The problem
A recruitment agency was growing but struggling to scale. Each recruiter could manage approximately fifteen active roles, limited by the time required for sourcing, screening and communication. As demand increased, the agency faced a choice: hire more recruiters or find a way to increase recruiter capacity.
The administrative work surrounding recruitment — searching job boards, screening CVs, scheduling interviews, sending updates — consumed time that could have been spent on the high-value activities: assessing fit, building relationships with candidates and clients, and closing placements.
What we built
We built an AI recruitment assistant that handled the administrative aspects of the recruitment workflow:
- Candidate sourcing. The assistant searched job boards, LinkedIn and the agency's internal database for candidates matching role requirements, producing prioritised shortlists for recruiter review.
- CV screening. Applications were screened against role criteria, with qualified candidates flagged for review and borderline candidates identified for recruiter assessment.
- Communication management. The assistant handled routine candidate communication — application acknowledgements, status updates, interview confirmations, feedback requests.
- Interview coordination. Scheduling across multiple calendars and managing changes was handled automatically.
The results
Six months after deployment:
- Roles managed per recruiter increased from fifteen to twenty-two — a capacity increase of nearly fifty per cent
- Time to fill reduced because sourcing and screening were faster
- Candidate experience improved through consistent, timely communication
- Recruiter satisfaction improved because they spent more time on the relationship-building and assessment work they found most rewarding
How it worked
The assistant did not make hiring decisions. It handled the administrative work that surrounded the decisions — finding candidates, screening for basic qualifications, managing communication and coordination. Recruiters focused on the work that required their expertise: assessing cultural fit, understanding candidate motivations, managing client relationships and negotiating offers.
The assistant improved over time as it learned which candidates the recruiters selected and which they passed on. Its sourcing and screening became more aligned with what the agency's recruiters actually valued.
What we learned
The most important lesson was that recruitment automation works best when it is designed to support recruiters rather than replace them. The recruiters were initially sceptical. Their scepticism evaporated when they saw that the assistant handled the work they found least valuable — the administrative coordination — and freed them for the work they found most valuable — the human interaction.
We also learned that candidate communication consistency has a significant impact on candidate experience. When every candidate received timely, professional communication, the agency's reputation improved and candidate engagement increased.
For a broader look at how AI supports recruitment, see AI automation for recruitment. For an overview of the assistant category, see AI recruitment assistant.
This case study describes a composite of real implementations. Results vary based on the specific process, team and context.