Making website improvements: Is it as simple as A-B-Let’s C?

Image source: https://firstsiteguide.com/tools/analytics/ 

There’s always room for improvement and better engagement. Do you want to make changes on your website but not sure what to do? Like everything else in life, it’s tempting to just Google it: “A/B test ideas.”

I did. The ideas are endless – 276,000,000 results, in fact. But where do you start – change the page layout, make the home page longer or shorter, a funnier headline, bolder text colors, adjustments to image size and placement, etc? Optimize for voice search? Those are just a few examples to improve the user’s experience (UX), but, well, maybe you should focus on changes to maximize search engine optimization (SEO). But, experts say not to focus too much on the tricks and treats of some SEO tactics or the site’s rankings will actually be penalized. Wow, there’s a lot to consider. We must make changes carefully.

To put it simply, your focus for improvement should be to create site/pages that are uniquely valuable to your target searches/audience. Those are the words of Rand Fishkin, an SEO expert from Moz, who writes about building the perfectly optimized page.

“A user's experience is made up of a vast array of elements, not unlike the search engines' ranking algorithms. Satisfying all of these perfectly may not be possible, but reaching for a high level will not only provide value in rankings, but through second order impacts like shares, links, and word of mouth" (Fishkin 2013).

How to begin making website improvements

Just like the rest of your marketing and communications, you need to be strategic to achieve optimization that you can then measure. Focus on what the customer is thinking rather than what your gut feelings and wild guesses may be. I recommend a tool such as ChangeAgain to help you manage and strategically make changes to your website that will create big returns, a better UX, and not to mention SEO rankings.

ChangeAgain helps small and medium business do just that. Their first experiment is even free. The visual optimization tool helps marketers do experiments with A/B and split testing without coding. Rather than coding, it uses a visual editor interface. Check out the below video to understand better.

Although it’s easy to get started, as they show us, ChangeAgain wants everyone to be successful so they suggest a game time out (so to speak) – develop a Testing Strategy by doing a poll or survey with the site’s key audiences first. Once you know what your users are thinking, it’ll be much easier to come up with a “winning B version,” as they say on the company blog. “It’s a direct response to feedback from your customers. You’ll get to the bottom of the real issues on your site so you can start improving your conversion rates. And then you can bask in the glory,” wrote Natalia.

While setting up the experiment you can set a goal for the experiment and then set targeting parameters: who will see the experiment by URL, county, and device. You can also define the percentage of site visitors that you want to see it. This customization really allows you to fine-tune the potential learnings and really get in the minds of your users/customers.

I’d like to remind you that it’s good practice to only make one change at a time that might affect others, especially if it’s in a funnel or flow that you are trying to improve.

How did the experiments perform?

All of the experiment data fully integrates with Google Analytics. This is an excellent feature, because you can then use all of GA’s features, standard reports, and your custom reports, just as if it’s a new page on the site. This is particularly helpful when segmenting visitor demographics of “page A” versus “page B” and their behavior in relation to time spent on the page, flow, funnel, KPIs, and so much more.

Image source: Change Again

Another excellent GA data point for many experiments is mobile viewing compared to desktop. Focusing an experiment on mobile devices could help determine if you are getting search engine referrers from voice search. These long tail searches act differently than semantic searches, so you’d need to make changes in key word/key phrase, tags, etc. Did those changes make a difference via referrer or device? If you are like me, then, you’d want to review the GA data of the bounce rate of these pages to determine if the user was getting what they expected. If the bounce rate is high, you could run another experiment by adding internal links at the top of the article, if it’s a blog, for example, to see if they click through to see more content.

To see GA and ChangeAgain data side by side and in a visual way is a great benefit for marketers to more easily understand, make decisions, and take action. According to the Social Science Research Network, a visual interface can be especially helpful to 65% of the population who are visual learners, and are more stimulated when information is visually presented to them, (Bradford, 2011). Furthermore, visuals are processed 60,000 times faster in our brain than text (Pant, 2015). This is important not only for analyzing the data – but for building and designing our sites! Think about it this way – we are users of Google Analytics. They want us to have a good UX, too, and utilize their tool, even though it free. I’m not sure how, but they are making money on us.

Let’s talk reliability 

The ChangeAgain tool supports websites, domains, intranets, extranets, and e-commerce sites. And, the data is 99% accurate, according to their marketing materials. I wish I knew more about the limitations of ChangeAgain’s reliability, so if you are going to buy their services this should be discussed with the company before signing a contact. You’ll want to know how it may or may not line up with the known issues of GA. The known issues that marketers cannot fix include, but are not limited to: browser disabled JavaScript, lack of cookies (for a few reasons), cookies timing out, same device with different user, same user with different device, and data sampling (Henneberry 2013).


In closing, I recommend ChangeAgain to strategically make changes that users want and will improve their experiences and your result. Avoid making changes that just give you more data to review and then are unsure what to do with. Data is only good if you can take action from it. ChangeAgain has the ability to make that happen.


Reference

Bradford, W., (2011, Sept. 1). Reaching the visual learner: Teaching property through art. The Law Teacher Vol. 11, 2004. Available at SSRN: https://ssrn.com/abstract=587201

Fishkin, R. (2013, August 6). A visual guide to Keyword Targeting and On-Page SEO. Retrieved from: https://moz.com/blog/visual-guide-to-keyword-targeting-onpage-optimization

Henneberry, R.  (2013, January 31).  Why is Google analytics inaccurate?  The Daily Egg Blog Site.  Retrieved on December 28, 2013 from http://blog.crazyegg.com/2013/01/31/why-is-google-analytics-inaccurate/

Natalia. (2016, May 7). Let’s Start A/B Testing. ChangeAgain. [Web blog]. Retrieved from  https://www.changeagain.me/lets-start-ab-testing/

Pant, R. (2015, January 16). Visual Marketing: A Picture's Worth 60,000 Words. Retrieved November 5, 2018, from https://www.business2community.com/digital-marketing/visual-marketing-pictures-worth-60000-words-01126256