The Issues tab: errors and warnings
The Issues tab lists every check a page does not fully pass, marked as an error or a warning and grouped under the pillar it belongs to. An error is a check that blocks citation; a warning is one that weakens it. Each issue carries an evidence line: the exact thing Rubric found on the page that made the check fire.
- →Errors are checks that block citation. Warnings are checks that weaken it. Both count against your score; errors carry more weight.
- →The evidence line behind each issue is the specific finding on the page, so you can act on it, not just read the rule.
- →Issues group under Known, Findable and Trusted, and the tab flags your weakest pillar.
What is the difference between an error and a warning?
In the Issues tab, every failing check is marked as either an error or a warning. An error is a check the page fails outright, so an engine has a concrete reason not to quote it. A warning is a check the page only partly meets, so it weakens citability without blocking it.
Both count against your score, but errors carry more weight, because they describe the signals that stop a page being cited at all. The header of the tab counts them for you: how many checks are failing, how many are errors, and how many are warnings. Work errors before warnings, and template-level errors before either.
What is the evidence line?
The evidence line is the specific thing Rubric found on the page that made a check pass or fail. It is the difference between a rule and a finding. Instead of only telling you that a check exists, the evidence line quotes what Rubric actually read: the block of schema that is missing, the opener that runs too long to lift, the statistic with no source next to it.
That is what makes an issue actionable. A rule tells you what good looks like; the evidence line tells you where your page falls short of it. Each issue pairs the evidence with a fix, so you can go straight to the change. To understand why a given check earns its place, see where every check comes from.
How are issues grouped and filtered?
Issues are grouped under the three pillars, Known, Findable and Trusted, so you can see which part of citability is weakest. Each pillar section shows that pillar's score and how many of its checks are failing, and the tab flags your weakest pillar so you know where to start.
Three filter pills sit at the top:
The tab also surfaces which checks fail on the most pages and a short list of your worst pages by failing checks, so a single tab tells you both what is wrong and where it is most common.
Why are some checks not listed?
The Issues tab lists only checks a page does not fully pass. Checks that pass on every crawled page are not shown; the tab notes how many passed so you know they were run. This keeps the list to the work that is left, not the work already done.
Two things are also kept out of the scored issues. Informational signals such as llms.txt are checked but not scored, so they do not appear as errors or warnings. And advisory views such as Agent-ready and Common Crawl presence are shown in their own tabs, deliberately outside your Rubric score, because they are forward-looking rather than settled citation signals.
How do I turn an issue into a fix?
Every issue in the tab pairs its evidence line with a fix, so reading an issue and acting on it is one step. The fastest wins are template fixes: a single change to a shared template can clear the same check on every page at once, which is why the tab highlights the one change that touches the most pages.
Once a fix is live, clear your CDN cache and re-run the crawl. Rubric reads what an engine reads, so a cached page re-scores the old markup and the issue will still show. See finding your weakest pages to spot which pages carry the most failing checks.