
Earlier than you begin: if you happen to’re unfamiliar with the rules of statistical search engine optimisation split-testing and the way SplitSignal works, we’re suggesting you begin here or request a demo of SplitSignal.
First, we asked our Twitter followers to vote:

This time the vast majority of our followers have been WRONG! Learn additional to search out out why.
The Case Examine
The worth of search engine optimisation content material on ecommerce class pages, whether or not on the prime or backside, has been the topic of a lot debate. These items of content material are sometimes created primarily for search engine optimisation functions, to extend visibility for related key phrases, relatively than to boost the person expertise. Google’s personal John Mueller has commented that these texts are generally seen as makes an attempt at key phrase stuffing, which may negatively affect the web page’s general view. Whereas some content material may be useful, he warned in opposition to overdoing it with massive article-style content material beneath product listings, as it could actually make it tougher for Google to know the web page’s intent. Google additionally acknowledged that there will not be a big distinction by way of search engine optimisation between inserting the content material on the prime or backside of the web page however {that a} well-crafted piece of content material will help customers higher perceive the class and its choices.
Naturally, these discussions and remarks from Google have piqued our curiosity. Creating search engine optimisation content material for a whole lot of 1000’s of class pages could be a important funding by way of time and assets. Due to this fact, it’s essential to know the true worth of such a content material. These insights have prompted us to conduct a sequence of cut up checks to look at the affect of various approaches to search engine optimisation content material on ecommerce class pages.
We lately performed a cut up check for certainly one of our main ecommerce purchasers, the place we eliminated the search engine optimisation content material from the highest of their class pages to guage its affect on natural efficiency and search engine visibility.
The Speculation
The web site in query wished to check if eradicating their search engine optimisation textual content, about 100 to 200 phrases on the prime of their class itemizing pages, would have a big affect on natural visitors.
Internally, our shopper had intensive discussions in regards to the worth of the search engine optimisation content material and hypothesized that eradicating it could enhance their search engine optimisation efficiency. They believed that this alteration would permit their class pages to raised concentrate on offering customers with the product listings they got here to the web page for, placing extra emphasis on search intent. Moreover, they anticipated that Google’s algorithms can be higher in a position to perceive the intent of the web page and, in return, may presumably enhance its visibility in search outcomes.

The Check
Our shopper sought to evaluate the affect of eradicating search engine optimisation content material from the highest of their class pages, and so they used the SplitSignal instrument to conduct the check. They chose a pattern of class itemizing pages as both management or variant.
The cut up check ran for 30 days, throughout which Googlebot crawled and listed round 97% of the pages.
The Outcomes
After 30 days of testing, we discovered that eradicating search engine optimisation content material from the highest of class pages didn’t have a big impact on natural visitors on the 95% confidence interval. The cumulative view of the check confirmed that the blue shaded space didn’t carry out beneath or above the x=0 axis, indicating that the check was not statistically important on the 95% stage. In consequence, we can not confidently attribute any improve or lower in visitors solely to the change in search engine optimisation content material.

Based mostly on these findings, we conclude that eradicating search engine optimisation content material from the highest of class itemizing pages doesn’t have a big affect on search engine optimisation efficiency.
Word that we’re not evaluating the precise management group pages to our examined pages. However relatively a forecast based mostly on historic knowledge. The mannequin predicts the counterfactual response that might have occurred had no intervention taken place. We evaluate this with the precise knowledge. We use management pages to present the mannequin context for developments and exterior influences. If one thing else adjustments throughout our check (e.g., seasonality), the mannequin will detect and think about it. By filtering these exterior components, we achieve perception into the true affect of an search engine optimisation change.
Why
Based mostly on the outcomes of the cut up check, we are able to conclude that the small piece of search engine optimisation textual content faraway from the highest of the ecommerce class pages didn’t have a big affect on natural visitors for our shopper. The speculation that eradicating the search engine optimisation content material may enhance search engine optimisation efficiency and assist Google higher perceive the intent of the web page didn’t maintain up underneath testing.
Moreover, the textual content itself appeared to haven’t any search engine optimisation worth, as its removing had no discernible impact on search engine optimisation. This means that the content material was not contributing meaningfully to the web page’s general search engine optimisation efficiency.
Moreover, we performed an evaluation on natural impressions utilizing the SEO A/B test analyzer to guage if there was any affect on visibility in Google search outcomes. The outcomes of this evaluation confirmed that the removing of the search engine optimisation content material didn’t considerably affect natural impressions in comparison with our modeled management group.

By analyzing each natural clicks and impressions, we are able to confidently conclude that eradicating the search engine optimisation content material from the highest of ecommerce class pages didn’t have a big affect on search engine optimisation efficiency or visibility in Google search outcomes.
In abstract, our findings counsel that this explicit ecommerce enterprise might not want to take a position important assets in creating and optimizing search engine optimisation content material for his or her class pages, particularly if the content material supplies restricted worth to customers. As a substitute, the enterprise can concentrate on creating high-quality product listings and guaranteeing that their class pages are well-organized and straightforward to navigate.
We’re excited to proceed difficult Google’s statements in regards to the worth of search engine optimisation content material on ecommerce class pages by way of our ongoing cut up testing efforts. Now we have been conducting checks on eradicating bigger items of content material positioned beneath the product listings. These outcomes will likely be shared in future case research, offering additional perception into the true affect of search engine optimisation content material on ecommerce class pages.
There’s no one-size-fits-all resolution in search engine optimisation. What works for one web site might not work for one more. So, if you wish to succeed, check what works for you!