The kinds of fix, and how to triage

Everything the audit finds lands in one list, and the list is filtered by kind. Understanding the four kinds is how you turn a hundred rows into a plan.

The four kinds

Technical covers whether a page can be read, indexed and shown at all. Broken links, pages telling Google not to index them, canonical tags pointing at the wrong page, a broken sitemap, missing structured data.

Content covers whether the page answers the search it is getting. Thin pages, pages aimed at the wrong kind of search, two pages competing for the same search, a question the page never answers, a topic with no lead page.

Links covers how your own pages point at each other. Pages nothing links to, pages linked only from the menu or the footer, an important page with few links to it, links that say "click here", links that redirect before arriving.

Snippets covers what Google prints in the results page. Titles and descriptions: missing, duplicated, too short, too long, or saying nothing useful.

Each filter carries a count, so you can see what a session would cost before committing to it.

What each is worth in practice

Snippets is usually the best hour on any site. The ranking is already there, the change is text, and one page's title can be rewritten in two minutes. If a page is shown two thousand times a month at position six and almost nobody clicks it, the title is the problem and the title is cheap.

Links is the one lever you fully control. Nobody else decides how your pages link to each other. It is also the most under-done, because it needs a view of the whole site, which is exactly what the audit builds. See internal links.

Technical is mostly a floor rather than a ceiling. Fixing a broken link does not win you traffic, it stops you losing it. Do the urgent ones and move on.

Content is the most expensive and the slowest to pay back, which is why it is worth doing deliberately rather than opportunistically. A thin page on a search with real demand is a good afternoon. A thin page nobody searches for is a page to delete.

How priority is decided

Priority is about what a problem is likely costing you, not how hard it is to fix.

That means some high-priority rows are the cheapest work on the list, and some medium ones are a week. Read the steps before deciding a row is expensive.

Where search data exists, the estimate uses how often the page is shown, where it ranks, and a published curve for how often people click a result at that position. Without Google Search Console connected there is no impression data, so those findings do not appear at all. See connecting Google Search Console.

Grouped rows

Some rows cover many pages: "No main heading, 15 pages", or "Add links between your own pages, 28 pages". <!-- example -->

Those are the ones to start with. The detail panel lists every page affected, so the row is a work order rather than a summary. One template change often clears the lot.

An order worth working in

1. Anything marked urgent. There is usually little of it.
2. Snippets, filtered to the pages with real impressions. Cheapest return on the list.
3. The grouped Links rows, especially pages nothing links to that are already getting traffic.
4. Technical, to clear the floor.
5. Content, deliberately, one page at a time, and only where the search demand justifies it.

Then run an audit and read Since your last audit.

Where to go next

The Fixes tab is the list itself. Everything Rankbench looks for is the full reference of every finding.