Reducing your BizReach (and per-placement database) cost with AI sourcing

What you actually pay BizReach in a year

Most agency principals quote a single number when asked about their BizReach line: the subscription. ¥1.2 million per year for the standard headhunter plan (¥600,000 per six months, paid in advance). That number is right but incomplete. The full annual cost has three layers, and the layer most often missed is the largest.

The three layers, in order of size:

1. Subscription — about ¥1.2M per year. The standard headhunter plan on BizReach. Smaller platforms (doda X, AMBI) run lower; enterprise tiers run higher. Paid in lump sum, refund-free if you cancel mid-term.
2. Per-placement performance fees — typically 20–30% of your recruiting fee. The big one. On a ¥10M placement at the 25% midpoint, you owe the database ¥875K — every single placement that originated on the platform.
3. Contingent attribution exposure — variable. Most platform contracts allow fee attribution when a candidate has been touched on the platform, even if the placement was finalized through another channel. This layer rarely shows up in budget conversations until an audit forces it to.

For a 12-biller agency doing 90 placements per year through BizReach at the 25% per-placement rate, the second layer alone is roughly ¥78.75M per year. The subscription is a rounding error against that number. The actual unit cost of a placement is what matters.

The For Startups precedent — what an audit looks like

The clearest public case study of what the third cost layer looks like when exercised is For Startups, Inc. (TSE: 7089), a publicly listed Japan agency. In January 2023, the company disclosed a retroactive restatement covering FY3/2018 onward, recognizing ¥402 million in additional cost of sales owed to multiple human resource database operators. ¥118M was unpaid principal. ¥283M was contractual damages — a 2.4× penalty.

The cause, in the company's own words, was a 規約の誤認 — a misinterpretation of fee-attribution terms in cases where a single career-changer was registered on multiple databases. Under several operators' contracts, fees could be owed to each operator the candidate had touched, not only to the one through which the placement was finalized.

This is the only public, on-the-record example of what an agency's contingent exposure to per-placement-fee databases looks like once audited and exercised. Most peer firms have similar exposure profiles. They are simply not yet under securities-disclosure pressure to surface them. (For the deeper structural argument, see The Database Tax briefing.)

The margin math: same fee, +¥608K per placement

Walk through one ¥10M placement, two ways. Same client, same fee from the client. The only thing that changes is which sourcing layer surfaced the candidate. Take BizReach at the 25% midpoint per-placement rate, against a Headhunt.AI-first path where the candidate came from the AI scoring layer.

That spread compounds across every placement on the desk. A 12-biller agency doing 90 placements per year converts +¥608K per placement into approximately ¥55M of annual margin recovered. The number scales with placement volume. A 25-biller desk doing 200 placements lifts about ¥122M per year. None of that requires a single client to pay more. It requires a single sourcing layer to charge less for the same result.

Why AI sourcing covers most of what BizReach does

The argument for per-placement-fee databases has always been candidate access — a registered pool of active or passively interested job-changers that the platform aggregates and the agency taps into. That argument was structurally true for most of the 2010s. It is now only partially true.

The pool itself was never the unique value. Public profile data on LinkedIn has been the largest pool of working professionals in Japan for years. What was hard, until very recently, was the throughput of working that data — Boolean strings, manual longlisting, hand evaluation of tenure and company-tier patterns. Working that pool by hand cost more in recruiter labor than the per-placement-fee database alternative cost in fees, for many roles. That trade-off made the database channel the rational primary path.

In the last 24–36 months that calculation has shifted. AI scoring against full profiles at scale is now viable. Headhunt.AI's database covers 4M+ Japan-focused professional profiles, primarily public LinkedIn data layered with public X, GitHub, Facebook, and Instagram signals where candidates have visible activity. The matching engine reads JD-to-profile fit at depth that Boolean cannot express — tenure patterns, company-tier sequences, bilingual register, adjacent-industry context. (See AI candidate scoring, explained for the mechanics.)

The candidates on Headhunt.AI overlap substantially with the candidates on BizReach. Both pools include working senior professionals. The cost per qualified meeting, however, is structurally different — measured against the ¥107,676 expected revenue per meeting in our 2026 cohort, Headhunt.AI runs at roughly ¥6,250 per meeting at the Enterprise Annual rate. Per-placement-fee databases, when amortised across the meetings they produce, typically run several multiples higher. (For the unit-economic framework, see what a recruiter meeting is actually worth in Japan.)

