Website Ranking Issue: A Practical Diagnosis for Google Visibility Loss
An evidence-led workflow for separating Google indexing, migration, technical SEO, content, authority and algorithm factors behind a website ranking issue.
· 15 min read
A B2B services page that once performed for development queries can fall to an average Google position near 54 while the same URL still averages about position 2 for closely related consulting queries. That pattern is a website ranking issue, not automatically an indexing failure, and this guide shows how to isolate the most likely cause before changing content, redirects or site architecture.
The search results for “ranking issue” are unusually ambiguous. In Jira and Atlassian, ranking means changing the order of work items on a board; Jira’s LexoRank system is part of that operational ordering. In Google Search, ranking is an information-retrieval outcome: Google decides which indexed pages best satisfy a specific query in a specific context. This article addresses the second meaning: a website that has disappeared, slipped, or stopped appearing for valuable Google searches.
Define the symptom before trying to fix it
A website ranking issue is a measurable loss of Google Search visibility for a page, query group, directory, device, country, or search type. It is not the same thing as a page being unindexed, and it is not proven by a single rank-tracker screenshot.
The source scenario illustrates why specificity matters. One established B2B page remained indexed, had the intended Google-selected canonical, redirected old URLs to current URLs, and exposed its content and internal links in the initial HTML. Yet one family of development-related terms averaged around position 54 while consulting-related terms for the same page averaged around position 2.
That evidence changes the initial hypothesis. A URL that ranks strongly for some relevant queries is unlikely to have a simple sitewide block, universal rendering failure, or total deindexing problem. The more productive question becomes: what changed in Google’s interpretation of this page for this particular query set?
Start by describing the issue in one sentence:
- “The
/development-services/page lost impressions and positions for commercial development queries after March 2026, but retained top-three visibility for consulting queries.” - “A directory lost clicks in the United States on mobile, while desktop and UK performance stayed stable.”
- “Old WordPress URLs moved to Next.js URLs in May 2026, and only pages with altered page templates declined.”
This definition makes later comparisons meaningful. It also prevents teams from treating every traffic drop as an algorithm penalty or every average-position change as a technical SEO defect.
Establish a reliable Google Search baseline
Google Search Console should be the first evidence source because it reports Google Search impressions, clicks, queries, pages, countries, devices, and search types. Google’s own traffic-drop guidance recommends comparing a broad period and looking for patterns by affected pages and queries, rather than reacting to one volatile day.
For a suspected decline between March and June 2026, compare at least these views:
- The 28 days before the decline versus the 28 days after it. This identifies the sharpest change.
- The same period year over year, if seasonality is plausible for the service or industry.
- The page filtered against the affected queries. Averages across the whole site can hide a query-specific loss.
- Affected versus unaffected query clusters. In the source example, “development” and “consulting” clusters should be exported separately.
Read impressions before treating position as the verdict
Average position is a useful directional metric, but it aggregates many impressions. A page appearing at position 2 for a small number of branded or highly specific queries can still have little visibility for a broader commercial term where it now appears around position 54.
Check whether the lost query group also lost impressions. If impressions fell sharply, Google is serving the page less often for that intent. If impressions held but clicks fell, inspect changes in title links, snippets, competitors, ads, AI features, local results, or search intent.
A manual site: query can offer a quick clue, but it is not a complete index report. Google documents that site: results are not guaranteed to be exhaustive. Use Search Console’s URL Inspection and performance data as the decision-making evidence.
For a repeatable measurement process, the workflow in Google rank tracking broken? How to verify a real ranking drop is useful before assigning a cause to a chart movement.
Separate crawling, indexing and ranking
Crawling, indexing and ranking are related but distinct stages. Googlebot must be able to access a URL; Google then decides whether and how to index it; Google’s ranking systems then decide where that indexed page may appear for individual searches.
An “indexed” result in URL Inspection answers an important but narrow question: Google has a version of the page in its index. It does not guarantee that the page will rank for a particular keyword, retain previous rankings, or be selected for every related intent.
