, Windsurf, Decagon, Abridge, Hightouch.
- Marketplace skew, disclosed voluntarily: the top 12% of candidates take more than 25% of all offers (Paraform).
- More than half of customers already have an in-house talent team (TechCrunch). This is the healthiest fact about the company: it is elastic capacity, not a fight with the TA function.
Why it is structurally different from the dead ones
Triplebyte and Hired both made candidates the scarce, expensively-acquired side. Candidates leave the market the moment they succeed, so the acquisition cost never amortises. Paraform's scarce side is recruiters, who are durable and repeat. And the platform does not own vetting quality — the recruiter does, and is paid on outcome — so Triplebyte's death spiral of growth diluting the screen is structurally unavailable here.
The macro supplied the sellers: roughly 100,000 recruiters laid off in 2022–23 became independents in need of demand.
The two unresolved things
Is there any independent evidence it works? No. No third-party placement counts, no external fill-rate data, no substantive recruiter or employer sentiment corpus; Trustpilot pages returned 403 on every attempt. Every performance claim above comes from Paraform's own blog or its funding PR. The one externally-legible signal is customer identity — Palantir, Rippling, Cursor and Decagon have real in-house TA and do not buy vanity vendors — which is meaningful and not quantitative.
Does it enforce exclusivity per role? Frontlines.io frames the model as "forced exclusivity," but the page returned 403 and only the framing could be verified. Dover, a direct competitor, describes the opposite: "five different recruiters are all sourcing for your senior engineer role." Both cannot be true. If it is the second, realised recruiter earnings are a fraction of the headline 70% and churn follows.
Pressure and posture
Dover puts Paraform's fee at 15–25% — $22,500–$37,500 on a $150K hire — against its own $2,000–$7,000 per hire (Dover). Wellfound Autopilot anchors a managed outcome at 10% plus $500/month per role (joinnextdev) [WEAK], and Juicebox sells self-service sourcing at $99–179/month. Paraform's defence is that these all serve the middle of the role distribution while its $260K average comp sits in the scarce tail.
On Getting cut out: the public Terms contain no explicit bar on an employer hiring a recruiter's candidate off-platform after introduction. The real terms sit in a private Recruiting Agreement that is not public [UNVERIFIED]. The working defence is economic — a defecting recruiter gains ~30 points on one deal and loses inbound reqs from 1,000+ companies — and it holds only while req supply is abundant.
The per-role listing fee is the most underrated design choice. Kim says it exists to force "buy-in from startups to the two-way marketplace" (TechCrunch) — a commitment device against employers spamming reqs they will not close, which is the failure mode that burns supply-side goodwill fastest.
The wedge is real, the structure is better than its predecessors', and the sizing is the problem: a ~$1–1.2B US net-revenue pool at 100% share of perm placement. Watch what it does with the Series B — whether it pushes toward subscription revenue, or follows Mercor, Turing and micro1 toward the lab budget line, which is where every other talent marketplace with real supply has ended up. See What the model actually is.