SEO Analysis:
How to Find and Fix Issues

SEO analysis is a review of a website’s technical accessibility, content, keywords, links, and organic search performance. Its purpose is to identify what prevents relevant people from finding and using the site, then choose specific improvements. A useful analysis produces a prioritized list of pages and fixes, with evidence for each recommendation and a way to measure the result.

Google’s SEO Starter Guide describes SEO as helping search engines understand content and helping people discover a site and decide whether to visit. An SEO analysis turns that goal into checks: whether important pages are accessible, whether they answer relevant searches, how they appear in results, and what visitors do after arriving.

Gather the data for an SEO analysis#

For a review focused on Google Search, start with the following tools. Keep the evidence from each tool attached to the question it answers.

An optional third-party tool can supplement the review, but label its estimates and recommendations separately. According to Google’s guidance on third-party SEO tools, these tools do not have access to Google’s internal ranking data, and their predictions cannot guarantee performance.

Google’s guide to combining Search Console and Analytics distinguishes activity in Google Search from recorded activity on the website and explains that clicks and sessions are calculated differently. Its comparison of the reports also identifies tag implementation, consent choices, time zones, and URL reporting as reasons for discrepancies. Compare their trends without requiring their totals to match.

Before examining individual pages, record the scope of the review:

• The website property and page groups being analyzed.
• The search engine, search type, country, and device filters.
• The current and comparison periods.
• The action the site needs visitors to complete.
• Recent changes to URLs, templates, content, or tracking.

Use equal-length, completed periods for a recent comparison, and keep the filters consistent. Also examine the same season in the previous year where data is available. Google’s traffic-drop investigation guide recommends examining longer history for seasonal patterns and comparing affected pages and queries. Choose a period long enough to contain meaningful activity for the pages under review.

1. Check whether important pages can be crawled and indexed#

Begin with the pages intended to attract search visitors. Google’s technical requirements specify that Googlebot must not be blocked, the page must return an HTTP 200 response, and its content must be indexable, although meeting these conditions does not guarantee indexing.

Inspect important URLs for access errors, unintended crawling restrictions, and unwanted noindex instructions. Google’s technical guidance distinguishes the controls: robots.txt prevents crawling but does not ensure a URL stays out of results, while noindex requires crawl access to prevent indexing.

Search Console’s Page Indexing report groups non-indexed URLs by reason and explains that duplicate or deliberately excluded URLs need not be indexed. Focus on whether the preferred versions of important pages are indexed.

Use URL Inspection to review a specific URL’s indexed status, crawl information, rendered content, and Google-selected canonical. The tool’s indexed result and live test answer different questions: the stored result describes Google’s indexed version, while the live test checks the current page without guaranteeing indexing or predicting the selected canonical.

Record a technical finding only after checking the exact URL and its intended role. Attach the reported reason, the affected pages, and the repair needed. Keep deliberate exclusions out of the repair queue.

2. Analyze keywords, search intent, and page content#

Use Search Console’s query and page analysis workflow: select a page and examine its Queries tab, or filter an important query and review the Pages tab. Record which URLs receive visibility for each priority search term. Sort the comparison by the change in clicks to locate affected queries or pages.

Google’s data grouping documentation explains two limits: some queries are withheld for privacy, and the report does not display every data row. Most page performance is assigned to canonical URLs. Use the visible queries to identify patterns; do not treat their sum as a complete count of demand or expect every reported URL to match an Analytics landing URL.

For each priority page, write down the main search need in plain language. Then review:

• Whether the page answers that need directly.
• Whether essential steps, specifications, explanations, or evidence are missing.
• Whether factual information remains accurate and current.
• Whether the headings make the answer easy to locate.
• Whether the page provides useful information beyond what other results offer.

These checks follow Google’s content assessment guidance, which asks about completeness, originality, reliable sourcing, factual accuracy, and value compared with other search results. Google also states that it has no preferred word count. Set the scope from the question the page needs to answer.

Compare competing search results#

Search the priority query and inspect the relevant competing pages. Record the search date, location, and device context. Because results vary with time, location, device, and search history, treat a manual search as a contextual observation.

Compare the page format, the answer provided, the evidence used, and the information needed to complete the search task. Record each gap as a concrete editorial task: an unanswered question, an outdated fact, an unclear explanation, or missing evidence. Keep the comparison focused on what would improve the answer.

When several URLs address a similar query, inspect their actual purpose before recommending consolidation. Use the query and page views to identify the overlap, then determine whether the pages serve distinct needs or repeat the same material. Preserve separate pages when their purpose is distinct; investigate duplication when it is not.

