Case Study Revealed the Hidden Weaknesses Behind a Used Cars Website

Car dealership SEO succeeds when your pages help buyers make confident decisions at every stage of the process. Google evaluates whether you offer trustworthy guidance, model insights, pricing transparency, and clear local signals. The more your content reduces uncertainty for the shopper, the stronger the behavioural metrics become — and Google uses those metrics to decide which dealerships deserve top placement in map listings and organic search.

Google 1st Page Ranking for a Pre-owned & Used Cars Website

Markov chains reveal how rankings transition through different states over time. This gives you a measurable way of understanding volatility and long-term ranking potential. I show how these models work using real examples from competitive sectors such as used cars, where ranking movement is often misunderstood. When you can see the probability of upward transitions, you can target actions that accelerate progress instead of relying on guesswork.

For another example of how state transitions behave, see my financial services case study.

With the birth of AI most SEO agencies offer to optimise your website for the AI-driven search landscape, ensuring you’re discoverable through ChatGPT and other AI assistants. This is not the only criteria and is the least important use of AI.
For years, search engines have used Markov Chains and Graph Theory to map how a website works—not just what keywords it contains. Markov Chains are based on probabilities. You don’t need to be a computer scientist any longer now you are able to use AI to calculate all the maths behind what Google uses for their PageRank algorithms. For example, if users are on a website’s homepage, there might be a 40% chance they’ll go to the “About Us” page, a 30% chance they’ll go to the “Products” page, and a 30% chance they’ll go to “Contact Us.” We wanted a clear, numbers-first picture of how Google’s crawler (and any visitor) moves through the site. These probabilities are calculated from historical data, such as how people have navigated the site. Our various package options all contain this analysis.

Today’s Google search isn’t driven by a single “algorithm”. It’s a collection of AI systems working together — language models, relevance scorers, link evaluators, spam filters and behavioural analysers. An update happens when Google adjusts how these systems communicate and influence one another. Even a small modification in one model can ripple across the entire ranking landscape, producing sudden changes in visibility for thousands of sites at once.

Google interprets a website as a connected graph of pages

where internal links determine how authority flows. This case study shows how a website ranking on Page One for “used cars” displayed structural weaknesses only detectable through Markov chain modelling and graph theory.

By analysing the internal linking structure mathematically, we uncovered why impressions were dropping, why vehicle pages took longer to index, and why certain make/model queries were losing ground — despite no changes in the website’s content or CMS. The mathematics of this model is, if we represent the Web as a graph, and a web surfer goes through this graph randomly, we will get a mathematical abstraction called a Markov chain. The interesting property of a Markov chain is, if this random surfer will go from page to page for a long enough time, the probability they end up on a certain page will be constant.

The simplified top-5 network graph highlights the five pages on Elephant.co.uk that currently act as the strongest internal “hubs,” based purely on how many links they send and receive within the site. These are the pages with the highest structural influence in the current architecture—not necessarily the most important commercial pages, but the ones the internal link network naturally pushes authority toward. The results were revealing in this case, the top nodes skew heavily toward

• Help & Support,
• Claims, and
• policy upgrade pages,
• Car insurance policy
• as well as the Sitemap

reflecting the imbalance uncovered in the Markov and PageRank analysis. This means the graph is showing the current internal linking reality, not the ideal structure. By applying the recommendations from the Markov analysis—such as boosting internal links to core car-insurance product pages, reducing over-linking to support content, building a central Car Insurance hub, and improving breadcrumb hierarchy—you can shift these top authority nodes away from support pages and into key commercial URLs. In other words, the graph visualises where authority flows today, and the suggested changes show how to reshape that network so the top-5 nodes become the pages that drive revenue and rankings.

The Background
The dealership’s website ranked well for high-value phrases such as “used cars”, “used cars Southampton”, and several make/model keywords. The site had strong photography, fast performance, and clean URLs. However, key signals suggested a structural problem:

• New stock was taking longer to index
• Older listings were drifting down the SERPs
• Page One rankings were unstable
• Impressions were falling in Search Console

Nothing had changed in content or technical setup — yet performance was shifting. This pointed to an internal linking imbalance.

Key Findings:

1. Help & Support dominates internal authority.
   The /help-and-support/ section receives significantly more internal link weight than any other area of the site. This suggests that many pages either link directly to support resources or rely on templates that push authority toward non-commercial pages. For a car insurance website, this is a misalignment between authority flow and revenue goals.

2. Commercial car insurance pages are underweighted.
   Core money pages — such as the main car insurance quote page — do not appear in the top authority nodes. This weakens their ability to rank competitively for high-value search terms like car insurance, car insurance quotes, or multi-car insurance.

3. Upgrade and ancillary product pages receive disproportionate weight.
   Pages for breakdown cover, legal protection, hire car cover, and personal injury cover appear highly ranked. This pattern likely comes from repeated sidebar or menu links during the quote process. While helpful for users, they absorb internal PageRank that would be better distributed to broader commercial pages.

4. Claims pages also emerge as strong nodes.
   The /claims/ page ranks high due to frequent links across help content and policy documents. This is expected, but again pushes authority away from quote and purchase pages.

 

What the Markov model shows when your site isn’t ranking yet

A Markov chain doesn’t judge quality — it only predicts how likely a page will move between ranking “states” over time, based on how similar pages behave.

Meanwhile:

• The Used Cars page was underweighted
• Vehicle detail pages were dead ends
• No make/model clusters existed
• Authority was leaking into low-value utility pages
• The site lacked internal cycles and hubs

Google prefers organised clusters, hierarchical hubs, and linking cycles that connect inventory pages. This website had none of those features.

