Main idea

A defect queue doesn't sleep. It requires checking whether the defect still exists, whether it is still reproducible, whether the customer who reported it is still a customer, whether it is still a priority, whether it should jump ahead of another anomaly. You have to reread, requalify, replan, hold the committee, produce the dashboards.

This work is real, recurring, and proportional to the size of the stockpile. It improves nothing. Prioritizing a defect every month isn't handling it: it is accepting that it stays there, and paying for it every month.

Energy then goes to administering defects rather than restoring quality.

A second load accompanies the first, and it shows up on no dashboard: an endless list sustains a permanent tension — everything you haven't done, everything you might do, everything you're keeping just in case, everything sitting there for no clear reason. A queue that keeps growing wears down the people who reread it, independently of the work it contains.

Layer added by "The Backlog Is Not a Dumping Ground: It's a Tool for Action" (2026-06-03).

Why it matters

The cost of a defect queue isn't read in the defects left unfixed. It is read in the bookkeeping they impose — invisible, because it has every appearance of steering: filters, priorities, dashboards, a committee.

It also gives an exit criterion: if maintaining the stockpile exceeds the cost of the fixes it defers, the stockpile has stopped being a trade-off.

Nuances and limits

For a small number of items, in a transition phase, this maintenance stays marginal and the stockpile stays legible.

And part of that work is useful independently of the stockpile: knowing whether a defect is still reproducible also serves the fixing.

Open questions

• Past what size does the periodic rereading of the stockpile cost more than the fixes it defers?