Your Site Audit Found 400 Issues. Here's What Matters

Your first site audit finishes and hands you 400 issues. Forty-one are red, another hundred and something are orange, and the rest are yellow. They are sorted by type, which is to say sorted by nothing you can act on.

You fix a few of the reds, lose the thread somewhere around “missing meta description” on page 60, and never open the report again. That is the standard outcome of a first audit, and it is a problem with how the list is presented rather than with your attention span.

There is a sort order that makes it tractable, and it comes from Google rather than from any tool.

Sort by the stage it breaks

Google’s description of how Search works sets out three stages, in order: crawling (Google downloads the page), indexing (Google analyses it and stores it), and serving search results(Google returns it for a query). The same page adds the sentence the whole of this post rests on: “Google Search works in three stages, and not all pages make it through each stage.”

The stages are sequential, and that is what turns them into a priority order. A page Google cannot fetch will never be indexed, so nothing you do to its title matters. A page Google fetches but declines to index will never be served, so its structured data is decoration. Work the stages in order and most of the 400 issues stop being urgent — not because they don’t matter, but because they cannot pay off yet.

Almost every audit tool flags all three at similar visual severity. A blocked directory and a short meta description both get a coloured chip, and the report gives you no way to tell that one of them is a fire.

Stage one: can Google fetch it?

These are the emergencies. Every one of them makes a page invisible regardless of its quality, and there are usually very few of them.

• Blocked in robots.txt.Google notes plainly that “some pages may be disallowed for crawling by the site owner”. A rule intended for a staging path that also matches a live directory is the classic version, and it is silent.
• Server errors. A 500 on a template affects every page using it.
• Redirect chains and loops. A loop is fatal; a long chain is waste and risk.
• Pages with no internal links pointing at them.If nothing links to it and it isn’t in a sitemap, discovery is left to chance.

If your report has anything in this group, it is the whole of this week’s work. Everything below waits.

Stage two: will Google index it?

The subtler group, and the one that swallows the most effort when it’s missed, because these pages look completely healthy in a browser.

Google is direct that this stage is not automatic: “Indexing isn’t guaranteed; not every page that Google processes will be indexed.” Some of the reasons are your doing:

• A noindex robots meta tag.Google lists “robots metarules disallow indexing” among the reasons a processed page doesn’t make it in. The dangerous case is a tag left on a template after a staging deploy — it renders perfectly and quietly excludes a section.
• A canonical pointing somewhere else.If every product variant canonicalises to a category page, you have told Google the variants aren’t the page to keep.
• Duplicate or near-duplicate pages. Several URLs saying the same thing leaves Google to pick one, and it may not pick yours.
• Soft 404s.A page returning 200 with “no results found” on it.

Anything in this group outranks every stage-three item in your report, however minor it looks in the interface.

Stage three: how well is it served?

This is where the other 380 findings live. They are real, they are worth doing, and none of them is an emergency.

It is also the group where Google’s answer stops being binary. Relevancy, it says, is “determined by hundreds of factors, which could include information such as the user’s location, language, and device” — so a missing meta description does not cost you a fixed amount. It shifts a probability.

• Titles and descriptions. Missing, duplicated, or long enough to be cut. Worth checking what a snippet actually looks like at the width Google renders it — our free snippet preview measures the pixel width rather than counting characters, which is the measure that decides where the cut lands.
• Hreflang problems on international sites.Broken annotations fail silently: Google ignores what it can’t use and reports nothing. The free hreflang generator validates the codes and builds the tags, including the self-reference people miss.
• Heading structure and internal linking. Cheap to fix, genuinely useful, never urgent.
• Structured data warnings.Errors that block a rich result are worth attention; warnings about recommended properties usually aren’t.
• Image alt text. Do it for accessibility, which is the better argument anyway.

The site audit triage table

The common findings, sorted the way the report should have sorted them.

Reachavo’s Site Audit groups its findings on this model rather than by issue type, so the crawl-blocking and index-blocking problems arrive at the top instead of alphabetically among the alt-text warnings.

The two that look like stage three and aren’t

Two findings get filed as cosmetic and belong much higher.

Duplicate titles across a section.Read as a snippet problem, it’s a backlog item. Often it’s the visible symptom of near-duplicate pages, which is a stage-two problem about which URL Google keeps.

Several of your own URLs competing for one query.No crawler can see this, because it isn’t a property of your HTML — it only shows up in performance data. It is also a common reason a technically perfect page won’t move. Our free cannibalisation checker reads a Search Console export and shows where the impressions are splitting, and the post on striking distance keywords argues that check belongs before content work rather than after it hasn’t worked.

A word on the count itself

Four hundred issues is not four hundred problems. Audit tools count findings per URL, so one wrong line in a template becomes one issue multiplied by every page that renders it. Sort by the fix rather than by the row and the list usually collapses to a dozen or so actual changes.

It’s also worth remembering what an audit can and can’t promise. Google’s Search Essentials sets out technical requirements, spam policies and best practices — and meeting all of them still doesn’t guarantee crawling, indexing or ranking. A clean audit removes reasons for Google to skip your pages. It does not make them win.

The short version

Sort every finding by the stage it breaks. Anything stopping Google fetching a page is today’s work; anything stopping it being indexed is this week’s; everything else is a backlog you work through in whatever order suits you. Collapse the count by fixing templates rather than rows, and treat duplicate titles and competing URLs as indexing problems wearing cosmetic clothing.

If your current report is a flat list of 400 rows, running a crawl in Reachavo will at least hand it back to you grouped by what it actually breaks — which is usually enough to see that a fortnight of work is really an afternoon.