What We Recommend Should Be Changed

The Markov and graph-theory analysis revealed clear strategic improvements.

1. Strengthen Homepage → Used Cars Linking

• Add a primary CTA to /used-cars
• Add featured stock on the homepage
• Add “Browse by Make” links

2. Create Make and Model Category Pages

• /used-bmw/
• /used-audi/
• /used-ford/
• /used-mercedes/

Plus model pages such as:

• /used-audi-a3/
• /used-bmw-3-series/

3. Add Related Vehicles Modules to Car Detail Pages

• 3–6 similar cars
• Links to Make and Model pages
• Link back to Used Cars

4. Add Breadcrumb Navigation

Home > Used Cars > BMW > 3 Series > Vehicle

5. Reduce Authority Leakage

• Limit Contact/Finance links to footer only
• Remove duplicated navigation links
• Fix Cloudflare-generated email-obfuscation URLs

6. Consolidate Canonicals and Pagination

Ensure pagination points to /used-cars to avoid fragmenting PageRank.

Projected Before & After (Markov Simulation)

The following table shows the expected gains after implementing the recommended structure.

Page	Before	After (Projected)	Change
Used Cars	3.87%	10.8%	+179%
Make Pages	~1%	4.2%	+300%
Model Pages	n/a	2.1%	New
Vehicle Pages	0.2–0.4%	0.8–1.4%	+250–400%
Contact Page	6.4%	2.1%	-67%

Expected SEO Impact

• New stock indexed faster
• Stronger rankings for make/model keywords
• Increased long-tail traffic (year, engine, trim)
• Greater visibility across all vehicle listings
• Better crawl efficiency and authority flow
• Improved Page One stability

Summary

This case study demonstrates how this used cars website can appear technically sound yet structurally weak. By analysing authority flow using Markov chains and visualising the site as a graph, hidden weaknesses become obvious — and fixable.

When the internal linking structure is redesigned around hubs, clusters, and cycles, Google understands the site more clearly. Rankings stabilise. Vehicle pages gain visibility. And the site becomes faster, stronger, and more competitive in search. The future of website management and SEO is interconnected — intelligent websites will require intelligent optimisation. Automation, AI-driven personalisation, and evolving search algorithms will reshape how businesses connect with their audiences.

Those who adopt these changes early will lead the next wave of digital transformation. In a world where online presence determines success, mastering these elements is not just an advantage … it’s a necessity.

Understanding the internal linking structure for this pre-owned Cars website

For an SEO analyst, understanding the internal linking structure is one of the most fundamental keys to achieving better rankings for this Used Cars example. It directly influences three core pillars of successful Search Engine Optimization (SEO): Authority Flow, Crawlability, and Relevance.

1. Authority Flow (Distributing Link Equity)

The most critical function of internal link analysis is mapping the distribution of Link Equity (often called “Link Juice” or, in academic terms, PageRank).

• The Concept: Search engines view links as “votes” of confidence and authority. An internal link passes a portion of a page’s authority to the destination page.

• The Analysis: A Markov Chain analysis, like the one we just performed, tells the analyst exactly which pages are receiving the most “votes” internally.

• The SEO Value: The analyst for this used cars website uses this data to answer a critical question: Are our most commercially important pages receiving the most internal authority? If the homepage is linking to a low-priority blog post instead of a high-converting product page, the internal link equity is being misspent. The analyst will use this insight to restructure links and funnel maximum authority to pages with high business value (products, services, lead forms).

---

2. Crawlability and Indexing

The internal link structure acts as a roadmap for search engine spiders (like Googlebot).

• The Concept: A crawler starts at the homepage and follows internal links to discover the rest of the site. If a page has few or no links pointing to it, the crawler may struggle to find and index it. These pages are known as orphaned pages.

• The Analysis: Analyzing the “From” and “To” data helps the analyst identify:

  • Orphaned Pages: Pages with an in-degree of zero, which may not be indexed and therefore cannot rank.

  • Deep Pages: Pages that are many clicks away from the homepage, which crawlers may visit less frequently.

• The SEO Value: By identifying these navigational barriers, the analyst can implement strategic linking to ensure all important content is easily accessible (ideally within 2-3 clicks of the homepage) and frequently re-crawled, ensuring timely ranking updates.

3. Contextual Relevance and User Experience

Internal linking is essential for telling search engines (and users) what a page is about and how it relates to other content.

• The Concept: The Anchor Text (the clickable text) of an internal link reinforces the topic of the destination page. When a cluster of pages about “car insurance policy upgrades” all link to the main “claims” page with the anchor text “File a Claim,” it strongly reinforces the claims page’s relevance for that topic.

• The Analysis: The analyst examines the topical clusters formed by the links. For example, all pages about “breakdown cover” should link to each other and up to the main “car insurance” hub.

• The SEO Value: This helps the analyst:

  • Build Topical Authority: Create authoritative content hubs by ensuring related pages link to a central, high-value resource.

  • Improve User Experience (UX): Provide relevant next-step links, increasing dwell time and lowering the bounce rate, both of which are positive ranking signals.

In short, analyzing the internal structure allows the SEO analyst to stop guessing and start engineering the flow of authority directly to the pages that drive revenue, thereby ensuring the right pages rank higher on the SERPs.

Related Reading:

• Travel & Tourism Case Study

• Financial Services Case Study

• SEO Case Study

• Car Insurance Case Study