Where Police Applicants Actually Drop Out

If you have been handed the staffing problem and told to come back with answers, start here. Three things hold in every agency that has published the numbers.

• Your biggest losses happen where nobody screens anybody. People just stop responding — mostly because you were slow.
• Testing is not the bottleneck. Getting people to show up for testing is.
• More applications may not fix it. Volume has already climbed in some of the largest departments, and agencies are still short.

Where the people go

A city auditor tracked every police application filed with one large metropolitan department from 2017 through September 2020. 12,129 went in. 171 officers came out.

That department is not alone. A West Coast city department counted 3,008 applicants in one year and graduated 40 — 1.3%, though that one counts graduates rather than hires.² Somewhere around 1 in 70 to 1 in 75 is the honest range.

If your number looks far better, check what you are counting. Many agencies count qualified applicants. Counting every application that came in is what tells you the truth.

There are only two ways to lose a candidate

Every candidate you lose leaves one of two ways. Either you turned them down, or they stopped talking to you. Those look identical in a headcount report. They need opposite fixes.

This is the part that changes what you do about it.

Six out of ten candidates that department lost were never rejected by anybody. 7,197 people filled out a full application and then never completed the background paperwork.

And 60% is the floor. The screened-out group also contains applications the audit filed as "incomplete or did not meet requirements" — and incomplete is abandonment too. Nobody knows the real number, because nobody separated them.

That second group is the cheapest thing in your pipeline to recover, because they already chose you once.

Why they go quiet

Counting tells you how many walked away. It does not tell you why. These are the five reasons that come up most often when we audit an agency’s recruiting operation, ranked by how often we find them.

Every one of the top four is inside your control, and none requires lowering a standard to fix. The delays are not always small, either — one big-city inspector general found 2,400 applicants still sitting at the background stage years after they tested, and nearly 60% of applications that died there had never gotten a rejection letter.⁶

The fifth reason is the one agencies underrate, because it does not look like a drop-out problem at all. It is a waste problem. Every hour spent working a candidate you were always going to disqualify is an hour not spent on someone you would have kept — which is exactly what makes an agency slow for everybody else. Screening earlier is not about being stricter. It is about being faster for the people who are going to make it.

The number that surprises people

Another large metro department invited 9,617 applicants to civil service testing. 2,717 showed up — 28%. Of the ones who did, 84% passed.³

Almost everyone who walks in the door passes. Getting them through the door is the problem. If your written exam is under review as a staffing fix, you are working on the wrong step.

The stages, in order

There is no single healthy background pass rate, and anyone who hands you one made it up. What matters is whether yours holds steady and whether your disqualification reasons cluster. If one reason drives most of your failures, that is a sourcing problem your background unit is paying for.

Background is slow because the work is genuinely hard, and no software changes that. But a candidate parked in a queue for a year is gone either way.

The academy

Nationally, more than 14% of recruits do not finish: 8% pushed out, 5% who quit, about 1% other.⁵ Split your own the same way — the two point in opposite directions.

More applications will not fix this

This is worth knowing before you propose a bigger marketing budget.

Hiring has recovered. In 2025, surveyed agencies hired 40% more officers than in 2021, and staffing is back to 2022 levels.⁷ Applications recovered too — one of the largest departments in the country went from 8,408 applications in 2022 to 13,047 in 2024.⁴

Hiring went up. Staffing did not close. The days did not shrink.

Find your own number

You do not need software to do this once. You need one person, one day, and whatever system holds your applicants.

1. Pick one batch. Every application from a single quarter at least a year ago, so they have all finished one way or another. Not a snapshot of who is in the pipeline now — that flatters you, because everyone you already lost has left it.
2. Count who is left at each step. Complete application, showed up to test, passed, entered background, cleared it, got an offer, started academy, graduated, finished field training, still here today.
3. Divide each step by the one before it. That is your drop-off. Divide the last number by your original applications for the whole-pipeline rate.
4. Split every loss into rejected, withdrew, or went quiet. This is the step almost everyone skips and the only one that tells you what to fix.
5. Write down how many days people sat at each step. Use the middle value, not the average — a few ancient files drag an average until it means nothing.

One pass gives you a baseline and a one-page answer for your chief. The hard part is repeating it monthly, for every recruiter and lead source, without somebody rebuilding a spreadsheet each time.

What a weak stage is telling you

One caution. Tighten who you let in and every rate on this page improves while you hire the same number of officers. Always read drop-off next to your actual hire count.