What to verify in URL Inspection
For the affected URL, record the following rather than simply marking it “indexed”:
- The inspected URL and the Google-selected canonical.
- The last crawl date and crawl status.
- Whether the live test returns a 200 HTTP status.
- Whether robots directives, noindex tags, login walls, or blocked resources affect rendering.
- Whether Google can see the principal copy, headings, internal links and structured elements expected on the page.
Google’s technical requirements state that pages need to be discoverable and accessible, return a successful status, and contain indexable content. Google also recommends using the Page Indexing report and Crawl Stats report when diagnosing blocked or inaccessible URLs.
The source case already cleared several high-value checks: no current noindex directive, no redirect loop, main content in initial HTML, internal links present, and intended canonical selected. Those findings should not be ignored, but they should move the investigation away from “Google cannot access the page” and toward intent, migration consistency, competing URLs, and query-level relevance.
If a page is indexed but receives no impressions, investigate the distinction carefully; page indexed but no impressions: what to check after SEO fixes covers that narrower diagnostic path.
Audit the WordPress-to-Next.js migration as a system
A May 2026 migration from WordPress to Next.js combined two meaningful changes: a platform rebuild and a hostname move from www to non-www. Even where individual URLs redirect correctly, a migration can alter internal linking, template signals, page rendering, response behavior, XML sitemaps, headings, metadata, and the set of pages Google sees as near-duplicates.
Google’s site-move guidance treats hostname and URL changes as migrations that need old-to-new mapping, redirect planning, updated internal URLs and monitoring. Redirects are a strong canonicalization signal, but they work best as part of a consistent set of signals rather than in isolation.
Migration checks that catch the less obvious problems
Test a representative sample, not just the homepage and one priority page:
- Crawl legacy
wwwURLs and verify each one makes a single-hop server-side 301 redirect to its closest non-wwwequivalent. - Check that canonicals, XML sitemaps, hreflang references, Open Graph URLs and internal links consistently use the non-
wwwhostname. - Compare old and new templates for title tags, H1s, primary copy, linked service pages, breadcrumbs, schema, pagination and footer navigation.
- Review server logs or Crawl Stats around May 2026 for increased 5xx responses, timeout patterns, blocked asset requests, or sudden crawl declines.
- Confirm that important legacy URLs were not all redirected to a category or homepage rather than their exact new counterparts.
The source material says that old URLs redirect to current URLs and Google selected the desired canonical. That reduces, but does not eliminate, migration risk. A redirect may be correct while the destination has changed enough in copy, information architecture or internal authority that it no longer competes for the same development queries.
A practical migration audit should compare rendered pages and source HTML side by side. Audra can make that comparison easier by grouping redirect findings, canonical data, internal-link discovery and technical checks into one local audit rather than separating evidence across several tools.
Check technical SEO beyond the “indexed” label
Technical SEO should be tested for patterns that align with the loss. One isolated page may have a malformed canonical or thin internal linking; a whole template may have a response-header, JavaScript, structured-data, sitemap or pagination defect.
The key is to ask whether the technical issue distinguishes affected URLs from unaffected URLs. If the consulting and development queries land on the same page, a template-wide rendering defect is less likely than a page-level relevance issue. If only pages rebuilt in a Next.js template declined, the template becomes a stronger lead.
High-priority tests
Run these checks across affected pages, unaffected peers and key competitors where possible:
- HTTP status, redirect chain length and soft-404 signals.
- Canonical target, including accidental canonicalization to a broader services page.
- Meta robots headers and HTML directives.
- Robots.txt accessibility for page assets and linked resources.
- Server-rendered versus client-rendered availability of the page’s primary text.
- Internal links pointing to the canonical URL with descriptive anchor text.
- XML sitemap inclusion and last-modified patterns.
- Duplicate titles, duplicated service-page copy, or parameterized URL variants.
Google notes that blocked resources can prevent proper crawling and recommends the URL Inspection rendering view for individual checks. That matters with JavaScript frameworks, even when a spot check suggests content is visible. The useful evidence is not “the page looks fine in a browser”; it is whether Googlebot can retrieve and interpret the intended version consistently.
