Your backlog has a score. Here's how to read it
Most teams can feel that their backlog is bad but can't say why. A health score turns that feeling into seven specific, fixable numbers.
Ask a team how their backlog is doing and you'll get a shrug. Everyone knows it's messy; nobody can say which part. That vagueness is why cleanups get postponed — you can't schedule work you can't describe.
VectorKan grades each project's backlog out of 100 and breaks the score into named signals, so "our backlog is a mess" becomes "we have 34 stale tickets and 12 with nobody assigned".
The seven signals
- Stale — no activity for 60+ days. The clearest sign a ticket isn't really planned work.
- Aging — created long ago and still untouched in an active column.
- Unowned — sitting in progress with nobody assigned. Work without a name doesn't move.
- Overdue — a due date in the past that nobody has changed. Every one of these teaches the team that dates are decorative.
- Unestimated — no story points, so it can't be planned into a sprint or counted in a forecast.
- Orphaned — no epic or parent, so nobody can tell why it exists.
- Duplicated — near-identical titles filed weeks apart by different people.
What a good score looks like
Don't chase 100. A perfect score usually means the backlog is small because someone just deleted everything, and it'll drift back within a month. What matters is the direction:
- Above 80 — healthy. Plan from it with confidence.
- 60–80 — normal for a working team. Worth half an hour a week.
- Below 60 — planning meetings are being run on bad data, and it shows up as scope surprises mid-sprint.
A backlog at 65 and climbing is in better shape than one at 85 and falling. Trend beats absolute number every time.
Fixing a low score without losing a day
The score is only useful if acting on it is fast. Each signal in VectorKan is a filter you can click, and the backlog view supports bulk actions — so a cleanup looks like this:
- Open the stale list, select everything that isn't a live commitment, and close it in one action.
- Open the unowned list and bulk-assign, or move it out of the active columns.
- Open overdue and clear the dates you're not going to honour.
- Estimate the top twenty only — nothing below that will be planned soon enough for an estimate to survive.
Most first cleanups take under an hour, and the score moves while you work so you can see when you've done enough.
Why this is worth having in the tool
Every one of these signals is computable from data your team already creates. The reason most trackers don't show it isn't difficulty — it's that they're built to record work rather than judge it. Once the judgement is automatic, backlog hygiene stops depending on somebody remembering to care.
Frequently asked questions
What is a backlog health score?
A single number, out of 100, summarising how well-maintained a backlog is. VectorKan calculates it from seven signals — stale, aging, unowned, overdue, unestimated, orphaned and duplicated tickets — and lists the specific items behind each one.
What is a good backlog health score?
Above 80 is healthy, 60 to 80 is normal for an active team, and below 60 means your planning is running on unreliable data. The trend matters more than the number: rising from 65 is better than falling from 85.
How do I improve a low backlog score?
Work the signals in order. Close stale tickets in bulk, assign or demote unowned work, clear due dates you won't honour, and estimate only the top of the backlog. VectorKan supports bulk actions from each filter, so a first pass usually takes under an hour.
Is the backlog health score available on the free plan?
Backlog health scoring is part of VectorKan's analytics. You can create a free workspace to try VectorKan, and the pricing page shows which plan includes each reporting feature.