Content debt score
Content health checks tell you about individual problems: a stale topic here, a broken reference there. The content debt score rolls those signals into a single 0-10 number so you can see, at a glance, how much accumulated maintenance work a topic (or the whole project) is carrying.
A higher score means more debt. Think of it the way an engineering team thinks about technical debt: not a failure, but a measure of work that has built up and will cost you later if ignored.
Where the score appears
Project dashboard, in the Content health card: the project score, the verdict for it, up to three open issue types, the factor bars, and a Content health report link to the full report.
Analytics ▸ Content health, in the Content debt card: the same score and factor bars, plus the tier legend, alongside health scan recommendations and other health insights.
Scores are read-only signals. Topicary never changes content to lower a score; you decide what to act on.
How to read the card
Both cards show the score as a single number out of 10, with a one-word verdict beside it:
Score | Verdict | What it means |
|---|---|---|
0 to 3 | Healthy | Debt is under control; work the individual issues as they come |
3 to 6 | Fair | Debt is building; plan a maintenance pass |
6 to 10 | Critical | Debt is dominating; prioritize paying it down over new content |
Below the score, three bars break the number down by factor. Each bar shows the factor's own 0-10 value on the right and its share of the score next to the label, for example Quality findings · 50%. A long bar on a 20% factor moves the total far less than a short bar on the 50% factor, so read the percentage before deciding where to spend effort.
Below the bars, a separate line counts open content gaps across the project. It is context, not a factor.
Before the first scoring run, the dashboard card shows a count of open issues and the verdict Needs attention or All clear instead of a score. Scores appear after the first run.
Every topic carries its own score, but no page displays it. Per-topic scores feed the ranking in your Prioritize what to write next, which is where debt turns into a work queue.
The three factors
Each topic's score is a weighted blend of three factors, each measured on its own 0-10 scale. Quality findings carry the most weight: a topic with real open issues always scores more debt than one that is merely old.
Factor | Dashboard label | What it measures | Weight |
|---|---|---|---|
Findings | Quality findings | Severity-weighted unresolved quality findings on the topic | 50% |
Staleness | Content age | Days since the topic was last updated | 30% |
Verification | Verification age | Days since the topic was human-verified | 20% |
A few things worth knowing about how the factors behave:
Findings rise with the severity of open Content health and governance on the topic. A warning counts three times as much as a suggestion, and informational notes count least, so one real problem outweighs a pile of minor notes. Reuse candidates are excluded: content good enough to reuse elsewhere is an opportunity, not debt. Two open warnings max out the factor.
Staleness climbs as a topic ages. It reaches the mid-point around 90 days and maxes out around 180 days.
Verification climbs with time since the last verification. A topic that has never been verified ages from its creation date instead, so a brand-new topic starts near zero and an old never-verified one carries the maximum. Verifying a topic always lowers this factor. See Track content freshness for how to verify a topic.
Content gaps are not a debt factor. Missing content is a property of the project, not of any single topic, so open gaps appear as their own count on the dashboard and analytics cards instead of inflating every topic's score.
The project aggregate
The dashboard card shows a single project-wide score. It is the average of each per-topic factor across all topics, blended with the same weights. This gives you one number to track over time. When it trends up, debt is accumulating faster than you are paying it down.
How the score is recomputed
Scores are recomputed weekly, on Sunday, in the same run that detects content gaps. They are also recomputed on demand whenever a gap-detection scan is triggered for the project.
The numbers reflect your content as of the last run.
How to act on the score
The score is a prioritization tool, not a grade. Use the per-factor breakdown to decide what kind of work will pay down the most debt:
High findings? Open the topic and work through its quality findings in the editor's Findings panel. This is the heaviest factor, so resolving warnings pays down debt fastest.
High content age? Review and update the oldest topics first. Even a small confirmed edit resets the staleness clock. See Track content freshness.
High verification age? Verify accurate topics, especially published ones that have never been verified. Verification is a deliberate human confirmation that content is correct.
Pick the factor with the highest aggregate score and clear that category in a focused pass. There is no sort or filter by debt on the Topics page, so build the queue from the health filter chips (for example Stale (90d+)) or from your Demand worklist, which already ranks topics by their per-topic debt among other signals. Tackling one factor at a time across many topics is more efficient than fixing one topic completely before moving on.
See also
Automated lifecycle recommendations: Nightly scans that turn the same signals into specific, actionable suggestions
Track content freshness: Staleness indicators and human verification, two of the three debt factors
Content health and governance: How automatic health tracking surfaces decay before readers notice