You already know what your current stack costs. A LinkedIn Recruiter Corporate seat runs north of $10,000 a year. A contingency agency bills a percentage of first-year salary on every placement. Your in-house recruiters absorb the rest in sourcing time, screening, and scheduling. Add it up and the number lands somewhere between painful and unknowable.
An AI sourcing marketplace rearranges those line items rather than erasing them. The platform charges a fee. Specialized recruiters still take a cut when they close. AI sourcing agents burn credits. Your team still spends hours in the loop. The question is not whether those costs exist, but what the stack adds up to for your role mix, and which lever actually moves the number.
So what does a hire actually cost on this model?
The Four Lines on a Marketplace Invoice
Cost per hire on a marketplace is a sum, not a sticker price. Four components do most of the work:
- Platform fee. A subscription, a per-req charge, or a success fee that unlocks the marketplace, matching, and the hiring command center.
- Recruiter take. The specialized recruiter who runs the search gets paid on close, usually as a percentage of base or a flat per-hire fee negotiated through the platform.
- AI sourcing credits. Metered usage for autonomous sourcing runs, contact enrichment, or outreach sequences. Some vendors bundle this; most meter it.
- Internal time. Your coordinator, your hiring manager, and your TA lead still read shortlists, interview, and sign off. This is the line most buyers forget to price.
For context, SHRM pegs average US cost per hire at roughly $4,700, and its 2025 benchmarking data separates that into $5,475 for non-executive roles and $35,879 for executive positions. Those are the floors to beat.
What Each Line Actually Costs in 2026
Platform fees for AI sourcing tools cluster in a wide band. Published rates range from entry seats around $99 to $170 a month on tools like Gem Lite or LinkedIn Recruiter Lite, up to LinkedIn Recruiter Corporate at $10,800 to $15,000 per seat per year with a three-seat minimum. Marketplace-style platforms more often price per role or per hire so the fee tracks activity rather than seat count.
Recruiter take is the biggest single line when it exists. Contingency fees typically run 15% to 25% of first-year base, and retained executive search sits near 33% with minimums of $80,000 to $100,000. On a marketplace, the same specialists compete for your req at negotiated rates, which is where the compression usually shows up: a 20% contingency on a $120,000 role is $24,000, while a marketplace-brokered flat fee on the same role often lands at $8,000 to $12,000.
AI sourcing credits are the newest line and the easiest to misread. Fetcher's Growth plan runs $379 a month for 500 candidates a year; per-contact models elsewhere price enriched profiles at a few dollars each. Credits scale with volume, so they punish spiky hiring and reward steady req flow.
Internal time is where the AI assist earns back the fee. Recruiters using AI tools reclaim roughly 20% of their working week by handing scheduling and first-pass review to agents. On a loaded TA cost of $100 an hour, that is around $400 per full week per recruiter back in your budget.
A Worked Example at 50 Reqs a Year
Picture a company hiring 50 roles a year. Mix: 30 individual-contributor roles at $90,000 base, 15 senior IC and manager roles at $140,000 base, 5 leadership roles at $200,000 base. Current stack is LinkedIn Recruiter Corporate for the TA team, a job board or two, and contingency agencies on roughly half the reqs.
On the current stack, agency fees alone on 25 placements at a 20% contingency on a blended $115,000 average land around $575,000 a year. Add LinkedIn seats, job board spend, and internal time, and total recruiting cost sits well above $700,000, or about $14,000 per hire.
On a marketplace, the same req load typically breaks down as follows: a platform fee of $40,000 to $80,000 a year, marketplace recruiter take averaging $8,000 to $12,000 per filled role (so $200,000 to $300,000 across 25 brokered fills), AI sourcing credits of $20,000 to $40,000 depending on outreach volume, and internal time roughly a third lower because agents handle shortlist QA and scheduling.
Total: $350,000 to $500,000 for the same 50 hires. That is a per-hire range of roughly $7,000 to $10,000, versus $14,000 on the legacy stack. The gap narrows for teams that rarely use agencies and widens for teams that lean on them.
