Should You Hire an In-House Recruiter or Outsource? The Break-Even Math

Here is the short answer, before any of the maths. As a working heuristic: if you make fewer than roughly 10 hires a year, agencies or fractional support are usually cheaper than a full-time salary. Between roughly 10 and 25 hires a year, embedded recruitment or RPO tends to win on cost per hire. At 25 or more sustained hires a year, hiring your own in-house recruiter starts to make financial sense. Those are ranges, not laws, and the rest of this article shows you the workings so you can run the calculation for your own numbers.

The reason this question deserves real maths rather than gut feel is that the three options have completely different cost shapes. An in-house recruiter is a fixed cost that does not care how many hires you make. An agency is a variable cost that scales with every placement. Embedded recruitment sits in between: a flat monthly fee you switch on and off as hiring demands. Which shape fits your business depends almost entirely on your hiring volume and how predictable it is.

I have sat on every side of this decision: as an in-house talent lead, as the embedded recruiter brought in to replace agency spend, and as the person advising founders on when to make their first TA hire. What follows is the framework I actually use.

The True Cost of an In-House Recruiter

The number on the job advert is not the number on your P&L. A UK in-house recruiter at the mid-to-senior level typically commands a base salary of £45,000 to £60,000, and experienced talent partners in London tech regularly go beyond that. On top of base salary you are paying employer National Insurance, pension contributions, and the tooling without which a modern recruiter cannot do the job: a LinkedIn Recruiter seat alone runs at roughly £8,000 to £10,000 per year, before job board credits and an ATS licence.

Build it up transparently and the fully loaded picture looks like this. These are estimates based on typical UK market rates, and your own numbers will vary with location and seniority.

Estimates based on typical UK market rates. For a US comparison, Paraform's cost analysis puts in-house recruiter bases at roughly $70K–$110K, or $100K–$150K fully loaded.

That £65,000 to £85,000 is committed whether the recruiter closes 30 roles this year or three. That is the single most important property of the in-house model: it is brilliant value at high volume and painfully expensive at low volume. The CIPD's resourcing research has long made the broader point that UK organisations consistently underestimate the full cost of their resourcing function, and the fixed-cost recruiter is the clearest example.

What One Recruiter Can Actually Handle

Before you can do any break-even maths, you need a realistic capacity number. The industry rule of thumb is that one full-time in-house recruiter handles somewhere between 15 and 25 active requisitions per year. Treat that as a range with a heavy dependency: the seniority mix.

A recruiter filling customer support, junior sales, and operations roles can run a high-throughput pipeline and land at the top of that range or above it. A recruiter working niche engineering searches, leadership roles, and first-of-their-kind hires will close far fewer, because each search involves longer sourcing cycles, more stakeholder management, and higher offer-stage risk. If your hiring plan is ten senior engineers and a VP, one recruiter at 25 requisitions is a fantasy. If it is 25 entry-level roles from a strong inbound brand, one recruiter may have slack capacity.

This capacity number is the hinge of the whole decision. Divide your realistic annual hiring volume by it and you know how many recruiters you would need to employ, which is the fixed cost you are comparing against fees.

The 7-Point Decision Framework

Cost per hire is the headline metric, but it is not the only variable. Before the break-even table, run your situation through these seven factors. They are the questions I ask every founder who tells me they are "thinking about hiring a recruiter".

1. Hiring volume. The dominant factor. Under roughly 10 hires a year, a full-time salary is hard to justify. Over 25 sustained hires a year, it is hard to justify not having one.
2. Volume predictability. A steady 20 hires a year suits a permanent recruiter. The same 20 hires compressed into one post-funding burst suits an embedded model you can switch off afterwards. Fixed costs punish lumpy demand.
3. Role seniority mix. Heavy senior and niche hiring drags one recruiter's capacity toward the bottom of the 15 to 25 range and increases the value of specialist external search for the hardest two or three roles.
4. Employer brand maturity. A recognised brand with inbound applicants makes an in-house recruiter dramatically more productive. An unknown company hiring against household names needs outbound sourcing muscle, which favours experienced embedded or agency support.
5. Internal HR capacity. A recruiter without coordination support becomes an expensive diary manager. If nobody internally can own offers, onboarding, and scheduling, factor that into the in-house cost or choose a model that brings process with it.
6. Budget model. Headcount budget and services budget are approved differently in most companies. A recruiter is a permanent payroll commitment with employment rights and redundancy cost; embedded and agency spend is opex you can stop with notice. CFOs are rarely indifferent between the two.
7. Speed to start. Hiring a good in-house recruiter is itself a 2 to 3 month recruitment process, followed by ramp time. An embedded recruiter or agency can be working your roles within days. If the hiring need is now, the in-house option arrives late to its own party.

