Where the 63 hours in a 1,000-applicant hire actually go

One vacancy, a thousand applications. Worked line by line against published sources, that comes to 63.5 hours of someone's attention. This article shows every line, links the source it came from, and then does the same arithmetic on a 200-applicant role to check the model against an independent estimate of what a standard hire costs. Every figure below is either linked to its publisher or marked as our own assumption.
First, the one number only you can supply
An hour of your recruiter's attention has a price, and nobody publishes it. We have refused to benchmark it before and we are not going to start: the salary aggregators that come up in search are restating national averages at second hand, not what your HR coordinator costs you.
So this article picks one example and shows its working, the same way:
| Input | Example used here |
|---|---|
| Monthly cost of the person doing the hiring | Tk 20,000 |
| Working hours in a month | ÷ 176 |
| Cost of one hour of their attention | ≈ Tk 114 |
Tk 20,000 is an illustration, not a benchmark. It is roughly a junior coordinator. Put your own salary in and every taka figure below moves; the hours do not.
The model: 1,000 applications, one hire
Each row is a stage, a per-unit time from a published source, and a count. The counts are the shape of a funnel and are stated as assumptions where no source sets them.
| Stage | Per unit | Count | Hours | Where the per-unit time comes from |
|---|---|---|---|---|
| Initial scan | 30 sec | 1,000 | 8.3 h | StratusHR: 30–90 seconds per resume. We take the bottom of their range. Corroborated by ResumeGo's 2024 survey of 418 hiring professionals and by Jan Tegze's 2023 tracking study of 114 recruiters. |
| Detailed review | 3 min | 200 | 10.0 h | 200 is our assumption: one in five survives the scan. StandOut CV puts rejection at the initial screening stage higher than that, so we are being deliberately conservative — see the note below. |
| Pre-interview contact | 10 min | 100 | 16.7 h | StratusHR: 5–15 minutes per applicant. Midpoint. |
| First interviews | 30 min | 40 | 20.0 h | StratusHR: 20–40 minutes per applicant per interview. Midpoint. |
| Team discussion | 30 min | 15 | 7.5 h | StratusHR: 20–45 minutes per applicant. Below their midpoint. |
| Posting and admin | — | — | 1.0 h | StratusHR: 20–30 minutes per job, plus the admin around it. |
| Total | 63.5 h | × Tk 114 = Tk 7,240 |
A note on that second row, because it is ours and not a source's
We say 200 of 1,000 get a proper read. StandOut CV's statistics page is harsher than that in every direction — it cites LinkedIn for 70% of resumes rejected at the initial screening stage, and its own shortlisting data implies that only around one applicant in ten reaches a shortlist at all.
We chose the gentler figure on purpose. A bigger detailed-review pile would make the manual column more expensive and our saving look better. Taking 20% instead of 30% costs us five hours of the comparison. If this assumption is wrong, it is wrong against us.
The 7.4 seconds you have read everywhere is not screening time
You will have seen the claim that recruiters spend six, or seven and a half, seconds on a CV. It is used constantly to argue that screening is already so fast there is nothing left to save. It does not mean what it is used to mean, and this is worth getting right.
The number comes from an eye-tracking study — “Eye Tracking Online Metacognition: Cognitive Complexity and Recruiter Decision Making” by Will Evans at TheLadders, thirty professional recruiters observed over a ten-week period, updated in 2018 to 7.4 seconds.
Read what the study says it measured. In its own words, its “gaze tracking” technology showed that recruiters spent about six seconds on their initial “fit/no fit” decision. That is the duration of a look, captured by apparatus that records where eyes land and for how long. It is the moment of first judgement, not the task.
The task is: open the document, read enough to judge it, decide, record the outcome, close it, open the next one. Three independent measurements of that agree with each other and disagree with seven seconds:
- StratusHR estimate 30–90 seconds per resume for HR planning purposes.
- ResumeGo surveyed 418 US hiring professionals between January and March 2024: 47% said 30 seconds to one minute, 33% said 10–30 seconds, and only 1% said under 10 seconds.
- Jan Tegze tracked 114 recruiters across three CVs in 2023 and recorded averages of 17, 27 and 46 seconds — the shortest single view 12 seconds, the longest 2 minutes 27.
So this article uses 30 seconds: the bottom of StratusHR's range, the bottom of ResumeGo's modal band, and below two of Tegze's three averages. Not seven seconds, which measures a glance, and not four minutes, which is what recruiters say when asked to remember.
The same role, screened by software
The second column is our Starter plan at this exact volume: 1,000 applications, 520 CVs fully read and scored, 52 AI interviews. We have costed it at Tk 2,000 rather than the published Tk 1,700, rounding up so the comparison errs against us.
| Stage | Human hours | What changed |
|---|---|---|
| Initial scan | 0 h | All 1,000 are read and scored by the platform. |
| Detailed review | 0 h | 520 get a full read rather than 200. Nobody is rejected by it — it produces a ranked pile with reasons. |
| Pre-interview contact | 0 h | Invitations go out automatically. |
| First interviews | 0 h | 52 interviews conducted by the platform instead of 40 by a person. |
| Reading the interview reports | 2.6 h | 52 reports × 3 minutes. New work that did not exist before. |
| Final interviews | 2.5 h | 5 finalists × 30 minutes, met by a person. |
| Team discussion | 7.5 h | Unchanged. 15 candidates × 30 minutes, exactly as above. |
| Posting and admin | 1.0 h | Unchanged. |
| Total | 13.6 h | × Tk 114 = Tk 1,550, + Tk 2,000 platform = Tk 3,550 |
Look at the bottom two rows before the totals. The hours in which the decision is actually made do not move. The team still spends seven and a half hours talking about fifteen people, and a person still sits across from the finalists. What collapses is the reading, the chasing and the scheduling — the part that was never judgement in the first place.