The Variables That Move the Number Most
Four inputs do most of the moving. Model them before you model vendors.
Req Volume
Platform fees are mostly fixed. The more reqs you push through, the lower the fee amortizes per hire. A $60,000 annual platform fee is $6,000 per hire at 10 hires and $600 per hire at 100. If your req volume is lumpy or seasonal, pricing that tracks roles (per-req or per-hire) protects you from paying for empty months. For teams managing spikes, this is the same logic behind handling hiring surges without permanent overhead.
Role Seniority
Contingency bills scale with salary; marketplace fees usually do not. That is where the compression is biggest. A $180,000 engineering lead at 22% contingency is $39,600 in agency fees; the same role at a $12,000 marketplace rate is a $27,600 saving on one hire. On junior roles under $70,000, the gap narrows and sometimes flips.
Fill Rate
If your current agencies close half the reqs they take, your effective cost per hire is double the sticker rate because the losses still cost you time. Marketplaces that match multiple recruiters to a single req raise fill rate by design. Model this before you model price.
Time-to-Fill
Vacancy cost is the line nobody puts on the invoice. Northwestern research found that leaving key sales roles vacant can reduce company revenue by 5% or more. Engineering roles already take an average of about 62 days to fill. Every week shaved off that cycle is lost revenue recovered, and it rarely shows up in a side-by-side pricing sheet.
Where the Marketplace Model Breaks Even
The economics work cleanly in three profiles:
- Mid-volume teams hiring 20 to 150 roles a year with a mix of IC and senior roles. Platform fee amortizes well, and recruiter take compresses most on the senior slice.
- Teams carrying heavy agency spend. If more than a quarter of your hires involve contingency, the recruiter-take line is where the savings land.
- Teams with uneven req flow. Per-req or per-hire pricing turns a fixed cost into a variable one, which matters when a quarter goes quiet.
The model stops working, or at least stops winning, when you hire fewer than 10 roles a year (platform fee does not amortize), when nearly all hires are entry-level under $60,000 (contingency was never the alternative), or when your team is already running a mature in-house sourcing function with low agency reliance. In that last case, the marginal value is AI agents on top rather than the full marketplace.
A bad hire costs at least 30% of first-year earnings per the US Department of Labor, which is a reminder that cost per hire is only half the equation. Speed and fit matter as much as fee. For a deeper read on the trade-off, the piece on balancing speed and quality in hiring is worth the five minutes.
How to Price This for Your Own Stack
Three steps, done in a spreadsheet before you take a sales call:
- Pull last year's numbers. Total agency spend, total platform spend, total internal TA cost, total hires. Divide. That is your real cost per hire, which is usually 1.5 to 2x what finance thinks it is. The guide on hidden recruiting costs covers the lines teams routinely miss.
- Segment your reqs. By seniority band and by whether an agency touched them. Price each segment on a marketplace model using the ranges above.
- Model vacancy cost. Pick your three highest-impact role types. Estimate weekly revenue or productivity loss per unfilled seat. Add that to the fee comparison. This is where the metric you are actually optimizing for gets honest.
Buyers who do this exercise usually find the marketplace wins on blended cost per hire by 25% to 50%, loses on isolated high-volume junior roles, and wins by a lot on senior roles where contingency fees scale with salary. The spread is why generic ROI claims are useless and your own numbers are not.
The Honest Answer on ROI
A sourcing marketplace does not make hiring free. It replaces the two most expensive line items (full-rate contingency and manual recruiter sourcing time) with two cheaper ones (negotiated marketplace fees and metered AI credits), and gives you a command center to see what you are actually spending. The savings are real but concentrated in specific role types and volume bands.
Price it against your own req load, not a vendor's case study. If the math works for your seniority mix and your fill rate, the switch pays for itself inside a quarter. If it does not, the honest recommendation is to keep the pieces that are working and layer AI agents under them rather than replacing the whole stack.