“Rank tracking for us is an important metric for measuring the success of our (technical) SEO changes/implementations. With the day-by-day / week-by-week comparisons we can discover the impact of several changes made on our websites and how this contributes in increasing organic rankings. In times where search engines are changing their algorithms every day, it’s important to understand how and when this affects our rankings, so we can react on it.”
Floris Gouw, SEO Coordinator Thomas Cook BE
Established in 1841 with the dream of one man, Thomas Cook has grown into one of the world’s most successful agencies for international travel. The in-house SEO team has been working for more than 4 years with AWR as the main partner for keyword monitoring in Belgium, tracking rankings for different search engines on multiple devices and languages.
Raising a red flag
It all started when Floris’ team noticed inaccurate ranking positions for their neckermann.be domain.
They contacted our customer care team and sent us a report which showed the domain had ranked in the top 5 positions for several days in a row, but suddenly dropped out of top 50 on the next day. Just one day later, it went back up in top 5. The shifts in positions were too sudden and too extreme. It was uncanny, considering that they’d made no changes for those keywords.
One of AWR’s core values lies in the accuracy of the data we provide, so we immediately set out to make an in-depth analysis on the support case that Floris reported. Together with our Search Engine engineers, we started looking at the SERP files that had been retrieved for the neckermann.be project.
24 hours later, the results from the SERP analysis were ready. All the data had been gathered with the correct search definition, which meant that the cause for this behavior was external to AWR.
Yet, we felt like an “it’s not us” approach would not be too helpful for Floris and the team. We continued the investigation at a granular level, trying to look at some of the problematic SERPs from an SEO perspective.
This is how we noticed that on the specific dates when neckermann.be dropped from top 50 rankings, Google was listing URLs from thomascook.be. These pages had almost similar content with their competitor pages from neckermann.be.
Understanding the problem
Once we reported these findings to the Thomas Cook team, they confirmed that the two sites belong to the BE Business Unit, and the offer for Belgium differs slightly between them. It was becoming clear that the major ranking fluctuations were the result of duplicate content on two separate domains.
“Neckermann is part of Thomas Cook, and in Belgium both websites have the same offer (but Thomas Cook has extras). The content on some pages is exactly the same, and we’re struggling with this. The thing we see now is that Google is showing one week the neckermann URL, and the next week the thomascook URL.”
At that point, Floris and his team started to investigate how many hotel pages were cases of duplicate content so they could decide how to approach the issue.
They needed to pinpoint the actual ranking URLs that had duplicate content and causing Google to be unsure of which ones to rank. The team used AWR’s exports to CSV and sorted the data by ranking URL.
Choosing the solution
Now that the stretch of the problem was clear, it was time to think about the fix.
One of the solutions they had was to develop unique content for all hotel pages on both websites, which was more demanding from a human resources point of view.
The second was to make sure the hotel offer pages on one of the websites were not indexed, thus de-indexing the pages in question.
Though full implementation required some time, they set out to work, thus building a strategy with long-term goals in mind. The results would be immediate and solved the duplication issue in line with Google’s best practices.
Measuring the impact
After the change was rolled out on the domains, Floris turned to AWR for monitoring the effects.
Again, the CSV exports proved their utility, showing the actual ranking URLs for each keyword together with the current ranking position, and the change in comparison with the previous ranking data updates.
In a few weeks, the high fluctuations on the problematic keywords stopped. The individual pages from each domain started to rank independently, and the search visibility trends returned to normal.
Some final words
While AWR isn’t an auditing tool, the ranking data it provides can offer a lot of clues about the health of an SEO strategy. In this case, having multiple projects for different domains on the same AWR account led to the discovery of duplicate content which influenced rankings for both domains.
It’s just like Floris said:
“Ranking Tools like AWR aren’t just for ranking, but also for showing us the technical SEO issues we’re dealing with :).”
What about you?
Have you ever dealt with duplicate content problems cross domains? It’d be interesting to learn how you discovered and resolved them!
Feel free to share your thoughts in the comments section 🙂