What that comes to
| By hand | With screening software | Difference | |
|---|---|---|---|
| Hours of someone's attention | 63.5 h | 13.6 h | 50 hours saved (79%) |
| Cost at Tk 114/hour | Tk 7,240 | Tk 3,550 | Tk 3,690 saved (51%) |
| CVs given a full read | 200 | 520 | 2.6× the depth |
| Candidates interviewed | 40 | 52 | 12 more |
Fifty hours is more than a working week. At the volumes this article describes it is also roughly three weeks of calendar time, because those hours are not available in one block — they are spent in the gaps between everything else the same person is doing.
The percentage saving on cost depends on the salary you put in, because the platform fee is fixed while the hours are not. The hours saved — 79% — do not depend on it at all.
Sanity check: run the same model on a normal role
A thousand applications is a lot, and a model tuned to an extreme can quietly stop describing anything real. So run the identical method on a 200-applicant role, where an independent published estimate exists to check it against.
| Stage | Per unit | Count | Hours |
|---|---|---|---|
| Initial scan | 30 sec | 200 | 1.7 h |
| Detailed review | 3 min | 40 | 2.0 h |
| Pre-interview contact | 10 min | 30 | 5.0 h |
| First interviews | 30 min | 15 | 7.5 h |
| Team discussion | 30 min | 8 | 4.0 h |
| Posting and admin | — | — | 1.0 h |
| Total | 21.2 h |
The later stages do not shrink in proportion, and they should not: you interview the number of people you need to meet, not a fixed percentage of whoever applied.
Now the check. StratusHR say a company devotes “around 30 hours over 4 weeks” to recruit one employee, and closer to 40 for senior roles. Our 21.2 hours covers fewer stages than their 30 does — we never counted reference checks, background checks or post-interview contact. Add those back at StratusHR's own published rates:
| Our model, 200 applicants | 21.2 h |
| + post-interview contact (5–15 min × 15) | 2.5 h |
| + reference checks (20–30 min × 3) | 1.3 h |
| + background and social checks (20–30 min × 3) | 1.3 h |
| Comparable total | 26.3 h |
| StratusHR's independent figure | ~30 h |
Two routes to the same place, from different directions, landing within a few hours of each other. That is the reason to believe the thousand-applicant figure: the method that produced it also reproduces somebody else's number for an ordinary hire, without being tuned to.
What this article is not claiming
Three things, stated plainly, because the arithmetic above is easy to stretch:
- Nothing here replaces human judgement. Team discussion is identical in both columns — 7.5 hours, 15 candidates — and a person still interviews the finalists. The stages that collapse are reading, chasing and scheduling.
- Nobody is auto-rejected. The platform produces a ranked pile with reasons attached; a person decides every outcome. We have written separately about why, and the product is built that way rather than configured that way.
- The savings are 79% of hours and 51% of cost, at this volume, with these assumptions. Not more. Change the volume, the salary or the funnel shape and the numbers move — which is why every input above is shown rather than summarised.
Sources
- Per-stage time estimates and the ~30 hours per hire: StratusHR, Estimated Time on HR: Recruiting Tasks.
- Resume review time survey, 418 hiring professionals, January–March 2024: ResumeGo.
- Tracked review times, 114 recruiters, 2023: Jan Tegze.
- The eye-tracking study behind the six / 7.4 second figure: Will Evans, TheLadders (PDF); 2018 update.
- Screening-stage rejection and shortlisting rates: StandOut CV resume statistics.
- Our own plan volumes and price: TopCandidate pricing.
All links checked 23 September 2026. If a figure has moved since, trust the source rather than this page.
Questions we get asked
How many hours does it take to hire one person?
Worked stage by stage against StratusHR's published per-task estimates, a 1,000-applicant vacancy comes to 63.5 hours. The same method on a 200-applicant role gives 21 hours, which lands close to StratusHR's own ~30 hours per hire once reference checks and post-interview contact are added back.
Do recruiters really spend only 6 or 7 seconds on a CV?
No. That figure is from an eye-tracking study and measures gaze duration during the initial fit/no-fit judgement, not the screening task. Measured end to end, ResumeGo's 2024 survey puts most reviews at 30 seconds to a minute and Jan Tegze's 2023 tracking recorded averages of 17 to 46 seconds.
How much time does AI CV screening actually save?
In the model in this article, 50 of 63.5 hours — 79% — at a cost saving of 51%. The hours that disappear are reading, chasing and scheduling. Team discussion and the final interviews are identical in both columns.
Read next
What it actually costs to hire one person in Bangladesh
Job posting prices, measured screening costs, and the one line item only you can price. A worked breakdown rather than a borrowed average.
High-volume screening500 applications and one person: how to actually screen them
Reading them by hand is 25 hours. Our measured batch rate is 800 CVs an hour. What gets faster, what does not, and what breaks.
Cost of hiringCost per hire vs cost per screen — the two numbers HR confuses
They used to move together and no longer do. What each contains, the trap that sets, and what to measure instead.
High-volume screeningWhy the best CV in the pile often never gets read
In three of five real job postings the top-scoring CV arrived in the second half of the pile. What reading in arrival order actually costs.
AI in hiringWhy an AI should never reject anyone by itself
A CV score is evidence, not a decision. What an AI genuinely cannot see in a CV, and why a person should press the button.