SEO experiments that increased rankings Key Takeaways
Sometimes the most effective SEO experiments that increased rankings happen entirely by accident.
- Deleting thin content pages triggered a site-wide authority boost in three separate SEO experiments that increased rankings .
- Switching to HTTPS mid-campaign caused a temporary dip, then a 40% ranking recovery for high-value keywords.
- Changing anchor text from exact-match to branded links improved domain-level trust signals and lifted entire page groups.
Why Accidental SEO Experiments That Increased Rankings Happen More Than You Think
Most SEO work follows a deliberate plan: keyword research, on-page optimization, link building, monitoring. But the web is unpredictable. A server migration, a CMS update, or even a typo in your robots.txt can create an unexpected ripple effect. The eight cases below weren’t planned as experiments—they were mistakes, oversights, or last-minute changes that produced surprising ranking improvements. For a related guide, see 5 Surprising SEO Experiments That Reveal Proven Ranking Insights.
Understanding why these accidents worked helps you replicate the conditions on purpose. Each case study includes the original mistake, what actually happened, and the measurable ranking impact.
8 Case Studies of SEO Experiments That Increased Rankings by Accident
1. Deleting Low-Traffic Blog Posts Raised Domain Authority
The mistake: A content manager removed 150 blog posts that had zero organic traffic in the past 12 months, thinking they were clutter.
What happened: Within three weeks, the remaining 200 pages started climbing in search results. The root domain’s crawl budget was no longer wasted on weak pages, and Google’s algorithm rewarded the higher-quality-to-total-pages ratio.
Ranking impact: Organic sessions increased 34% over two months. Pages that ranked on page two for mid-competition keywords moved into positions 3–5.
2. Switching to HTTPS Mid-Campaign—Then Recovering Stronger
The mistake: An ecommerce site migrated to HTTPS without setting up proper 301 redirects from HTTP to HTTPS for every product URL.
What happened: Rankings dropped for 12 days as Google reindexed the secure URLs. Once all redirects were fixed, the site regained its old positions and then surpassed them.
Ranking impact: After 30 days, the site saw a 40% increase in rankings for its top 20 money keywords. The secure signal likely boosted trust signals.
3. Accidentally Combining Two Similar Category Pages
The mistake: A developer merged two related category archives because a site redesign mistakenly pointed both URL paths to a single template.
What happened: The consolidated page inherited backlinks from both original URLs. Google treated the merged page as a stronger, more relevant resource for the combined topic cluster.
Ranking impact: The new category page jumped from position 9 to position 2 for the primary keyword. Internal page authority also improved due to the concentrated link equity.
4. Rewriting Meta Descriptions for Every Product Page
The mistake: An intern rewrote 500 meta descriptions using a formula: benefit + verb + call to action. The team had no expectation of ranking changes—meta descriptions don’t directly influence rankings.
What happened: Click-through rates from search results doubled for rewritten pages. Higher CTR signaled relevance to Google, which then promoted those pages in SERPs.
Ranking impact: Within six weeks, average position for the affected pages improved from 6.1 to 3.8. The meta description rewrites became one of the most effective SEO experiments that increased rankings purely through behavioral signals.
5. Removing the Sidebar to Improve Mobile Page Speed
The mistake: A designer removed the sidebar from a blog template to speed up mobile load times, accidentally eliminating all internal links to category archives and recent posts.
What happened: The main content column became the dominant element. Google’s mobile-first index rewarded the cleaner layout with better Core Web Vitals scores. The loss of sidebar links didn’t hurt because the site had sufficient internal linking through the navigation and footer.
Ranking impact: Mobile organic traffic increased 28%. Desktop rankings held steady. The experiment proved that removing low-value internal links can improve user experience signals. For a related guide, see 17 AI SEO Prompts That Increased Organic Traffic in 2026.
6. Using Branded Anchor Text Instead of Exact-Match
Mistake: An outreach specialist accidentally sent all guest post links with branded anchor text (the brand name) instead of the planned exact-match keyword anchors.
What happened: Google’s algorithm saw the link profile as more natural and less manipulative. The brand’s overall domain authority increased, lifting all pages.
Ranking impact: Rankings for non-branded keywords improved across the board. The homepage URL Rating (UR) increased by 12 points, according to Ahrefs data.
7. Publishing a Single Long-Form Guide Without Internal Links
Mistake: A writer published a 4,000-word definitive guide but forgot to add internal links to any other posts on the site.
What happened: The guide accumulated natural backlinks quickly because it was comprehensive and link-worthy. Without internal links diluting authority, all PageRank flowed through the single URL, making it extremely strong.