Diagnose content intent and query relevance
A page can be technically healthy yet cease to rank because the search results have changed or because Google now finds a different page more useful for the query. Google describes its ranking systems as page-level systems using many signals to understand how to rank individual pages. That makes the source pattern—strong consulting rankings and weak development rankings on one URL—especially relevant.
The immediate task is to compare the affected search results, not merely rewrite the page. For each lost development query, record the first 10 organic results and classify their dominant intent:
- Service landing pages seeking leads.
- Guides explaining a development process or framework.
- Job, software, or product pages.
- Local providers and map results.
- Enterprise consultancies with proof, case studies and specialist pages.
If the top results are now detailed development-service pages while the affected URL reads like broad consulting copy, Google may be interpreting it as a poor fit for the more specific commercial need. Adding the word “development” repeatedly will not solve that mismatch.
Compare the page against the result set
Review concrete elements: the H1, title, opening paragraph, service scope, buyer problems, delivery process, technologies, project examples, outcomes, pricing approach, locations, qualifications and calls to action. For B2B services, an About page can also provide corroborating experience and ownership signals; see About page SEO: turn company history into search proof.
Also check for internal competition. A migration can create several similar pages—such as /development, /software-development, /digital-development, and an older blog article—with unclear roles. Do not assume cannibalization from overlap alone. Confirm it by seeing whether another site URL gains impressions or rankings for the same queries while the target page declines.
Evaluate links, authority and site architecture
Links remain useful diagnostic evidence because a hostname migration can fragment internal and external signals even when redirects function. Google explains that canonicalization helps consolidate signals from similar URLs, including links, but canonical selection is ultimately Google’s decision rather than a guaranteed instruction.
Start internally. Identify whether the affected page lost prominent navigation links, breadcrumbs, contextual links from related services, or links from high-authority case studies after the redesign. A key development page buried three clicks deeper than before may receive weaker contextual reinforcement than it did on WordPress.
Then examine external-link continuity:
- List the most-linked legacy URLs in the affected area.
- Confirm their redirects resolve directly to equivalent new pages.
- Look for valuable links that now redirect through multiple hops or land on generic pages.
- Compare referring-page topics: development-focused editorial links may matter more to a development query set than general corporate links.
No fixed number of links guarantees recovery. The evidence to seek is a meaningful change between the pre-migration and post-migration link graph, especially where the page lost both relevance signals and internal prominence.
Treat performance and accessibility as evidence, not automatic ranking explanations
Performance and accessibility problems can impair user experience, crawling reliability and the ability of users or assistive technology to use a page. They are worth fixing, but a slow Largest Contentful Paint or missing form label is not, by itself, proof of why one keyword cluster dropped from page one to position 54.
Audit the affected page for concrete defects: render-blocking JavaScript, oversized hero media, layout shifts around the primary heading, broken keyboard navigation, inaccessible accordions that contain key service details, missing image alternatives, low-contrast calls to action, and page sections unavailable without client-side interactions.
For a Next.js migration, compare the initial HTML and rendered experience on mobile. The source case says main content and internal links were in initial HTML, which is a good sign. Still, other page components may have changed: service proof could be deferred, testimonials may load late, or key navigation may no longer expose related pages clearly.
A combined audit is useful here because a client-ready report should show which findings are likely causal, which are quality improvements, and which have no demonstrated connection to the visibility loss. Audra’s local-first audits can place technical SEO, performance, accessibility, links and answer-engine visibility beside each other without requiring a subscription-based reporting workflow.
Check Google algorithm updates and reporting anomalies last, not first
Google makes broad core updates several times per year, and its documentation explains that these updates are not designed to target particular sites or individual pages. A timing correlation can justify investigation, but it does not establish causation.
For a decline described as March through June 2026, plot the exact day or week when impressions and positions changed. Then compare it with Google Search Status Dashboard incidents, documented Google Search ranking updates, deployment dates, release notes, redirects, content rewrites and analytics-tag changes.
The diagnostic standard should be higher than “traffic dropped around an update.” Look for a pattern:
- Did many unrelated site sections decline on the same day?
