Candidate experience as a funnel metric, not a vibe
Why CX is now measurable, what the CandE benchmark actually says, and the five drop-off points where 80% of damage gets done.
On this page▾
- The Talent Board's CandE Benchmark tracks 200,000+ candidate responses annually — companies in the top quartile see 38% higher offer-acceptance rates.
- Five drop-off points cause the bulk of CX damage: long unexplained gaps, generic rejections, recruiter ghosting after interview, surprise compensation, and slow offer turnaround.
- Net Promoter Score for candidates (cNPS) is now the dominant single metric — rejected candidates' cNPS matters more than offered candidates' because there are 20x more of them.
- Rejected candidates who get a personal, specific rejection are 2.5x more likely to refer others to the role within 12 months (CandE 2024).
Candidate experience used to be a values statement. Now it's a measured funnel — and most recruiting orgs are leaving 20–30% of their offer-acceptance rate on the table because they treat it as PR rather than process.
Why CX stopped being soft
The Talent Board's annual CandE Benchmark, now in its 13th year, has made candidate experience quantifiable across 250+ employers. The headline finding is consistent: top-quartile CX employers have offer-acceptance rates 30–40% higher than bottom-quartile employers in the same industry, controlling for comp.
The five drop-off points
| Stage | Damage signal | Median fix |
|---|---|---|
| After application | Auto-reply silence > 7 days | Status email at day 3 |
| After recruiter screen | Generic 'we'll be in touch' with no timeline | Concrete next-step date + name |
| After interview loop | Disappearing for 7+ days post-final round | 24-hour decision SLA on the loop |
| At offer | Surprise base/equity below candidate expectation | Comp range disclosed at recruiter screen |
| At rejection | Form letter or 'we went with another candidate' | Specific rejection with optional feedback offer |
Measuring it without a vendor
- Single-question post-stage survey: 'On a 0–10 scale, how likely are you to recommend applying here to a friend?' — that's cNPS.
- Segment by application stage, recruiter, and disposition (offered/declined/rejected).
- Track time-in-stage as a leading indicator — CX score correlates more strongly with time-in-stage than with any qualitative variable.
The rejection letter test
If your rejection email is the same email every candidate gets, you're failing the cheapest CX intervention available. Specificity costs the recruiter four minutes per candidate and produces measurable referral lift in the following 12 months.
Frequently asked questions
Should we give interview feedback to rejected candidates?
If they ask — yes, narrow and behavioral, no diagnoses of capability. The legal risk most TA leaders fear is overstated for honest, behavior-specific feedback; the actual risk lives in 'you weren't a culture fit' which can be construed as protected-class discrimination.
How short can the application be?
LinkedIn data: every additional field beyond 5 reduces completion rate by ~8%. Most ATS-mandated fields (gender, ethnicity, veteran status) can be optional and post-screen — make them optional or remove them from the front door.
Does Glassdoor still matter?
Yes for established companies; less for early-stage. The most useful Glassdoor signal isn't the score but the recency and consistency of CEO responses — that's what candidates actually read.
- Talent Board CandE Benchmark Research — Talent Board
- LinkedIn Global Talent Trends 2024 — LinkedIn
- The Robot-Proof Recruiter (Katrina Collier) — Kogan Page
Read next
All playbooksThe minimum operating system for fair, fast, and predictive hiring at any company size.
A scorecard turns 'I liked them' into 'they demonstrated X'. Here's how to write one that calibrates a whole loop, reduces bias, and survives legal scrutiny.
How to build a multi-channel pipeline that doesn't depend on one job board, how to write outreach that gets replies, and how to measure source quality honestly.