Working the Action Plan
The Action Plan is Rubric's ranked to-do list. It turns the checks your pages failed into a phased sequence of fixes, ordered by how much score each one moves per hour of work, not by page. It runs in three phases: Phase 01 Foundation (days 0 to 30), Phase 02 Structure (days 30 to 60) and Phase 03 Polish (days 60 to 90). Work it top to bottom.
- →The Action Plan is built from the 44 checks. It ranks the failures and warnings into an ordered list of fixes.
- →Fixes are ranked by score movement per hour of work, not by page, and grouped into three phases: Foundation, Structure, Polish.
- →Template fixes come first, because one change to a shared template can clear a check on every page at once.
- →To see the score move after a fix, clear your CDN cache, then re-run the crawl or wait for the Watch-list re-check.
What the Action Plan is
The Action Plan is the part of your report that tells you what to do next. Rubric runs 44 scored checks across the three pillars, Known, Findable and Trusted, and every check that fails or warns becomes a candidate fix. The Action Plan collects those candidates, ranks them, and groups them into phases so you always know the single highest-value thing to do with your next hour.
It is a work queue, not a page list. The report has a separate Pages view for looking at individual URLs against the 70 line. The Action Plan answers a different question: given everything that is wrong across the whole site, what should I fix first.
How the three phases work
The plan is phase-tiered, not a flat "1 of 30" list. Each phase is a block of work sized to roughly a month, and the phases are ordered by leverage.
| Phase | Window | Name | What sits here |
|---|---|---|---|
| Phase 01 | Days 0 to 30 | Foundation | The highest score-per-hour fixes: entity and schema basics, crawler access, sitemap, the answer-first and structural wins that unblock everything else |
| Phase 02 | Days 30 to 60 | Structure | Deeper structural work across templates and key pages: headings, self-contained sections, tables and lists, internal linking |
| Phase 03 | Days 60 to 90 | Polish | The last refinements: freshness, author attribution, sourced statistics, review schema, then re-crawl and compare |
Foundation comes first because those fixes both move the most score and remove blockers. An engine that cannot resolve who published a page (a Known failure) is reluctant to attribute a quote to you, so fixing entity clarity early raises the ceiling on your Findable and Trusted work.
Why fixes are ranked by score movement per hour
Rubric ranks fixes by estimated score movement per hour of work, so the top of the list is always the best use of your time, not the lowest-scoring page. A change that touches one template and clears a failing check on 400 pages ranks above an edit that fixes a single URL, because it lifts far more of your median for the same effort.
This is why you should never start with your worst page. Your site score is the median of your page scores, so raising it means lifting the middle of your site. The fixes that do that are usually template fixes, and the Action Plan surfaces them at the top.
How estimated lift works
Each fix shows a projected score gain, the "+N" on the row, along with a per-engine gain beneath it. That number is how far your site score should rise if the fix passes on every page it affects. The projected total at the top of the plan is today's score plus the summed per-fix gains, capped at a believable ceiling. Lifts do not simply add up: once most pages clear the 70 line, the median moves more slowly, so the last few fixes gain less than the first. Treat every number as an estimate of movement, not a promise of a citation or a ranking.
Work template fixes before page fixes
A template fix is a single change to a shared layout or config that applies to every page built from it. A page fix changes one URL. Template fixes come first because they clear a check everywhere at once, which is exactly what moves a median. Adding Organization schema to the shared <head>, allowing the AI crawlers in robots.txt, or publishing an XML sitemap are all one-and-done template fixes. See template-fixes-vs-page-fixes for how to tell them apart.
How to see your score move after a fix
Rubric reads the live page, so the score only changes when the page changes and you re-check it. There is a specific order, because caching will otherwise hide your work.
There is no automatic weekly re-crawl of sites you have not added to your Watch list, so re-run manually when you want to see a change straight away.
What the plan does not do
The Action Plan is a ranked, exportable checklist that reflects your last crawl. It is honest about what it is, so it helps to know what it is not.
- There is no "mark as shipped" button, and marking a fix does not trigger a re-crawl within the hour. You confirm a fix by re-running the crawl.
- The per-fix "+N site score" gain is an estimate, not a promise. It projects how far your median should move if the fix passes on every affected page. It is not a guarantee of a specific score, a citation or a ranking outcome.
- You cannot reorder or hide fixes, or mark one as "not for this site". Work the order given, or export the plan and manage it in your own tracker. See handing-fixes-to-developers to export it as a developer checklist.