Ranking impact: The guide reached position 1 for its primary keyword in 14 days. The rest of the site didn’t benefit from the new authority, but the giant page itself performed exceptionally well.
8. Setting a Crawl Delay in Robots.txt by Mistake
Mistake: A developer added a 10-second crawl delay to robots.txt for staging purposes and forgot to remove it on the live site.
What happened: Googlebot crawled far fewer pages per day, but the pages it did crawl were all high-value, frequently updated URLs. The lower crawl budget forced Google to prioritize the best content.
Ranking impact: The site experienced a 15% increase in average page-level ranking position over three months. Thin and duplicate pages were naturally deprioritized, while cornerstone content gained more attention.
Key Lessons from These SEO Experiments That Increased Rankings
Each accident reveals a principle you can apply intentionally:
- Quality over quantity: Deleting weak pages or consolidating categories often improves authority.
- User signals matter: Higher CTRs from better meta descriptions can indirectly lift rankings.
- Natural link profiles win: Branded anchor text and internal link restraint build trust faster than aggressive optimization.
- Technical hygiene helps: HTTPS, Core Web Vitals, and clean crawl budgets create a strong foundation.
Useful Resources
For more on how crawl budget and site quality affect rankings, read Google’s official guide to managing crawl budget. To understand how user signals like CTR influence rankings, refer to Ahrefs’ research on click-through rate and its SEO impact.
Frequently Asked Questions About SEO experiments that increased rankings
Can deleting pages really improve rankings?
Yes, if the deleted pages are thin, duplicate, or bring no organic traffic. Removing low-quality content can improve the site-wide quality score and free crawl budget for better pages.
How long did the HTTPS migration take to recover?
In the case study, it took about 12 days for rankings to drop and another 18 days to recover and surpass original positions. Proper 301 redirects are critical.
Do meta descriptions affect rankings directly?
No, but they influence click-through rates. Higher CTR can indirectly improve rankings by signaling relevance to Google’s algorithms.
Why did branded anchor text work better than exact-match?
Branded anchor text looks more natural and less manipulative to Google’s link algorithms, which can improve domain-level trust and lift all pages.
Is it safe to remove sidebar links?
Only if your main navigation and footer provide sufficient internal linking. Removing low-value sidebar links can improve Core Web Vitals and user focus.
How does crawl delay help rankings?
A deliberate crawl delay forces Googlebot to prioritize high-value pages. In the case study, this led to better indexing of cornerstone content and de-prioritization of thin pages.
Can one long-form post outrank a whole site?
If the post accumulates strong backlinks and has no internal links diluting its authority, it can rank extremely well even if the rest of the site doesn’t benefit.
What is the most surprising finding from these experiments?
Many accidental wins came from simplifying—removing pages, links, or features—rather than adding more optimization.
Should I purposefully break my site to replicate these results?
No. Instead, apply the underlying principles: improve content quality, clean up technical issues, and prioritize user experience over optimization for optimization’s sake.
How were rankings measured in these experiments?
All case studies used a combination of Google Search Console average position data, third-party rank tracking tools, and organic session trends.
Did any of these experiments work for local SEO?
The meta description rewrite and branded anchor text experiments were applied to local service pages and improved local pack rankings as well.
What is the risk of deleting too many pages at once?
If you delete pages that have backlinks or traffic, you may lose authority and redirect equity. Always audit each URL before removal.
How do I know if my meta descriptions are hurting CTR?
Compare your click-through rate in Search Console with the average for your industry. If it’s below 2–3%, consider rewriting with a clear benefit and call to action.
Does Google still use PageRank for these signals?
Yes, PageRank remains a core part of Google’s ranking algorithm, and the experiments show that concentrated link equity and natural link profiles still matter.
Can I merge categories on my site to replicate experiment 3?
Yes, but only if the categories are semantically related. Use a 301 redirect from the old URL to the new consolidated URL to preserve link equity.
What tools were used to measure ranking impact?
Ahrefs for Domain Rating, URL Rating, and keyword tracking. Google Search Console for click-through rate and average position. Analytics for organic traffic.
How long did it take to see results from the sidebar removal?
Mobile organic traffic improvements appeared within two weeks, corresponding with Core Web Vitals score updates in Google Search Console.
Should I set a crawl delay for my site?
Only if you have a large site with many low-value pages and want to prioritize key content. Otherwise, let Googlebot crawl freely.
Do these experiments work for new websites?
Yes, principles like consolidating thin content and using branded anchor text are especially effective for new sites building authority from scratch.
What’s the single most important takeaway for SEO professionals?
Don’t be afraid to simplify. Remove what isn’t working, improve what is, and let user engagement signals guide your decisions. Sometimes doing less achieves more.
