A/B testing in B2B: a truly effective strategy?

“We just have to do an A/B test”. How many times have you heard this phrase, when launching an email campaign, a new version of the website or a conversion form? In all companies, A/B tests are spreading at lightning speed. The idea: to compare the effectiveness between two versions of the same digital content.

But is it an effective strategy?

Especially in B2B? Is it (really) a email data  relevant lever, when you are an overwhelmed marketer?To answer these questions, Benoit (Growth marketer at Plezi) interviewed Orian Roturier (Head of Growth & Demand Generation at GitGuardian). If you missed this Plezi Decode special A/B test , don’t panic: catch-up session in this article.
A/B testing (also called A/B testing or split testing) is a marketing method for comparing two versions of

the same content. Orian Roturier

email data

also talks about “testing two hypotheses of the same variant”. This testing strategy can be implemented to measure the effectiveness of:
In digital marketing, A/B testing aims to measure the impact of a change, and its performance. For example, let’s take two Internet users: Paul and Marion. Today, they both decide to visit the website of a SaaS software publisher.

Paul will have access to the multiexperience: cx of the future? original version of the website (“A”), and Marion to the alternative version (“B”). These variations of an element A to an element B are presented randomly to our two visitors, as well as to all web traffic. They will have access to different elements: marketing arguments, button colors, images, CTAs or even page structure.
Let’s take an email campaign as an example. The second step is to create two pieces of content, then define a hypothesis. For example: “email version #2 will generate more signups.”

Here we find the performance chleads  indicator to achieve (generate registrations) and the hypothesis to follow (it is version #2 of the email, which will perform the best). To test the hypothesis, it is now necessary to define the “P value” : this value indicates the probability of obtaining the desired conversion result.

Leave a comment

Your email address will not be published.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