Recruitment Agency Metrics: 5 Measures for Better Decisions

Five recruitment agency metrics that help you decide what to fix

Useful recruitment agency metrics connect an observable result to an action the team can take. Define the measure, its denominator and the group being measured before comparing performance. A number becomes useful when you can explain what changed and decide what to investigate.

Calls made and emails sent describe activity. They do not, on their own, explain whether a search is progressing or a relationship is being handled well.

The five measures below are a suggested operational set. They are definitions to adopt or adapt, not industry benchmarks or claims about a particular software dashboard.

1. Time to first qualified shortlist

Define this as the elapsed time between an agreed, actionable brief and the first shortlist that meets the documented criteria for client review.

Choose whether you measure working hours or calendar time, and use the same convention throughout. Mark assignments whose scope changes materially rather than pretending their timing is directly comparable.

If the measure rises, inspect where work waited: intake clarification, candidate responses, screening or internal review. The number does not tell you which stage caused the delay.

Use the median alongside the number of assignments observed. One unusually difficult search can distort a simple average, while a tiny sample may not support a strong conclusion.

2. Existing-database contribution

For a defined period, calculate:

Shortlisted candidates first identified through existing agency records ÷ all shortlisted candidates × 100.

Set an attribution rule before counting. For example, define “existing” as present before the assignment opened, and count each candidate once per assignment.

An illustrative month with 12 such candidates among 40 shortlisted candidates produces a 30% contribution. That is arithmetic, not a target or proof of quality.

If the measure is low, investigate search habits and data relevance. A new specialism might reasonably rely more on fresh sourcing than an established desk.

3. Shortlist decision time

Measure the time between a candidate submission being shared and the client recording an actionable decision, such as interview, decline or a specific request for more information.

Separate those decision types when reviewing the results. A request for clarification advances the process differently from an interview invitation.

Also show candidates still waiting and their current age. Looking only at completed decisions can hide the longest unresolved cases.

If delays cluster around a client, review the agreed decision owner and response process. If they cluster around missing information, improve the handover.

4. On-time follow-up completion

Calculate completed follow-ups on or before their due date divided by follow-ups due during the period.

Specify how cancellations and rescheduled tasks are treated. Review overdue tasks that were repeatedly moved rather than allowing date changes to make the process appear healthier.

The aim is reliable commitments. An agency can score well by setting very loose dates, so sample whether those commitments were useful to candidates and clients.

5. Submission-to-interview conversion

Use a defined submission cohort: the proportion of submitted candidates who receive an interview invitation within your chosen observation window.

Track candidates still awaiting decisions separately. Do not divide this week’s invitations by this week’s submissions if the invitations came from older work.

Interpret the result alongside assignment difficulty, agreed criteria and client feedback. A change could reflect stronger screening, a different mix of roles or delayed decisions.

Use the measures in a short review

Choose one change to investigate each week. Look at the underlying records, agree a likely explanation and decide on a small operational adjustment. Revisit it after enough comparable work has passed through the process.

Avoid turning every movement into a recruiter ranking. These measures are most useful when they expose a process the team can improve.

Bring your reporting definitions to a Connekt demo and ask which measures your proposed setup can support.

A common measurement question

What is a good benchmark? Start with your own clearly defined baseline and comparable work. External numbers are only useful when their definitions, time windows and recruitment context match yours closely enough to support a decision.