3. Review titles and search snippets#

Compare the page’s title, main heading, description, and visible search result with the queries it serves. Google’s title-link guidance recommends descriptive, concise, distinct titles without keyword stuffing or boilerplate, and explains that displayed titles can come from several sources, including the title element, headings, and link text.

Check whether the title accurately describes the answer and distinguishes the page from other URLs on the site. Review a low click-through rate alongside the query, position, country, and device before choosing a title change.

Google’s snippet documentation explains that snippets primarily come from page content, sometimes use the meta description, and can vary by search. Improve the relevant page text and write an accurate description rather than assuming a description edit controls every displayed snippet.

4. Review internal links and backlinks#

Use Search Console’s Links report to identify linked pages, linking sites, and link text. Google defines a backlink as a link from another site to a page on the site being analyzed, and notes that the report provides a sample that can include links that have since disappeared.

For backlinks, inspect the referring page and destination before recording a finding. Check whether the link still exists, whether its context describes the destination accurately, and whether the destination remains available. Use verified references to understand which content attracts links and which destinations need attention.

Google’s link best practices explain that links support page discovery and relevance, and recommend crawlable links with descriptive, relevant anchor text. Apply its internal linking guidance by ensuring that each important page has a link from another page on the site.

Review navigation and contextual links to priority pages. Check that their destinations work and that the link text describes what the reader will find. Add a link where the destination supplies a useful next step or supporting explanation. Record the source page, destination, and intended change so the recommendation can be implemented precisely.

5. Check page performance and mobile usability#

Run PageSpeed Insights on important landing pages and on pages representing different site templates. Its documentation distinguishes field data from real visitors and Lighthouse lab diagnostics from a simulated test. The field data covers a trailing 28-day period and may fall back to origin-level data, or be unavailable, when samples are insufficient.

Check the reported scope before attributing a result to one page. Review the Core Web Vitals classifications for LCP, INP, and CLS, then examine the diagnostics for actionable repairs. Record which URLs or templates are affected and which element or behavior needs investigation.

Also open the pages on a mobile device. Check navigation, text readability, overlays, forms, and the visibility of the main answer. After a repair, rerun the relevant test and confirm the affected interaction works. Assess the subsequent search performance separately in Search Console.

6. Connect visibility to organic visits and useful actions#

Keep the measurements distinct. Search Console’s Performance report defines the following search metrics:

Google’s performance analysis guidance recommends focusing on trends in impressions and clicks over position alone. Review a specific query and page using consistent filters.

In GA4, open the Landing page report and add Session source / medium. Filter to google / organic for a Google-focused analysis, following Google’s combined-data guide. The Landing page documentation describes how this secondary dimension connects entry pages with traffic sources. Use the appropriate broader organic-search segment only when the review covers other search engines too.

Choose a meaningful action and verify that it is collected consistently. Examine sessions, the relevant key events, and revenue where applicable. Google’s Traffic acquisition report documentation defines session key event rate as sessions containing a key event divided by total sessions. The report’s key event count measures recorded occurrences; dividing that count by sessions does not calculate the share of sessions completing the action.

For a conversion review, record the selected event and the rate’s denominator. Evaluate each page against its purpose, and inspect the route to the intended action. Check the measurement setup before interpreting a sudden decline in recorded actions as a content problem.

Prioritize the fixes the evidence supports#

Use the findings to choose a first investigation or repair. The table below is a triage method: each pattern directs the next check, rather than proving a cause.

To check the demand explanation, use Google Trends with the relevant topic, geography, and time range. Its data guide explains that values are normalized measures of relative interest, not absolute search counts. A decline in impressions alone cannot establish that fewer people searched for the topic.

Order the work by the importance of the affected pages, the size and persistence of the problem, the strength of the evidence, and the effort required. Put confirmed access problems on essential pages ahead of speculative edits. Prioritize a content change when the missing answer is identifiable and relevant to the searches the page serves.

For each selected fix, retain a short record:

• The affected URLs and queries.
• The finding, evidence, date range, and filters.
• The proposed change and its owner.
• The metric or technical check that will show whether the problem improved.
• The change date and review date.

Recheck after changes#

Confirm technical repairs with the relevant inspection or performance test. Review content and search-result changes against the same queries, pages, and filters used in the baseline. Retain seasonal context and document other changes that could affect the comparison.

Allow time for search effects to appear. Google’s SEO Starter Guide says some changes may be reflected within hours while others take months, and it generally recommends waiting a few weeks to assess their effect. Keep technical verification separate from that later performance review, and record an inconclusive result when the data does not support a decision.