In-house recruiting buys accumulated pipeline, control over candidate experience, and a fixed cost that improves with volume. Agency search buys immediate access to a network you have not built and converts recruiting to a variable cost. The right choice follows from your volume, the scarcity of your roles, and whether you need the capability permanently.
The comparison is usually framed as a cost question and answered with a spreadsheet that proves whichever answer the founder already preferred. Cost matters, but the two models differ in what they leave behind after the role is filled, and that difference compounds. An agency placement gives you a person. An internal function gives you a person plus a pipeline, a reputation with candidates, and market knowledge you keep. This guide separates what each model buys, then shows how to price both against your own volume.
An agency sells access and speed. They maintain relationships across a market segment continuously, so on the day you need someone they are starting from a warm list rather than from nothing. For a scarce role in an unfamiliar market, that head start can be worth several months, and months are the currency that matters when a seat is blocking revenue or delivery.
An internal function sells accumulation. Every search adds to a candidate database you own, a set of people who said no this time and might say yes next year, an interview process that gets sharper, and a reputation among candidates who went through it. None of that shows up in the cost of the first hire, and all of it shows up by the tenth.
There is a control dimension that founders discover late. The agency owns the candidate relationship, which means the candidate's impression of your company is mediated by someone whose incentive is the placement. That is not necessarily bad, and it is not yours. When a finalist hesitates, the person best placed to change their mind is someone who works at your company and can speak to it credibly.
Build the internal number first because it is the one people skip. Take the fully loaded annual cost of the recruiting seat, meaning salary plus employment costs plus tooling plus any sourcing subscriptions, then divide by the number of accepted offers you realistically expect them to produce in a year. Be conservative on that denominator: ramp time is real, and the first quarter usually produces less than the steady state.
Build the agency number as total spend rather than headline percentage. Take the fee structure you have actually been quoted, apply it to the roles you expect to run through that channel, and add the founder and hiring manager hours the agency does not remove, which is most of the interviewing and all of the decision-making. A model that appears to cost nothing until it succeeds still consumes your calendar.
Then compare at three volumes rather than one: your current run rate, double it, and half it. The crossover point is what you are looking for, not a single answer. Most founders find that at low volume the variable model wins comfortably, at high volume the fixed model wins comfortably, and there is a wide middle band where the numbers are close enough that the deciding factor is something other than cost, such as how badly you need control of the candidate experience.
When the role is scarce and you have no network in it. A specialized regulatory, clinical, or deep-domain hire in a market you have never recruited from is exactly the case where someone else's existing relationships beat your ability to build them from zero. Paying for a network is rational when building it would take longer than the problem allows.
When the volume is genuinely low or spiky. Two hires this year and possibly four next year does not amortize a fixed cost, and a fixed cost that idles is worse than a variable one that is occasionally expensive. This is also the case for a company between funding events where committing to permanent overhead is imprudent.
When confidentiality matters. Replacing someone who is still in the seat, or exploring a market entry you do not want signalled, is difficult to run internally without the search becoming visible. A third party can approach candidates without naming you until late in the process.
Put the hiring plan into a system your team can actually run
Free 1-user plan · No credit card
When you are hiring repeatedly into the same profile. The tenth engineer of a similar shape is dramatically cheaper to find than the first, but only if the same team ran all ten and kept the notes. Every search through an external partner restarts that learning curve from the partner's side rather than yours.
When candidate experience is a competitive lever. If you are recruiting against larger employers who pay more, the thing you can beat them on is how it feels to go through your process: fast, personal, and honest about the trade-offs. That experience is difficult to deliver through an intermediary whose incentives differ from yours.
When you want the people who said no. A well-run internal function treats a declined offer as the start of a relationship rather than the end of a transaction. Some meaningful share of senior hires at any company come from someone who turned them down eighteen months earlier and then circumstances changed. That asset only accrues to whoever owns the relationship.
The replacement or rebate provision is the one to read carefully. It defines what happens if the person leaves or is let go within a defined window, and the differences are material: a free replacement search, a partial refund on a sliding scale, or nothing at all. Understand whether the window runs from start date or offer date, and what conditions void it.
Exclusivity and engagement basis change both cost and behavior. A contingent arrangement, where the fee is paid only on placement, costs nothing until it works and gives the partner an incentive to prioritize whichever client is easiest to fill. A retained arrangement, where fees are staged through the search, buys committed attention and usually a more rigorous process, and it costs money even if the search does not conclude.
Also settle ownership of candidates in writing. If a person appears in an agency submission and also applied directly two months earlier, who is that candidate attributed to? Agree the rule and the evidence standard up front. This is the single most common source of disputes, and it is entirely avoidable with one clause and a dated application record.
Yes, and most companies past a certain size do, but only with a clear split rule. The workable division is by role type rather than by urgency: your internal team owns the profiles you hire repeatedly, and external partners take the one-off, scarce, or confidential searches. Splitting by urgency instead creates a pattern where every role becomes urgent and everything drifts external.
Write down what happens when both channels surface the same person, because they will. The rule should favor whichever channel had a documented, dated contact first, and your applicant records need to be good enough to establish that. Without it, you will pay a fee for someone who was already in your database, and the argument will damage the partner relationship.
Watch for the incentive problem where your internal team is measured on cost per hire while external spend sits in a different budget line. That structure quietly encourages people to route hard searches outward, which is the opposite of what you want, since the hard searches are where internal capability is built.
Transition on a role family rather than all at once. Pick the profile you hire most often, bring that one inside, and keep everything else external until the internal channel is demonstrably producing accepted offers for that family. Attempting the whole switch at a single moment means every search degrades simultaneously, and the pressure to revert becomes overwhelming.
Extract what you have already paid for. Any partner who has run searches for you holds a map of your market: where the people are, what they earn, what objections come up, and who turned you down. Ask for a written market summary as part of closing out an engagement, and make it a term of the next engagement rather than a favor you request afterward.
Expect a temporary dip. The internal channel is slower for its first two quarters while relationships and knowledge accumulate, and if you judge the switch on the first quarter you will conclude it failed. Set the review point at least two quarters out and define in advance what success looks like, so the decision to continue or revert is made against a written standard rather than against whoever is most frustrated that week.
Pitch N Hire is an applicant tracking system. Run job posting, screening, structured interviews, and offers from one pipeline — starting free for 1 user.
Free for 1 user · No credit card · Talk to a real hiring expert
Pitch N Hire gives a founder-led hiring process the structure it needs — one pipeline, structured interviews, and a record of every candidate. Start free with the 1-user plan.
Prefer to talk? Book a demo · Talk to sales · View pricing
Free 1-user plan · No credit card · Talk to a real hiring expert
See your true cost-per-hire and how much Pitch N Hire could save you — our free Recruitment ROI Calculator gives you the numbers in under a minute. No signup required.
Open the free ROI calculatorPrefer a tailored walkthrough on your real roles? Drop your work email:
★ Free 1-user plan · No spam · Talk to a real hiring expert
