A/B testing

With A/B testing you use a scientific approach to investigate which changes on your website or webshop lead to more conversions. This allows you, for example, to improve the conversion rate and thus increase turnover.

A/B test variations
Test results
Simulation

Do you recognize this?

  • A lot of money goes to SEA and SEO, but the results are below expectations
  • You do have an A/B testing tool, but you don't do enough with it
  • Conversion optimization does not 'live' within the organization

And want the following?

  • A structured approach to conversion optimization
  • Based on valid, reliable, and significant AB tests
  • Designed, set up and run by an experienced a/b testing specialist

€10 million+ in annual recurring revenue generated for customers

Highlighted benefits of A/B testing

Fixed costs per A/B test → No surprises afterwards

We agree on the costs per test in advance. These costs include the following activities: Communicating, researching, prioritizing, devising, building, implementing, checking, launching, monitoring, analyzing, reporting and documenting. No hassle with hours of estimates for me and no surprises for you afterwards, it's a pleasure to work together.

Structured process → More results

Recent research by RedEye shows that companies with a structured CRO process achieve better results. That is why we use a well-thought-out process to run every A/B test. Current A/B tests, test ideas and results are of course all tracked in a structured manner.

Extensive reports → Better learnings

The test reports contain statistically well-substantiated recommendations based on the A/B test results. Advanced techniques such as Monte Carlo simulations are also used to predict the impact of a test on turnover. Clear visuals and management summary then make the reporting accessible and valuable to all stakeholders.

Frequently asked questions

Can't we just set up A/B tests ourselves?

No, in any case not A/B testing where the results are valid and reliable. Nowadays, every online marketer or front-end developer can easily set up an AB test in one of the well-known A/B testing tools. However, without knowledge of issues such as sample size, significance, power, flickering, hypotheses and heuristics, the outcome will most likely be useless. A/B tests whose results you can rely on therefore require an expert.

Is my website suitable for AB testing?

It is possible to run AB tests on almost all websites (including web shops on popular platforms such as Magento, WooCommerce and Shopify). As a rule of thumb, a minimum of approximately 1,000 conversions (transactions, leads, donations, etc.) per month is used. So-called MVT (multivariate) tests even require much higher numbers of conversions. If you are far below the recommended number of conversions, AB testing is probably not the best solution to improve your website. In that case, a website analysis or user testing may yield a better return.

Which A/B testing tools are you familiar with?

I have extensive hands-on experience with setting up A/B tests in almost all known A/B testing tools. In total, this concerns several hundred tests that have been set up in Google Optimize, Convert, VWO, Adobe Target, SiteSpect, AB Tasty, and Optimizely. If you happen to use another testing tool, this is no problem. Because A/B testing tools are ultimately very similar in principle, valid and reliable tests can be set up with that tool in no time.

Who will build the A/B tests?

In the vast majority of cases, I will be able to build the devised A/B tests myself. I then place the JavaScript and CSS code with which the tests were built in the testing tool so that the variants can be compared against the control. Exceptions to the foregoing are tests for which information from the backend is required in the A/B test. Consider, for example, tests in which the prices of products are tested to determine whether new input fields need to be added to a form.

How is progress documented?

All experiments are tracked in a so-called A/B testing roadmap. The experiments are each assigned a status here, such as test idea, planned, or completed. This roadmap has been developed as a Google Sheets document. This makes it easy to provide various stakeholders (such as product owners, web builders, managers, and online marketers) with insight into which AB tests are live on the website. An additional advantage of using Google Sheets is that there are no monthly costs.

How much additional turnover does AB testing generate?

This amount of additional turnover depends on several factors. Logically consider the current turnover, but also, for example, the amount of tests that are run per month, to what extent the hypotheses for these AB tests are substantiated with research such as user tests, how well the AB tests are built, etc. However, it certainly happens that a single AB test increases revenue per user by > 10%.

Interested in A/B testing for your organization?

Contact me

Testimonials

Very positive results achieved

Tsjibbe Wiersma, Online Marketing Manager, Profile

Theo has a particularly sharp view of conversion optimization and its added value for the organization's objectives. He wants to get the most out of it (in a positive sense). We worked together at Profile Banden en Onderhoud for a longer period of time and achieved very positive results. A pleasant person to work with, clear in his communication and strong in terms of content. (LinkedIn)

Maintain an excellent overview

Marjolein van den Broek, Account Director, Yonego

I have had a great time working with Theo in recent years. He has helped us in the CRO area on various large accounts. No challenge is too much for Theo and the more complex the more he seems to enjoy it. What struck me is that he was excellent at maintaining an overview of what was going on within the projects. In addition to his substantive knowledge about conversion optimization, customers were also very enthusiastic about his technical skills and his sharp eye. Theo is a very good and pleasant specialist to work with, very loyal and also with a dry, pleasant sense of humor. (LinkedIn)

Helps organizations further

Wim Vis, Management team, DGN Publishers

It may sound a bit rude, but Theo is by far the biggest test nerd I know. And that not only works fantastic, but is also mega-contagious. He knows absolutely everything at the intersection of data, testing and conversion, digs deeper than many and then knows how to easily translate this into clear problem statements. But where many freelancers stop there, Theo goes further. He formulates practical solutions for the problems identified. In this way he helps organizations further. And all with a dose of humor and an extremely pleasant manner. Yes, I can highly recommend Theo without any doubt. (LinkedIn)

Helped with entire CRO process

Dennis Nederlof, Digital Marketing, Nexeye

Theo helped us with the entire CRO process for Hans Anders. From setting up a backlog to building tests and conducting analyses. And he usually has a solution to achieve this for the more technical tests as well. If you want to start experimenting or move to the next level, Theo is a great fit! (LinkedIn)

Extremely skilled professional

Jan de Roode, Communication Advisor, Middin

I got to know Theo as a very skilled professional. It is pleasant to work with him. Theo is solution-oriented and knows how to translate between the client's wishes and the (im)possibilities of the technology. And above all, not unimportant for me as a client, Theo meets every deadline. (LinkedIn)

Pleasant cooperation

Marijke Verkaik, Web Designer, KovokS

We hired Theo to investigate some websites and I highly recommend him. Not only does he offer pleasant cooperation, but also in-depth knowledge about how to optimize a website. (LinkedIn)

Clients

  • Nike
  • Greenchoice
  • Vodafone
  • Financieele Dagblad
  • Pieter Pot
  • Profile
  • Citroën
  • ZorgKiezer
  • Virgin Media
  • Haarshop
  • Liander
  • Schoonenberg
Free and non-binding consultation
Curious about the possibilities? Then schedule a free 1-hour consultation.
Contact me

What is AB testing?

With AB testing, a CRO specialist compares two or more variants of a website or webshop. What has been adjusted in such an AB test variant is determined on the basis of preliminary research and a hypothesis. Statistics will then be used to determine whether there is a variant that leads to significantly more conversions.

How do you perform an AB test?

Carrying out AB testing involves quite a few steps. I have roughly defined it as: researching, building, implementing, checking, initiating, monitoring, analyzing, reporting, documenting. Please keep in mind that some steps take much more time than others.

What is the added value of AB testing?

The added value of AB testing is that it gives you more certainty about the effect of changes to a website. If you simply implement changes without testing them first, they may have a negative impact on things such as the conversion rate, average order value, or turnover.

How accurate is AB testing?

Generally speaking, AB tests are very accurate. However, the extent to which AB tests are accurate depends on how well they are set up. An experienced ab testing specialist will take into account issues such as significance, power, outliers, runtime, etc. This ensures that experiments become more reliable and valid.