Glossary
Definition and usefulness of A/B Testing
The need to test
Subject Q (Subject X was otherwise engaged) has a problem: they have a first date and they don’t know whether to come with roses or with a box of chocolates. First impressions are really important for subject X and they couldn’t get enough information about the other person’s preferences.
The ideal thing to do would be to carry out some tests with both options and decide which works best in the end. Unfortunately, for subject Q, first impressions only happen once… the good thing about marketing is that it’s not as complicated as love.
Definition
A/B Testing (also known as split-run testing) is a marketing strategy based on the previous example: we have a situation and two possible solutions, both launched at the same time to analyse which performs better and meets the goals in mind.
In other words, a portion of the audience try different messages to analyse and evaluate the results of each one. In this way, you can see preferences, likes and preferred communication methods of your audience in a more tangible way. That is A/B Testing in a nutshell.
Creativity is all important in a campaign, as long as it’s used thoughtfully and not randomly, this is where the importance of A/B Testing comes in. The main goal of A/B Testing is to make your time and resources more efficient, as we rarely have the opportunity to test through trial and error during a camapign without generating losses, but A/B Testing allows us to do this as we go.
Our A/B test can be done with two completely new messages, or with one new and one old with different variations so as to see what differences arise in the results (in this case, the already existing message would be the control and the new message would be different).
What can I use A/B Testing for?
The use of A/B Testing can occur at different times or under different strategies such as the following:
- In a mailing campaign: A/B Testing will help us see if our clients prefer a banner or a video more, as well as if they prefer more formal communication or something more relaxed.
- In a social media campaign: we can determine different aspects such as images, to the type of content to communication that will be done through A/B Testing. In this case, tests can also be done to segment audiences that will be most beneficial to our goals.
- On a webpage or in an app: we can try different designs, and test both graphics and User Experience.
- In a message: perhaps, for a certain group, a long information-heavy message works better, but another group may need specific, simple nformation with supporting graphics throughout.
- In a call to action: A/B Testing can help us decide if a more direct approach to communication is more effective or one that appeals to our emotional side.
We can finish our A/B Testing once either of these two outcomes happens:
- We reach the deadline for our analysis.
- We have a large enough number of interactions to draw conclusions.
Once we have discovered which message was more successful, we can proceed to apply it to the rest of our audience.
And like love, there are no certainties in marketing
As in many other cases, when carrying out and A/B Test, we don’t get the secret ingredient to success, if something worked for a small portion of the audience, it may well not work for the rest. Or maybe a message turned out to be very effective at one point, but in a few weeks (or even days) that same message won’t be so lucky.
It’s always easier to draw conclusions when we have data to back us up. A/B Testing is an excellent way of doing this.
For A/B Testing to be effective, it’s of utmost importance that we know exactly who our audience is. It may well be that our campaign is successful, but we need it to be successful with our audience. To return to the initial example, some prefer flowers and others prefer chocolates, but subject Q only cares about the preferences of one person (at least, for now).
A/B Testing is a marketing tool that allows us to get a feel for our audience, to understand what moves them, what interests them or what works for them. With A/B Testing, we can get closer to our audience and make a firmer step towards choosing the right strategy.
