Week 10 Queue Lab: Audit the Priority Lane's usage release

Read the Office's released Priority Lane dataset against its stated purpose, and state plainly what it does and doesn't let you conclude.

26 April 2027

An audit is only as good as its willingness to be surprised by the data — and as honest as its willingness to say what the data can’t settle. This lab gives you the Office’s released dataset and asks you to read it carefully, keeping two different kinds of claim visibly separate.

Before the lab#

Reread your week 6 defence of whichever policy you chose, including your note on how the Priority Lane’s self-declared urgency could be misused.

In the lab#

Office of Everything file — synthetic case material. Released dataset, three months. All requests received, by category: “dispute a printer credit” 60, “collect a form” 600, “enrolment variation” 300, other 40 (1,000 total). Priority Lane declarations, by category: “dispute a printer credit” 8, “collect a form” 122, “enrolment variation” 60, other 10 (200 total). Tickets that declared Priority Lane waited a mean of 3 minutes before being served; tickets that did not waited a mean of 9 minutes — one figure for the whole Priority Lane population, not broken down by category. The lane’s stated purpose at launch was “for the small number of matters that are genuinely time-critical.”

First, the category-share analysis. Work out each category’s share of all requests and its share of Priority Lane declarations, and compare the two. A category that makes up 60% of ordinary demand and 61% of Priority Lane use is using the lane about as often as its size alone would predict; a category whose two shares differ more sharply is over- or under-represented among declarations. Do this for all four categories before deciding whether any of them looks disproportionate.

Second, the aggregate wait-time association, kept separate. The release gives exactly one wait-time comparison, for the Priority Lane population as a whole: declared tickets waited six fewer minutes on average than undeclared ones. Report that figure as what it is — a whole-population average difference, not a per-category saving — and write one sentence explaining why it cannot be attributed to any single category, and why an average difference like this is not, by itself, evidence that declaring urgency caused the six minutes rather than being associated with it for some other reason (for instance, if certain categories are both quicker to serve and more likely to declare).

Third, name what’s missing. Pick at least two specific things this release would need to add before the six-minute figure could be explained rather than just observed: wait times broken down by category, an independent way of verifying whether a declaration was genuinely time-critical, when each request arrived relative to others, how long each category typically takes to serve regardless of the lane, or how consistently staff apply the rule at the counter. For each one you name, say specifically what it would let you check that you can’t check today.

Then write your conclusion, limited to what this release actually supports: a specific statement of which category’s share of declarations differs most from its share of demand, reported alongside the separate, whole-population wait-time association — without claiming a specific category saved a specific number of minutes, without claiming the association is caused by declaring urgency, without claiming any declaration was or wasn’t truthful, and without concluding that the lane measures urgency rather than something else the data doesn’t rule out. The data can support more than one conclusion, including that the lane is running close to as intended.

Afterwards#

As a controlled contrast, open High-declaration Day in the Queue Policy Lab. Compare Quick Enquiries and Priority Lane, naming one ticket whose wait changes and one whose wait does not. Explain why this fixed synthetic experiment cannot establish what caused the audit population’s six-minute association. Declared priority is not a measurement of genuine urgency.

Name one additional field the Office could release next time — a within-category wait-time breakdown, a verified-urgency flag, an arrival timestamp — that would let a future audit test one of the limits you named today.

Keep this audit. Week 12’s redesign of the Office of Everything has to either defend the Priority Lane as it stands, given this data and its limits, or change it — and either choice needs to cite this week’s finding, not simply repeat week 6’s original defence. Week 11 asks what happens to every rule considered so far on the Office’s worst day.