What to do when your online store’s traffic volume is too small for split testing?

When you think about conversion rate optimization, the first thing that comes to your mind is A/B (split) testing or multivariate testing. Although those are commonly used techniques of achieving a higher conversion rate, the conversion rate optimization actually doesn’t have a word “testing” in it.

I often see very small online stores trying to run – often very complex to integrate – split tests on elements with really small impact on conversion rate. Test implementation costs a fortune compared to how much they can possibly gain from the outcome, and the duration of test in order to achieve statistically significant results can often be in range of several months, years or even never.

Did you, at any point ask yourself:

  • Why do you even run your ideas through the split testing software?
  • Why don’t you just try them and see how it goes?

When your traffic volume is really low, these tests are usually very unlikely to be a smart investment. Before we continue, I’d like you to read one of the best posts I’ve read this year. It’s a post by Neil Patel explaining what he learned after spending $252,000 on conversion rate optimization.

Split Testing is Experimenting
Image source.

In his article Neil explains that running conversion rate optimization tests is a hit and miss game. And within a year of testing you might get one or two big hits. Might. After hundreds of tests. After hundreds of thousands of dollars spent.

Why do the “big guys” use split tests?

Sites with large search volume can not only afford to use split testing software and test their ideas, but they basically have to do it thorugh testing software.

  • The goal is usually to get a small lift, where on big traffic volume it means lots of money.
  • It’s just as likely if not even more likely you’ll get a small drop instead of a lift while testing ideas.
  • This means they can’t risk and implement changes on their website without testing software.
  • They have to test versions on a smaller sample of their visits so that when inevitable drop occurs they don’t suffer a lot.

You don’t have enough visitors to do it like the big guys? Good news is, you don’t really have to!

The reason they don’t just implement their best ideas and hope for the best is because they have a lot to lose. A small online store with up to a few hundred transactions per month, on the other hand, will not be endangered from a 5% drop, nor will gain unimaginable profit form a 5% gain. You can just implement it and if it turns out to be worse then it was, remove it! Start with a new idea and repeat. You’ll get 2% here, 3% there, but at the end of the year, you might just find yourself with a big number in total.

You made it all the way down here so you must have enjoyed this post! You may also like:

Using Zend for intensive data processing Tomislav Nikcevski
, | 0

Using Zend for intensive data processing

Keeping an eye on small things in eCommerce projects Goran Bokun
Goran Bokun, | 1

Keeping an eye on small things in eCommerce projects

Version control systems in UI design Katarina Dijakovic
Katarina Dijakovic, | 3

Version control systems in UI design

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <blockquote cite=""> <code> <del datetime=""> <em> <s> <strike> <strong>. You may use following syntax for source code: <pre><code>$current = "Inchoo";</code></pre>.

Tell us about your project

Drop us a line. We'd love to know more about your project.