“Time to fill” isn’t one number — it’s at least three, and naming them separates strong candidates from weak ones.

Definitions (pick one and be explicit):

Time-to-fill (TTF). Days from req open to offer accepted. Recruiter-owned metric, reflects pipeline and sourcing effectiveness.
Time-to-hire (TTH). Days from candidate first contact to offer accepted. Candidate-experience metric, reflects interview loop efficiency.
Time-to-start (TTS). Days from req open to new hire start date. Includes offer acceptance and notice period. This is the business-impact number — it’s when the role is actually filled.

The three can diverge significantly. A role with 20-day TTF can have 50-day TTS if offer acceptances sit on long notice periods.

The right cuts (report by):

Role level and function. Senior engineering TTF is structurally longer than entry-level support. Don’t aggregate.
Priority / req tier. P0 roles should be faster than P2. Report both, and compare against target.
Source channel. Referred roles vs inbound vs agency vs recruiter-sourced. Each has different cost, quality, and timeline.
Location / remote-status. US roles vs international differ in both TTF and compliance-driven delays.
Distribution, not average. Median is more honest than mean (one hard-to-fill role can skew). Report P50, P75, P90.

Common traps:

Closed-lost reqs. Reqs canceled or filled internally skew averages. Decide whether to include.
Req re-opens. A role that re-opens because the first hire didn’t start — is that one TTF or two?
Pipeline status effects. Some roles sit open for 60 days waiting for a specific qualified candidate; average drags. Report median and P90 alongside.
Hiring freezes mid-recruitment. A paused req distorts the number dramatically. Exclude or flag.
Benchmarking. External benchmarks exist (SHRM reports median TTF ~36 days across industries, but tech specialized roles run 40–60+). Don’t import a benchmark without checking the peer-set definition.

What to put on the dashboard:

Headline: median TTS, by function.
Trend over 6 quarters (not last quarter alone).
P90 alongside P50 so the tail is visible.
Segmentation by role level, priority tier, and source channel.
Funnel-conversion metrics alongside (% of candidates reaching offer, offer-accept rate) so you can diagnose where the time is going.

How do you compute time-to-fill, and what’s the right way to report it?

Stuck? Peek at a hint