An SEO audit is not an 80-page report that nobody reads. It is an honest conversation with your website: what is working, what is dragging problems along and, above all, what you can fix this week with the resources you have. If you are starting out in this, the temptation is to open a tool, launch the crawl and drown in a list of 3,000 issues. That path leads to paralysis.
Here you will find a different approach: separating what you control within your site (on page) from what depends on third parties (off page), prioritizing by real impact and avoiding the classic mistakes of first-time auditors. According to official Google documentation, no single signal guarantees rankings, so we steer clear of promises and focus on reducing friction.
Before touching anything: define what questions your audit should answer
An audit without prior questions is a collection of data with no direction. Before opening Search Console or any crawler, write three or four specific questions on a sheet of paper. Real examples:
- Why do category pages get impressions but almost no clicks?
- Which pages have lost traffic in the last 90 days and share a technical pattern?
- Do inbound links point to URLs that redirect or return 404?
This initial step makes the difference between a useful analysis and a PDF that stays in the drawer. In small projects, with two or three well-chosen questions you already have material for weeks of work.
On-page audit: what you can fix without asking anyone’s permission
On page covers everything that lives within your domain: content, structure, tags, internal linking and basic technical health. It is the part where a beginner can generate the most value quickly, because it does not depend on external negotiations.
1. Crawling and indexing: the filter almost nobody checks first
Before looking at titles or keywords, confirm that Google can access what you want to rank. In Search Console, review the “Pages” report and distinguish between:
- Excluded because they are not relevant: normal in filters, parameters or thank-you pages.
- Excluded due to an error: this is where the gold is. An accidental noindex on a main category or a robots.txt that blocks CSS can explain entire traffic drops.
A common case that is only detected with experience: the development team deploys a new version of the site and, without meaning to, adds an X-Robots-Tag: noindex header in the production environment. The site keeps working, users notice nothing and organic traffic plummets silently for days. Checking the HTTP headers of a sample of URLs after every deployment is a cheap habit that prevents nasty surprises.
2. Titles, metas and heading hierarchy
You do not need a perfect title for every page. You need every title to describe the page accurately and not be repeated across twenty different URLs. Crawling tools flag duplicates, but the judgment is yours: two pages with the same title can be legitimate if they serve different intents and are well differentiated in their content.
Something similar happens with Hn headings. One H1 per page is a solid recommendation, not a law of physics. What does penalize the experience is using H2 and H3 as decoration, unrelated to the content they head.
3. Content: intent before keyword density
The right question is not “how many times does my keyword appear?” but “does this page answer what the user expects when searching for this?”. According to industry studies, most content that loses rankings fails not because of over-optimization, but because search intent changed or because the content remained superficial compared to more complete competitors.
A practical trick: compare your page with the top three in the SERP for that query. If all three include price comparisons and yours only explains the concept, you are probably outside the intent even if you have the perfect title.
4. Internal linking: the silent multiplier
Review which pages receive the most internal links and which are orphaned. A frequent mistake in blogs with years of content is that new articles only link to the home page and the contact page. Important URLs end up three or four clicks deep and receive few internal signals. A pass of contextual linking toward commercial pages usually moves the needle more than rewriting titles.
Off-page audit: what depends on others and how to evaluate it without obsessing
Off page includes inbound links, brand mentions, reviews and external signals. Here transparency is key: you do not control what others publish, and no audit can guarantee that a media outlet will link to your site. What you can do is diagnose the current situation and detect risks.
1. Link profile: quantity is not quality
Export inbound links from Search Console and from a third-party tool. Do not stop at the total number. Look at:
- Unique referring domains versus total links: a thousand links from a single domain are worth less than twenty from different, relevant outlets.
- Link anchor text: an excessive concentration of exact-match anchors is an unnatural signal that should be diversified over time.
- Destination pages: if 90% of your links point to the home page, your internal pages compete without external support.
2. Broken links and redirects at the destination
A recurring mistake: media outlets link to your site, you change the URL structure and those links end up in a 404 or in a redirect chain. Authority is diluted. A quarterly review of inbound links to old URLs and their redirection to the correct destination is a basic maintenance task that many teams ignore.
3. Brand mentions without a link
Not all mentions turn into links, and chasing every one can consume more time than it contributes. Prioritize mentions in outlets with real traffic and a related topic. A mention in a general-interest outlet without a link can have brand value, but it rarely moves rankings on its own.
4. Reviews and reputation
If you have a local business, reviews are part of off page and affect visibility on maps. According to official Google documentation, reviews influence local ranking, but they are not the only factor. Responding to reviews and keeping contact details consistent matters as much as getting new ones.
How to prioritize when everything seems urgent
A complete audit can generate hundreds of tasks. The filter that works is this:
- Does it affect indexing or crawling? If so, it goes first. Without indexing there is nothing else to discuss.
- Does it affect pages with traffic or conversions? An error on a URL that receives 10 visits a month is noise; the same error on the main product page is critical.
- Can you solve it this week? If it depends on a third party or on a two-month development, note it down and move on.
This order avoids the classic beginner’s mistake: rewriting the metas of twenty irrelevant pages while the main category is blocked by a noindex.
What an SEO audit does NOT solve
It is worth saying clearly: an audit does not fix a product nobody searches for, does not compensate for a slow site due to architecture decisions that require investment and does not replace a content strategy. Nor does it guarantee higher rankings. It serves to eliminate friction and make decisions with data, not to predict results.
In projects with a limited budget, a well-scoped on-page audit usually delivers returns before a link campaign, because technical and content fixes depend only on your team. Off page gains weight once you already have a solid foundation and need external signals to compete for difficult queries.
Tools: fewer than you think
To start with, Search Console covers indexing, performance and links. A free or low-cost crawler gives you duplicate titles, redirects and 404 errors. A spreadsheet where you note findings, priority and owner is more valuable than the most expensive tool. The discipline of documenting what you changed and when will let you measure the real effect, something no suite does for you.
Auditing is not about accumulating issues. It is about choosing the ones that matter and checking, over time, whether the effort translated into less friction for the user and more clarity for search engines.