Course › Module 10 · Recruiting analytics and KPIs

Forecasting hiring capacity

Module 10, Lesson 4  ·  4 min read ·  Updated 21 September 2026

Module 10 · Lesson 4

How much hiring can your team actually handle? Working it out prevents most recruiting failures. Almost nobody does it until the team is already drowning.

Counting jobs is the wrong measure

There is no right number of jobs per recruiter, because jobs differ enormously in effort.

A common job with a big pool and two interview stages is not the same work as a rare senior job with five stages and three people who have to agree. Counting them as one each tells you nothing.

Count hours instead. Work out roughly how many hours a week each kind of job takes:

Kind of jobHours a weekWhat eats the time
Common, plenty of candidatesLowScreening and booking
Mid-level specialistMediumSearching and messaging
Rare technicalHighSmall pool. Every person handled by hand
Senior or leadershipHigh, and for longerSlow conversations, many opinions

Get these numbers from your own last three months, not from a benchmark. Then capacity is simple division: hours you have, divided by hours the work needs.

Working it out from the hiring plan

  1. Take the plan. How many hires, of what kind, by when.
  2. Work out how many jobs will be open at the same time. Use your real time to fill — a job starting in June has to open in March.
  3. Apply the hours for each kind.
  4. Compare against the hours you have, minus a realistic allowance for jobs that get reopened, cancelled or changed.

Include replacements. Most plans count growth and forget that people leave, which also creates jobs. That gap is often the difference between a plan that works and one that does not.

What AI changes, and what it does not

Changes a lotBarely changes
Reading and screening CVsThe meeting to define the job
Booking and chasingInterviewer and manager time
Building searchesTalking to candidates
Drafting messages and notesClosing, negotiating, counter-offers
Reporting and adminKeeping managers happy

The left column is real and it is large. The trap is assuming it means one person can now carry twice as many jobs.

The moment screening stops being the slow part, something else becomes the slow part. It is almost always interviewer availability. A team that doubles its screening speed without adding interview time has built a longer queue, not a faster process.

Explaining a limit

Do not refuse. Put a price on it.

"We can run six of these at once with the interviewer time we have. The seventh either waits until March, or we cut the process from five stages to four, or we add another interviewer."

Three options with consequences is a decision someone can make. "We're at capacity" is a complaint, and it will be overruled.

Track one number

Interviewer hours available each week. It is usually the real limit, almost nobody measures it, and it turns an argument about capacity into a fact about diaries.