A/B Testing Part 2: The Importance of SEO
In our previous episode, we’ve discussed the basics of A/B testing and clarified its definition and value. Hopefully, you are fully aware by now why A/B tests are important for conversion rate optimization and when and how they should be employed. Once you start optimizing your website for conversions, you will want to conduct a successful test that will actually provide you with an insight about your visitor’s experience. However, there are a few factors that you should take into consideration when you start building variations of your landing pages.
For those who haven’t clicked on one of the above-linked pages, A/B testing is a process of comparing two versions of the same content to determine which one delivers better results. It is as simple as that. Now to conduct an actual test could be a tricky and time-consuming task, so pay close attention.
In the following list you will find frequently asked questions and answers that will help you determine on which SEO aspects you should focus and invest your time on to get the best possible results. Bear in mind that CRO, same as SEO, isn’t an exact science and although there are some strict rules and general guidelines behind the process – this is where your creativity will be tested as well.
Is SEO Important for A/B Testing?
Yes, it is. In fact, it is the crucial aspect of your tests and you should devote as much time as you would when optimizing a post on your blog. In general, people tend to disregard the SEO aspect of conversion rate optimization since it is easy to feel discouraged by the amount of work necessary for a successful campaign.
It is of crucial importance to understand why good SEO is vital for your A/B testing process, and on which particular features should you concentrate on. Here are the three rules which you should never break in order to stay on the Google’s good side:
1. No cloaking
If you love a good mystery please note that search engines don’t. Keep your content transparent for both Google crawlers and your visitors. Remember that misrepresented content will be penalized and if you want your visitors to actually purchase something from you – why bother with deceiving them at all? So make sure your title and your URL structure are not misleading your visitors just so you could have a better click-through rate.
2. No-follow or Canonical links
In all probability, you are running a test with multiple URLs. In this case, it is well advised to set canonical links that will guide Googlebots to your original and preferred version, instead of simply marking them as no-follow. If you have access to your webmaster tools then you already know how much search engines love a transparent sitemap with all important data available. Here is a great tutorial that will help you understand and build canonical URLs.
3. 302s instead of 301s
While we are on the subject of links, there is also a matter of redirects that you will use. The main difference between the two redirects, 302 and 301, is that one of them is a temporary redirect (302) and the other is a permanent redirect (301). There is a solid chance that you are using a plug-in that will simplify this process significantly, but nevertheless don’t forget to always point crawlers to the desired URL. Remember that you can’t create two pages of the same website and have crawlers indexing those two pages in the same manner. This is why it is important to pay attention to the following aspect as well.
4. Mind the Time
If you are taking your sweet time testing out multiple versions of your landing pages, Google is going to express concerns about your actions. It is only natural – showing two different pages to different groups of users is simply not acceptable nor it is user-friendly and don’t forget that search engines are all about people. To avoid gaining any unwanted attention, know that you should run your test for a limited amount of time and once you get your results you should as well clean all unnecessary elements and pages from your website.
5. Optimize for Success
No matter which one of your pages delivers the best possible result, it is important to optimize and prepare each page in accordance to its prime intention. Devote as much time as you would to any other official blog post, mind the title, the URL structure, the keyword density and positioning, your images and links or, in simple terms, optimize your page for success. You should know that although there are very small chances that your A/B testing will affect your SEO, your SEO will definitely affect your A/B tests. Keep this short checklist to make sure you got everything in order:
Clear URL structure: We’ve already mentioned that this part is vital, but to make sure that we are on the same page here is an example of how a clear URL structure should look like – http:// yourwebsite.com/your-url-structure
Title tag: Position your keyword at the very beginning and make sure that your title is transparently representing the content of your landing page.
H1 and H2: We are discussing basic stuff at this moment, but nevertheless let’s remind ourselves that your keyword should be presented in the H1 and H2 tags. Primarily because of the search engines, but also you will help your visitors providing them with a clear navigation.
Images Alt tags: Descriptive alt text should always contain your keyword or the entire title of your content.
The aforementioned SEO guidelines are especially important and troublesome as well when it comes to split tests, because the great majority of elements that you will be testing are the very same. You will definitely change your title, your URL structure consequentially, and your subheadings and images as well. Remember to keep all versions SEO optimized and friendly for both users and Google crawlers.
Always mind the final outcome. Once your testing is done, your winning page will affect your website so make sure it doesn’t stand out. If you are targeting a specific set of keywords, follow up on that pattern. On the other hand, you have to maintain a balance between your traffic and your conversion rates so don’t neglect the overall design of your landing page just to gain more non-converting traffic.
