The Permit-Review Metrics That Actually Matter

Executive summary

Building departments are usually measured on one number: average time from application to permit. It is the easiest number to produce and the easiest to distort, because it can be improved by deferring intake, splitting applications, or returning files early.

A defensible measurement set covers five dimensions and names the distortion each metric invites. Used together they show whether the service improved, whether quality held, and where capacity actually goes.

Why one number is not enough

Permit timeliness is politically visible; housing programmes and municipal action plans reference approval performance [1]. That visibility is exactly why the measure needs company: an unaccompanied timeliness target rewards behaviours that shift work to applicants or defer the clock.

National permit statistics provide context for volume and mix but not for internal performance [2]; the operational measures have to come from the department’s own stage data.

A balanced measurement set

Stage data is the prerequisite

None of this works without stage timestamps. Recording only receipt and issuance makes it impossible to separate departmental time from applicant time, which is the first question any external challenge will raise.

Implications for authorities having jurisdiction

Choose no more than six reported measures and keep them stable across years. A metric set that changes annually cannot show a trend, which is the only thing council actually wants to see.

Sample touch time rather than estimating it: two weeks of recorded review time per permit type, twice a year, is enough to plan capacity and to cost a business case.

Implications for applicants and industry

Applicants should ask for distributions rather than averages. A department with a median of three weeks and a 90th percentile of four months has a predictability problem that an average conceals.

First-submission completeness is a shared measure: it reflects both the clarity of published requirements and the discipline of the submission.

Risks, limitations and safeguards

• Any single metric under target pressure will be optimised at the expense of the others; publish the set together.
• Reversal and post-issuance measures depend on honest recording; protect the reporting culture that makes them possible.
• Small-volume permit types produce noisy statistics; aggregate carefully and state sample sizes.
• Automation metrics such as extraction accuracy are not service metrics; keep them in a separate technical set.
• Do not publish improvement claims attributable to a tool without a documented baseline and comparison method.

PermitAssure perspective

PermitAssure records stage transitions, deficiency causes, findings and their dispositions, so this measurement set becomes available from operational data rather than from a separate reporting exercise. PermitAssure publishes no performance figures on behalf of a jurisdiction; the jurisdiction owns and reports its own numbers.