The leaking bucket: why staffing agencies lose candidates they already have

A staffing leader we work with described their recruitment operation to us as a leaking bucket. "You pour more candidates in while you're not picking up on great candidates you might be able to interview already." Their team was spending serious budget on job boards every quarter, while sitting on a database full of people who had already applied, already been screened, and already shown interest in the brand.

That description stuck with us because we've heard variations of it from staffing leaders across the Nordics, the UK, and continental Europe. The bucket has a shape we recognise immediately. Staffing teams collect candidates at the top of the funnel through job ads, social campaigns, referral programmes, and walk-ins. Most of those candidates then drop out of the recruiter's attention the moment the role they applied for closes. The next role opens, the team sources from scratch, and the cycle restarts.

According to our own data we publish on our homepage, recruiters spend up to 70% of their time on repetitive tasks, and 80% of all job applications never get reviewed. Both numbers are symptoms of the same underlying problem. 

‍

What the database actually looks like inside a staffing agency

When we run discovery calls with new staffing customers, recruiters usually describe their database the same way. Tens of thousands of candidate records, often more, built up over years of activity, across multiple ATS migrations, across teams who have come and gone. The data is there. Nobody on the team has the time, the tools, or the search interface to use it.

Recruiters end up working with whoever applied this week rather than whoever is the best fit. Consider the candidate who applied for a warehouse role three months ago, finished a screening call, scored well, and didn't get placed. That person is technically in the system. Practically, recruiters can't find them the next time a similar role opens. Another agency posts the role, that candidate applies there, and the placement happens somewhere else.

We see this leak happen every week in the operations we work with.

‍

Why the leak gets worse in Nordic and European markets

A senior recruiter recently told us: "Enabling finding candidates that can be relevant for a job, even if they haven't applied for that specific or related job in the past, is a true gamechanger." Recruiters in the Nordics work in a different environment from their US counterparts. They see lower volumes of applicants per role. Language requirements and local employment law bind the candidate pool more tightly. Recruiters can't rely on the next wave of applications to bail them out.

For staffing teams in Sweden, Denmark, Finland, and Norway, the database is the asset. When recruiters can't search it properly, they lose every short-turnaround request to whichever competitor can.

‍

What it looks like when an agency stops the leak

OnePartnerGroup is one of the customers we work with on this exact problem. (You can read the full story in our OnePartnerGroup case study.) Their team handles project staffing requests where clients sometimes need a candidate presented inside 48 to 72 hours. Posting an ad and waiting for a pipeline to build doesn't fit that timeline. Searching the existing database does, when the search actually works.

The OPG team described three specific patterns to us during a recent product session:

1. Fast-turnaround project requests. A client emails on Monday and wants candidates by Wednesday. Recruiters need to surface relevant people from the database immediately, without composing a new job ad first.
2. Recurring warehouse and logistics roles. The requirements are simple and well-known. A driver's licence, a forklift certification, availability for shift work. The recruiters at OPG kept hiring the same profile over and over, with no easy way to find the people who had already applied for similar roles.
3. Expired ads holding the right candidates. Roles close, but the candidates who applied for them often fit other open roles on the same team's desk. Without a way for recruiters to search across closed and open roles together, those matches don't happen.

We built those three use cases directly into Talent Finder. Recruiters search the full candidate pool in plain language. No job ad required. No filters to configure first. They ask for what they need, and they get a ranked list back.

‍

The deeper problem sits behind the search box

A search interface only helps recruiters when somebody has structured the data behind it well. This is where most agency databases break down. Candidates apply for a role, fill in basic fields, attach a CV, and disappear into the ATS. The structured information about that candidate, what they actually said in an interview, what they wanted next, what their notice period was, what their salary expectation was, sits in free-text notes, recruiter inboxes, and the heads of consultants who have since moved on.

When our AI screens candidates, we capture that structured information at the moment of the interview. Every candidate answers the same role-specific questions. Recruiters then get those responses parsed into searchable fields and synced back to the ATS. The next time a similar role opens, recruiters can find those candidates based on what they actually said in their interview, even when their CV doesn't mention it.

That's the connection between screening and search. Our team built AI Screening so recruiters can capture the data, Talent Finder so they can search it, and Suggested candidates so they get automatic matches surfaced when new roles open.

‍

What staffing leaders should ask their team

Three questions every staffing MD or operations lead can put to the team on Monday morning:

1. When did we last place a candidate who originally applied for a different role? If the team can't remember, the database is leaking.
2. How long does it take a recruiter to search the database for a specific profile? If the answer involves the phrase "I'd rather just post a new ad," the database isn't usable in practice.
3. What percentage of our screened candidates from the last quarter have we contacted again? If the percentage is low, the agency is paying for those candidates twice.

Those questions tell you the size of the leak. Fixing it doesn't always need new software. Recruiters can sometimes patch it themselves by spending twenty minutes each week reviewing their closed ads against open ones. For staffing operations running at high volume, manual review stops scaling past a certain point.

That's where our team comes in. If you want to see what Talent Finder and AI Screening look like against your own data, book a demo and we'll walk through it with one of your live roles. For more on how we work with staffing teams specifically, see Kiku for Staffing Agencies.