Five best practice tips when it comes to A/B testing

The marketing sector can be a complicated place as new marketing tools and techniques are launched, almost on a weekly basis. Powered by The Drum Network, this regular column invites The Drum Network's members to demystify the marketing trade and offer expert insight and opinion on what is happening in the marketing industry today that can help your business tomorrow.

We’re here again – another Facebook update. Whether they help or hinder our social business goals, the recent stream of updates on Facebook have been coming in thick and fast; the latest rumours being a new experiment of A/B testing on standard page posts. What does this, and other recent Facebook updates, mean for brands?

A/B Testing

Up until now A/B testing, also known as split testing, has been a manual marketing tactic, where businesses can test which headlines, copy, call to actions and so on will work best for their chosen audience. Previously reserved only for Facebook paid ad campaigns, companies can now easily experiment with different target audiences and ad sets to work out which post delivered the best results.

This new exciting update allows businesses to test two pieces of content with two separate audiences (called control and test groups) and then analyse the results - this could be ideal for optimising your social posting plans, thoroughly investigating which piece of content, or copy, works best and therefore saving money when it comes to any paid efforts, and time spent in the face of feed algorithm changes holding you back!

Organic social reach has been turned on its head in light of the newsfeed refocus, promising ‘meaningful social interactions’. In business terms, it meant that family and friends posts would be prioritised, and businesses would struggle even more to have their voice heard.

Facebook announced that the updated news feed would focus on ‘connecting with people and less about consuming media in isolation’. Brands panicked about how to combat this issue, but other recent developments from Facebook and these basic A/B testing opportunities could be a push that brands need to revisit their content and copywriting for social to make the most of the platform and optimise their offering to the best of their ability. When was the last time you read an ad on Facebook, and felt really compelled to read the full post and click? This could be a great chance for all brands to up their game!

When creating ads and social posts for your brand, it can be tempting to do what you think will work - but using intuition, and basing decisions on a ‘feeling’, can be detrimental in marketing. Cold hard facts, and data backed results will be far more effective in optimising your social approach.

Here are some best practice tips to set off on the right foot with A/B testing

1) Test one factor at a time

Consider what variable you want to test - what image works better for clicks? Do emoji’s compel users to click?

These results you collect will be more conclusive if you only test one thing at a time, as you will know exactly what the contributing factor was - testing two completely different ads doesn’t give you anything conclusive as you have no way of knowing what aspect had the most impact.

2) Measure the impact

Keep a record of your ads, variables and the results you’ve done.

This means if you want to optimise other areas, or use a similar audience test, you can come back to what you know worked.

Also this could work in your favour, and if you aren’t doing paid social campaigns yet, you have the proof to convince those purse string holders to push some money into this platform.

Measurement is key in all areas of marketing, so get these basics right.

3) Significance check

If you want extra confirmation that your results are valid, you can check them for statistical significance.

There are many tools out there that can help you do this, just make sure your data pot was big enough to begin with - a general rule to go by is 50 - whether that is 50 clicks, 50 submissions, 50 likes. Decide your metric to focus on, and 50 (generally) will be a big enough data set to draw a significant conclusion.

4) Test everything!

Do lots of A/B tests. Test as many different things as you can. You want as much data to leverage as possible, because in the end you’ll be creating highly optimised social posts that are designed to convert.

This advice does not only apply on Facebook with these new updates, but once you’ve got used to testing social posts, this can spread to your email marketing techniques, your website, and this is CRO 101 - A/B testing is your friend, so make the most of it where you can! But we can only wait and see how our newsfeed content changes, in light of this new testing feature on Facebook! Hopefully brands use this as an opportunity to be more creative, which is always a good thing, for consumers and marketers alike.

Get The Drum Newsletter

Build your marketing knowledge by choosing from daily news bulletins or a weekly special.