What an Embedded Hiring Model Actually Looks Like: How It Differs from Agencies and RPO

What separates these three models

Agencies, RPO providers and embedded partners are often described using similar language, but the substance behind each label isn't the same. So what does an embedded hiring model actually mean in practice?

The answer can be difficult to pin down because agencies, RPO providers and embedded partners can all appear external on an organisation chart. All three can source candidates, manage interviews and help close roles. The difference becomes clearer when the search is no longer treated as the unit of work. It sits in what each model is designed to own, how success is measured, and what remains inside the business after the engagement ends.

For a CleanTech scale-up, that distinction matters. Hiring volume may still be uneven, while individual appointments can have an outsized effect on delivery. The business needs people in seats, but it also needs a way of hiring that does not have to be reinvented every time another role opens.

What agency recruitment owns

An agency is engaged to introduce candidates for a particular vacancy. It takes a brief, searches its network and presents people who may fit. If one is hired, the agency receives a placement fee, typically 15 to 25 per cent of first-year salary for standard permanent roles and often 25 to 30 per cent for senior or specialist searches.

That structure makes sense when the requirement is clear and the business needs access to a market it cannot reach easily on its own. The agency is being paid for the outcome of a specific search, not to redesign the environment around it.

Responsibility for that environment stays with the company. The quality of the role brief, the consistency of interviews, communication between hiring managers, pipeline visibility and the use of hiring data are all outside the placement transaction unless separately agreed. When the role closes, the relationship has largely done what it was set up to do.

This is why repeated agency recruitment can fill vacancies without creating a repeatable hiring process. Each successful placement solves an immediate need, but the reasoning, data and working habits behind the search do not necessarily become part of the business. The next role can therefore begin with many of the same decisions still unresolved.

Source: Recruitment & Employment Confederation; Sales Recruit UK, Energy and Renewables Salary Guide, as cited in Blog 5.

What RPO is built to manage

Recruitment process outsourcing goes further. Under an RPO arrangement, a company transfers all or part of its recruitment process to an external provider on an ongoing basis. Commercial models are commonly based on cost per hire, cost per resource, or a combination of the two, rather than a percentage fee attached to each placement.

That makes RPO structurally closer to embedded support than an agency model. The provider can manage requisitions across functions, coordinate activity, report against agreed measures and bring consistency to a large recruitment operation. It is designed to take responsibility for a defined process rather than a series of unrelated searches.

The point of separation is usually the operating environment. Traditional RPO works best where hiring demand is sustained and predictable enough to support a dedicated team, formal service levels and a delivery model built around volume. Measures such as time-to-fill, requisitions closed and candidates submitted per role provide useful control when hundreds of hires need to move through the same machinery.

A growing CleanTech business may have a very different pattern. Hiring volume can move well past the fifteen-to-twenty-role threshold where a permanent internal hire starts to make sense, and still not resemble the scale RPO is built for. Requirements keep changing as funding, projects and regulation move. Throughput still matters, but the harder part is often judgement: helping leaders define an unfamiliar role, challenging a brief before it reaches the market, and keeping assessment consistent when each hiring manager has a different view of what good looks like.

An SLA can measure whether the agreed process is moving. It is less suited to capturing the conversations that shape the process in the first place.

Sources: Robert Walters, What Is Recruitment Process Outsourcing, 2026; ADP, What Is Recruitment Process Outsourcing?

What embedded hiring looks like day to day

An embedded partner operates inside that earlier part of the work. It joins the rhythm of the business, works directly with leadership and hiring managers, and owns the hiring process from the point a need is identified. The distinction is not simply proximity. It is the scope of responsibility.

Before a search opens, that means turning a hiring request into a documented role brief that reflects what the business needs now, rather than recycling a job description from its previous stage. It means agreeing assessment criteria before interviews begin, so candidates are judged against the same requirements rather than a collection of individual impressions.

Once the search is live, the work includes sourcing and candidate management, but it also includes maintaining a pipeline that leadership can read without chasing for updates. It means showing where candidates are dropping out, where decisions are slowing down, how long roles are taking to fill and what each hire is costing. Hiring managers are part of the process rather than occasional recipients of CVs, with expectations, feedback and decisions managed consistently across the team.

These activities are deliberately connected, so a weakness in one tends to surface somewhere else. A weak brief produces an unfocused search, assessment drifts when criteria aren't consistent enough to catch it, and slow feedback along the way increases candidate loss on top of that. Without usable reporting, none of this reads as a pattern the business can correct. It just looks like a series of unrelated delays.

The embedded hiring model therefore produces two things at the same time. It delivers the hires the company needs now, while building the components of hiring as operational infrastructure: role brief templates, assessment frameworks, reporting, data and established ways of working. Those assets stay useful beyond the role that prompted them.

What the business is left owning

The practical choice between agency, RPO and embedded support is rarely a judgement on one model in the abstract. Each is built around a different hiring problem. An agency can provide reach and specialist search capability for a defined vacancy. RPO can bring control and consistent delivery to a large, sustained recruitment programme. Embedded support fits the period when a growing business needs hands-on delivery, but also needs the hiring function itself to take shape around that work.

This is why the most useful question is not simply how many roles an external provider can fill. It is what the company will be able to see, repeat and eventually own because the provider was there.

Founder-led hiring often begins with knowledge held between a small number of people. They know why someone was hired, which questions revealed the most, where the process became stuck and what finally moved it forward. Growth makes that informal knowledge harder to carry. An embedded phase gives it a structure that other people can use, including an internal TA hire when the time comes.

Hiring started out as something founders did themselves, because there wasn't another way to do it. The question worth asking of any hiring model isn't how well it performed while it was there. It's what the business can still do once it's gone.