- Did competitors in the same result set shift at the same time?
- Did only migrated templates or development-related pages decline?
- Was there a Search Console data anomaly or a rank-tracker collection issue?
Google’s traffic-drop documentation explicitly includes algorithmic updates, technical issues, security or spam issues, seasonality, changing interests and reporting glitches among possible causes. That list is a reminder to test competing explanations before declaring a penalty.
If no technical fault appears, use Google’s core-update guidance as a content-quality review framework. Compare the page honestly with the results now winning the query rather than making reactive, repeated edits every few days. Google does not publish a guaranteed recovery timeline, so reports should state that recovery timing varies and should be measured against a documented change log.
Run a compact website ranking issue workflow
A disciplined workflow prevents teams from making ten simultaneous changes and losing the ability to learn what mattered. For the source scenario, the first review should prioritize query segmentation and migration comparison over another generalized content refresh.
- Document the loss. Export Search Console queries, pages, countries, devices and dates; label affected and unaffected clusters.
- Verify indexing. Inspect representative affected URLs, confirm Google-selected canonicals, status codes, robots controls and rendered content.
- Compare before and after migration. Match legacy and new URLs, internal-link counts, template copy, sitemaps, redirect hops and crawl behavior.
- Inspect the live SERP. Classify search intent and identify which page types replaced the affected page.
- Audit competing site URLs. Check whether another page took over impressions for the same query group.
- Prioritize one evidence-backed fix. For example, restore a missing service section and contextual links, correct a template canonical, or split an overbroad consulting page into distinct intent-led service pages.
- Annotate and monitor. Record the release date and reassess impressions, rankings and conversions over subsequent crawls and reporting periods.
For agencies, the deliverable should separate facts from hypotheses. A strong client report can say: “Google indexes the page and selected the intended canonical; the decline is limited to development-intent queries; migration-era internal links fell from key service pages; the recommended action is to restore topical service architecture and monitor query-level impressions.” That is more useful than an unsupported claim that Google “penalized” the website.
FAQ
Why isn't my website ranking on Google?
A page may not rank because Google cannot crawl or index it, but it may also be indexed and simply not considered a strong result for the query. Check Search Console first: inspect the URL, confirm the Google-selected canonical, review affected queries and impressions, and compare the page with the current top results’ intent and format.
Why has my website dropped in rankings?
Ranking losses can follow a migration, redirects, server errors, lost internal links, changed search intent, stronger competitors, seasonality, security or spam issues, or a Google algorithm update. The useful diagnosis starts with the exact affected pages and query clusters. A date correlation alone does not prove an algorithm update caused the decline.
What does ranking mean in SEO?
In SEO, ranking means the position or visibility a page earns in Google Search for a particular query. It differs from Jira ranking, where teams order work items by priority using systems such as LexoRank. Google ranking is an information-retrieval decision based on the page, query, search context and competing results.
How do I diagnose and fix a website ranking issue?
Define the loss, export Search Console data, verify crawling and indexing, review migration signals, compare the affected page with current search results, then fix the highest-confidence issue. Avoid changing redirects, titles, copy and internal links all at once. Keep a dated change log so the team can evaluate what actually improved visibility.
Why isn't ranking #1 on Google enough anymore?
A high position for one query does not prove broad search visibility. The same page can rank around position 2 for consulting terms and around position 54 for development terms because Google interprets those searches differently. Measure query clusters, impressions, clicks, landing-page conversions and search-result features rather than celebrating one headline ranking.
Sources
- https://www.reddit.com/r/bigseo/comments/1wy6ib2/ranking_issue/
- https://developers.google.com/search/docs/monitor-debug/debugging-search-traffic-drops
- https://developers.google.com/search/docs/crawling-indexing/site-move-with-url-changes
- https://developers.google.com/search/docs/essentials/technical
- https://developers.google.com/search/docs/appearance/ranking-systems-guide
- https://developers.google.com/search/docs/appearance/core-updates
- https://developers.google.com/search/docs/crawling-indexing/consolidate-duplicate-urls