Is translating articles dangerous to my SEO?

SEO experiment: 15 theories put to the test

Since Google avoids revealing the ranking factors that influence the search engine and a lot of the information is often concealed - or let's say it is named as a blanket - insights in the SEO scene often had to be gained using the try and error principle . James Reynolds presented 15 of these SEO experiments for SEOSherpa, which have some surprises in store.

SEO experiment 1 - CTR influences the ranking

Search queries and CTR experiment - Part 1:

Many SEOs believe that the CTR affects organic ranking. And is that also true?

Rand Fischkin started a very interesting experiment on this in 2014. He asked his 164 thousand Twitter followers to search for IMEC on Google and click on it. Using Google Analytics, he was able to understand that 175 - 250 users responded to this call. After 3 hours, the IMEC page went from 7th to first place - clear evidence that the CTR has a direct influence on the ranking.

Further tests were carried out to analyze the processes. With these, the result was not so striking, which is why Rand feared that Google had reduced the influence of the CTR on the ranking after his call. Which is why he made another attempt.

Search queries and CTR experiment - Part 2:

In another tweet, he asked his followers to search for "the buzzy pain distraction" and click on the result from They should use a form to record the position in which the result was.

375 users took part in his experiment and 2.5 hours later the page moved from 10th to 1st.

This experiment also confirmed that the CTR influences the ranking, but how do you use this knowledge to your advantage?

  1. Improve the CTR, by writing your title and meta description in such a way that the user feels compelled to click on your page.
  2. Build your brand, because if the user knows your brand and trusts it, your brand will be noticed more quickly in the SERPs.
  3. Make sure you stay longer, because if the user immediately turns back shortly after they have reached your site, every advantage of a better CTR will have been taken care of.

SEO Experiment 2 - The Fallacy About Mobilegeddon

When Google rolled out the Mobile-Friendly Update, or Mobilegeddon, as it was reverently called in the SEO scene, on April 21, 2015, many SEOs expected this to be one of the most important updates of recent times and that everyone who does not participate here will suffer a severe drop in ranking with their site.

After Mobilegeddon there was general astonishment, because the press in particular reported that there were no major incidents due to the mobile-friendly update.

But what really happened, Eric Enge and the Stone Temple Consulting team were able to demonstrate with their experiment.

They collected the data for 15,235 searches for April 17th. On May 18th they repeated this process and a clear before and after picture could be drawn.

As you can see on the graph, almost 50 percent of all sites that were not mobile friendly have actually lost in the rankings. The others, however, have gained in ranking.

SEO Experiment 3 - How backlinks perform after they're removed.

The assumption that links that have already been removed - link echoes or link ghosts - still have an impact on the ranking in retrospect, was tested by the MOZ team.

For this purpose, Rand Fishkin's team took on two websites that were far behind in the rankings. 22 links were placed on different websites for both sides, which significantly improved the ranking. The two sides landed at number 1 and number 5 in the SERPs.

In the next step, these links were removed again, which suggests that the pages would fall again in the rankings. But miraculously, the opposite was the case, as they stayed in their new position for several months.

This shows us that good links are worth their weight in gold and that the influence of links remains valid even after they have passed away.

SEO Experiment 4 - You can rank with duplicate content, even with competing keywords

Dan Petrovic from Dejan SEO tested four websites with his SERPs Hijack Experiment by placing their content on a different website with a higher PageRank.

Amazingly, he was able to show that the page with a higher PageRrank has displaced the original page in the rankings. One of the tested sites was even completely removed from the SERPs.

To what extent the experiment is still valid today, however, remains uncertain. Incidentally, after the test, Dan received a warning from Google. So content theft is not worth it.

SEO Experiment 5 - Number 1 is not the best place

According to a study by Stone Temple Cosulting, rich answers have been gaining ground for several years. The immediate answers to a search query that can be found in the SERPs, which are still above the first place, can ensure a steep increase in traffic, as can be seen in the example of Confluent Forms.

In order to be listed in the SERPs yourself using a rich answer, you should think of a simple and up-to-date question. Your corresponding answer should be clear, simple and useful for search engines and users. Also like to go into detail and ensure enough added value.

SEO Experiment 6 - HTTPS can hurt your ranking

Websites that use HTTPS are more secure. According to Google, HTTPS should also be a ranking factor. Therefore, they should actually get a ranking boost.

But that's not the case. At least that's what a study by Stone Tempe Consulting and seoClarity shows. In this HTTPs study, in which 218,000 domains were observed, just 630 made the switch to HTTPs. But instead of receiving a ranking boost, these pages suffered a short-term slump.

HTTPS might make your site secure, but it doesn't do your ranking any better.

SEO Experiment 7 - NoIndex doesn't always work

To remove a page from the Google index, use the robots tag with the “NoIdex” command. This removes the relevant page from the index.

And to test whether this always works correctly, IMEC Labs carried out an experiment.

You can see that not all pages have been removed from the index.