The Break-Even Table: Cost Per Hire at 5, 10, 15, 25 and 40 Hires

Now the maths. Assumptions first, so you can swap in your own figures. Agency contingency fees run at 15 to 30 per cent of first-year salary, with 15 to 20 per cent the standard range for non-executive roles, as documented by Valuable Recruitment's UK fee guide. I will model agency at 20 per cent of a £60,000 average salary, which is £12,000 per hire. To see how fast that compounds in dollar terms: five placements at 20 per cent on $120,000 salaries is $120,000 in fees, for five hires.

For in-house, I will use £75,000 fully loaded per recruiter with capacity of roughly 20 requisitions a year, so 40 hires requires a second recruiter. For embedded, the market runs at a flat monthly fee of roughly $5,000 to $20,000 depending on seniority and scope, which is around £4,000 to £16,000 indicative; Paraform and EOR HQ's RPO cost guide both map this pricing territory in detail. I will model embedded at £8,000 per month, engaged only for the months the volume requires.

*At very low volumes, agency can still beat embedded if the roles are one-off and urgent, because there is no monthly commitment at all. The highlighted cell marks the typically cheapest model at each volume; treat all figures as illustrative ranges, not quotes.

Read the pattern, not the pennies. Agency cost per hire is flat at every volume: predictable, but it never gets cheaper, and at 25-plus hires the total becomes indefensible. In-house cost per hire collapses as volume rises and the fixed salary is spread across more hires, crossing below embedded somewhere in the low-to-mid twenties. Embedded occupies the wide middle band, and it is the only model where you can prune the months you do not need.

If you want the deeper comparison of how the three models differ operationally rather than just financially, Talentful's agency vs RPO vs in-house comparison is a solid companion read, and we cover the same ground from an SME angle in our own embedded vs RPO vs agency guide.

The Middle Path: Embedded as Try-Before-You-Build

Here is the honest positioning, including the part where the embedded model argues against itself. Embedded recruitment is not the permanent answer for every company. For a business heading toward 30-plus hires a year, the right end state is an internal talent function. The question is how you get there without paying agency fees on the way or hiring a recruiter into an organisation with no hiring process for them to run.

That is where embedded earns its place as the bridge. A senior embedded recruiter does not just fill the roles in front of them. Over a 6 to 12 month engagement they build the infrastructure your future in-house hire will inherit: a configured ATS with clean data, structured interview processes, calibrated scorecards, salary benchmarks, employer brand assets, and warm talent pipelines for your recurring role types. When your volume justifies the permanent headcount, you hire an in-house recruiter into a working system, and they are productive in weeks instead of spending their first two quarters building plumbing.

In several of my own engagements the explicit final deliverable was my own replacement: scoping the internal TA role, running the search, and handing over a documented playbook. A company that buys embedded recruitment this way is not outsourcing forever. It is renting senior capability while it decides, with real data, whether to build. Pricing for that kind of engagement is covered in detail in our embedded recruitment pricing guide.

The Risks Each Way

No model is risk-free, and anyone selling you one without caveats is selling too hard. These are the failure modes to price in.

In-house: the idle recruiter problem

If hiring slows, and it always slows eventually, you are paying £65,000 to £85,000 a year for someone with three open roles. Companies respond by giving recruiters HR generalist work they were not hired for, or by making them redundant, which costs money, morale, and the institutional knowledge you hired them to build. A recruiter hired in a growth spike is often the first line item questioned in the next budget review.

Agency: fees that explode with success

The agency model has no volume discount built in by default. Your best year of growth becomes your worst year of recruitment spend: 25 hires at typical contingency rates is a £300,000 line item for a function you still do not own. You also accumulate nothing. No pipelines, no process, no employer brand equity. Every search starts from zero, and the agency's incentive is the next placement, not your long-term hiring capability.

Embedded: lock-in to watch for

The embedded model's risk is dependency. If the provider keeps the process knowledge, candidate relationships, and ATS configuration in their own systems, you have rented a function without building one, and switching off the engagement means starting again. Before signing, confirm in writing that you own the ATS and its data, that processes are documented in your systems, and that the engagement includes a defined handover. Watch also for long minimum terms; a credible embedded provider should be comfortable with a 3-month initial commitment and rolling notice after that.

Frequently Asked Questions