Quick answer: A recruitment funnel tracks how many candidates move from each stage of hiring to the next: applied, screened, interviewed, final round, offer, hired. The useful number is stage-to-stage conversion, because it shows exactly where you lose people. Heavy losses early usually mean a job ad or sourcing problem. Heavy losses late usually mean a speed, interview experience or pay problem.
What is a recruitment funnel?
It’s a count of candidates at each stage of your hiring process, read from top to bottom.
Every hiring process narrows. Hundreds apply, a few get hired. The funnel doesn’t tell you that narrowing is bad. It tells you where it happens and whether it’s happening for the right reasons.
What are the stages?
Use your real stages. Here’s a common set for a small team:
- Applied. Everyone who submitted an application.
- Screened. Passed knockout questions and CV review.
- Interviewed. Had at least one interview with the hiring team.
- Final round. Reached the last interview or work sample.
- Offer. Received an offer.
- Hired. Accepted.
If you source candidates directly, add “Sourced” and “Replied” above “Applied”. Keep sourced and inbound candidates separate. They convert very differently.
How do you calculate funnel conversion?
Two numbers for each stage.
Stage conversion = Candidates at this stage / Candidates at previous stage
Share of applied = Candidates at this stage / Candidates who appliedHere’s an example for one role. The numbers are illustrative.
| Stage | Candidates | From previous stage | Of applied |
|---|---|---|---|
| Applied | 180 | 100% | |
| Screened | 36 | 20% | 20% |
| Interviewed | 12 | 33% | 6.7% |
| Final round | 4 | 33% | 2.2% |
| Offer | 2 | 50% | 1.1% |
| Hired | 1 | 50% | 0.6% |
The “from previous stage” column is the one to act on. It isolates where the loss happens. “Of applied” is useful for planning: if you need one hire and 0.6% of applicants get hired, you need roughly 180 applicants.
Where does the funnel usually leak, and why?
Each stage has its typical causes.
Applied to screened is very low
Most applicants don’t fit. That’s a top-of-funnel problem.
- The job title attracts the wrong people. See how to choose a job title.
- The ad is vague about must-haves, so everyone applies.
- One channel sends lots of volume and no fit.
The fix is better targeting, not faster screening. Knockout questions help. Read too many job applications, not enough real candidates.
Screened to interviewed is low
You liked the CV but couldn’t get them into an interview.
- Too slow to reach out. Good candidates are talking to other companies.
- Scheduling friction. Endless email ping-pong.
- Candidates withdrew because the role details, like pay or location, surprised them.
Interviewed to final round is low
Either your screening was too loose, or interviewers are using different standards.
- Check whether interview scores match screening decisions.
- Make sure everyone uses the same criteria. See what a structured interview is.
Final round to offer is low
Common when the team can’t decide. See how to decide who gets the offer.
Offer to hired is low
This is the most expensive leak. Everyone did all the work and the candidate said no.
- Pay below market or below what the candidate expected
- A competing offer that arrived first
- A long, tiring process
- Something in the interviews put them off
Record the reason every time a candidate declines. After five or six declines, the pattern is usually obvious.
What about candidates who drop out on their own?
Track withdrawals separately from rejections. A candidate you rejected and a candidate who left are different problems.
High withdrawals early often mean your application form is too long. See reducing application drop-off. High withdrawals mid-process usually mean you’re too slow or too quiet. See why candidates ghost and how to prevent it.
How do you see your funnel without building it by hand?
You need every candidate’s status changes recorded with dates. In a spreadsheet that’s hard. In an applicant tracking system (ATS) it happens as you move candidates.
In Canvider, the analytics page has a Pipeline Conversion Funnel that shows each stage both “of applied” and “from previous stage”. Custom stages are placed in the funnel by the category you assign them, so set categories carefully. Other panels that help with funnel work:
- Average days per stage shows where candidates wait longest
- Source effectiveness shows applications, hires and conversion per channel
- Offer rejection analysis groups the reasons candidates gave for declining
- Per-job analytics shows applicants over time and by job board for one role
The rule for any tool: the funnel is only as accurate as your status updates. If candidates sit in “Applied” after you’ve rejected them, your numbers are wrong.
How often should you look at it?
Per role, check it weekly while the role is open. Across roles, check it quarterly.
Compare the same period year over year, and give a new channel or process change a full hiring cycle before judging it. A stage with three candidates has no meaningful conversion rate.
For the metrics that sit around the funnel, read time to fill vs time to hire and quality of hire.
Try Canvider free. No credit card required.
Frequently asked questions
What are the stages of a recruitment funnel?
A typical recruitment funnel has six stages: applied, screened, interviewed, final round, offer, and hired. Some teams add a sourcing stage at the top and an onboarding stage at the bottom. Use the stages that match your actual process.
How do you calculate recruitment funnel conversion rates?
Divide the number of candidates who reached a stage by the number who reached the previous stage. If 40 people were screened and 12 were interviewed, screen-to-interview conversion is 30%. Also track each stage as a share of everyone who applied.
What is a good offer acceptance rate?
There's no universal benchmark worth trusting for every market, but a falling offer acceptance rate is always a signal. It usually points at compensation, a slow process, or a gap between what the job ad promised and what the interviews showed.