Google did not immediately remove any of the pages from the index, although the files from the Robots.txt were crawled several times a day. What is also irritating is that it is actually not crawled at all. And to make the confusion complete, the pages that were not removed from the index were crawled up to five times.

The experiment is weak in terms of conclusiveness, but at least one can say that Google's indexing process is not always the fastest and sometimes even does not do what it is supposed to. Robots.txt NoIndex should therefore be used with caution.

SEO Experiment 8 - Exact Match Anchor Text Links vs. Generic Anchor Links

To determine the impact on the ranking of Exact Match Anchor Text Links Compared to testing generic links, Rand Fishkin has made a number of attempts. As part of these experiments, he placed 20 generic links (B) on different pages that refer to a URL. He also has it at 20 Exact Match Anchor Text left (A) done.

The result is clear: Exact Match Anchor Text Links are significantly better than generic links.

SEO Experiment 9 - Links to other websites to increase your own ranking

Up to now, the SEO scene has partly held on to the belief that links on your site that refer to other sites reduce your ranking. Shay Aharony and the SEO agency Reboot wanted to know exactly and carried out their outgoing links experiment.

For this purpose, ten websites with the same structure and a similar domain format were optimized for the invented word "phylandocic". The same three links to trusted sites were placed on five of them.

After searching for the word "phylandocic", the situation was clear.

Each website that had links displaced the other pages without links in the rankings.

SEO experiment 10 - nofollow links improve your ranking

The starting question for the following experiment was whether nofollow links have an effect on the ranking.

Since nofollow links prevent link juice from being passed on, the question is justified whether they are relevant for your own ranking at all.

IMEC Lab was able to clarify this by means of a series of experiments in which several participants were asked to place nofollow links and ultimately came to the conclusion that nofollow links also have a value for their own ranking.

SEO Experiment 11 - Links from pages with many links work

Links that came from pages that already linked to many other websites were sometimes considered to be weak links. So the more outbound links a page has, the less link juice is delivered to your page.

This is exactly what Dan Petrovic tested in his PageRank Split Experiment and was able to prove with the help of an A / B test that such links from pages that already have many links can be very valuable.

SEO experiment 12 - image links are particularly helpful

In another experiment, Dan Petrovic looked at different types of links and wanted to find out which type of link had the best effect on the ranking.

In the experiment, four types of links were compared:

  1. Exact keyword text in the anchor link
  2. Exact keyword text immediately followed by the target URL
  3. Images link with the keyword text in the alt tag
  4. Keyword text followed by the link plus the CTA "click here"

The result showed that the picture link variant worked best. The link with the keywords in the anchor text also performed well. The other two variants did not appear in the SERPs.

So in the future you should think about the best way to get backlinks to your images, because they are tough.

SEO Experiment 13 - Press Release Links Work

In order to get to the bottom of Matt Cutts 'statement that links in press releases have no SEO value, SEO Consult published a press release in which there was a link to Matt Cutts' blog. The word "Sreppleasers" was used in the anchor text of the link. which is an anagram of the press release word for press release.

If you are looking for "Sreppleasers", you come across Matt Cutt's blog and since this term is not on his blog, you know that the press release has had its effect here. Apparently Matt Cutts should reconsider what he said.

SEO Experiment 14 - The surprising influence of anchor text on page titles

In this experiment, Dan Petrovic dealt with the question of whether and how the anchor text of links that refer to one's own page influences the title tag of this page if no tag was previously available.

After he had three links from other pages with the anchor text "Banana" to his site, he was able to observe that his page, which previously had no title tag, was named "Banana" in the SERPs. Thus, the influence of anchor text on the title tag of the linked page was clear.

SEO Experiment 15 - Negative SEO: How to damage your competitor's ranking. What not Heats that you should.

How easy is it to negatively influence the ranking of a page? To answer this question, Tasty Placement carried out an experiment. The target of their defamation was the cleaning service

In order to achieve its goal, Tasty Placement booked cheap backlinks for a total of just 40 dollars, which should point to

After the links were gradually indexed, the following consequences arose.

You can see that from the time the sidebar links were indexed, the page has suffered a drastic crash in the rankings. And this applied not only to the main keyword “pool cleaning houston”, but also to 26 other keywords.

Damaging a page in the ranking is therefore not that difficult.


These experiments show in an interesting way how little concrete we still know about the mechanisms that run in the background of Google's search engine. We can also draw valuable learnings from them and sometimes revise what we believe to be true or confirm existing knowledge. However, since the algorithm changes from day to day, the tests always only depict the currently applicable situation.

Did the experiments and their results also question your knowledge of SEO, or is all of this nothing new to you? We look forward to your comments!

Source: SEOSherpa

Alexander Marques Diogo studied literature and musicology at the University of Hamburg. The passion for online marketing came from a part-time job. He has been supporting the editorial team of since November 2015. In addition, he works as an online marketing consultant in the field of affiliate marketing and SEO for the Hamburg start-up ad-Shot.