Where BizReach still wins

The honest version: per-placement-fee databases retain genuine advantages in two specific segments. Pretending otherwise would not pass scrutiny on your team.

Active job-seekers. Candidates who have explicitly registered on BizReach are signaling active or imminent interest in changing jobs. For roles where active intent is the primary filter — junior-to-mid mass-market roles, volume hiring, certain non-bilingual specialties — that signal is real and useful. AI sourcing against passive public profiles cannot replicate the "ready-to-move" filter that registration provides.

Certain volume mid-market roles. For roles where the candidate base is broadly available across many agencies and the differentiator is speed-to-meeting, the active pool on BizReach often produces meetings faster than passive sourcing on public profiles. The trade-off is margin: speed comes at the cost of the 20–30% per-placement fee.

For passive senior candidates, which is where the highest-margin agency placements come from, the active-intent filter on BizReach becomes a liability rather than an advantage. The candidates you most want — the ones who would only consider a move for the right role — are precisely the ones who do not register on public job platforms.

The migration plan — sequenced, not abrupt

The cleanest way to bring the BizReach line down is sequenced replacement, not abrupt cancellation. Three phases, each independently measurable.

Phase 1 — Lead with AI sourcing on every new search

For the next quarter, put Headhunt.AI in front of every new search. Run BizReach as the fallback channel when the AI list does not produce enough qualified candidates within the search window. Track which channel each placement originated on. Track the fee paid to each channel per placement.

Most desks find within one quarter that 60–80% of new searches close out of the AI channel without needing the per-placement-fee fallback. The remaining 20–40% are typically the volume mid-market and active-intent roles where the database channel still wins.

Phase 2 — Renegotiate the subscription at renewal

Once you have a quarter of data showing what the residual usage looks like, the renewal conversation gets easier. BizReach's account team will renew at lower tiers before they let the contract walk entirely. Frame the conversation around the residual: you are not canceling, you are right-sizing to the residual roles where the platform genuinely wins.

Phase 3 — Track contingent exposure quarterly

The third layer of database cost — contingent attribution exposure — does not disappear when you reduce primary usage. It is a tail risk that compounds as the candidate base ages. Run a quarterly audit of which platforms your recruiters touched candidates on, against which platforms ultimately fee-attributed those placements. The For Startups precedent shows what the math looks like when this audit happens externally. Better to run it internally first.

The single-line test you can run this week

Before any of the migration plan, the simplest test is a side-by-side on one stuck role. Pick a role your team has worked through BizReach for two months or more. Run it through Headhunt.AI. The cost: ¥75,000 for 500 ranked candidates. Compare the AI top 100 against the candidates your team has already approached.

If even one in a hundred is qualified and new to your team, you have your answer. The AI is surfacing profiles your current BizReach-driven process is structurally missing — proof of concept on your hardest case, with zero added contractual liability.

The argument is not that BizReach is wrong. The argument is that the channel mix that made sense in 2018 no longer makes sense in 2026, because the AI-first alternative did not operate at this throughput in 2018 and does today. The 20–30% per-placement fee is the cost of a sourcing approach that has been overtaken by a better one for most segments. Stop being the firm that pays that tax on every placement when you do not have to.

Frequently asked questions

Sources

Production data from ExecutiveSearch.AI K.K. and ESAI Agency K.K. internal operations: 16-week 2026 outreach cohort (123,675 candidates contacted, 3,868 replies, 1,260 qualified meetings) and Q1 2026 desk results. BizReach platform-level financials drawn from Visional Inc. (TSE: 4194) H1 FY7/2026 results, TSE filing dated 17 March 2026; BizReach segment H1 revenue ¥38.3B (+19.2% YoY), 42.7% adjusted operating margin. For Startups, Inc. (TSE: 7089) ¥402M restatement: timely disclosure (TDnet), 20 January 2023; FY3/2023 Q3 earnings call. Per-placement rate ranges drawn from agency-side conversations with dozens of Japan firms through Q1 2026. Methodology, published-sample sizes, anonymisation policy, and statistical methods at our methodology page. For the full structural argument, see The Database Tax briefing.