How to Optimise Conversion and Customer Retention through A/B Testing

Read time: 8 mins

How well do you know your customers?

‘Better than anyone else!’ you might reply. Through careful ICP research, you know how old they are, what kind of thing they do for a living, what other places they like to shop, etc, etc. 

But do you know them well enough to predict the detail of how they’ll behave as they move through each stage of the customer journey? Are they more motivated by the colour red or blue? Do they find a descriptive paragraph inspiring or prefer a short, staccato sentence? 

You could rightly protest that these are unfair questions. The truth is no one knows their customers that well. Not at first anyway. 

That’s why A/B testing is such an important part of building your customer journey. It gives you the information you need to optimise conversions and build a customer retention-boosting experience that customers return to again and again. 

Careful testing of different options shows you which ones customers respond to best across your entire customer journey, it means you can stop designing your customer experience on hunches and educated guesses, and get real, hard data to base your choices on. 

Then, you can roll out the results of your tests to a wider layer of customers, boosting retention and conversion rates across your customer base.

What to test

When it comes to testing your referral campaign and what messaging nurtures advocates the best, there are some tried and, er, tested things that offer insights into how to boost referrals and drive retention: 

• Type of offer: Do your newly signed-up customers respond better to a cash discount or a percentage off their next purchase? 
• Type of image: Is a product image on the referral page the best way to encourage people to refer? Maybe a lifestyle image depicting friends socialising, for example, is a more motivating choice? 
• Copy and messaging: Is shorter or longer text more engaging? You can test whether getting straight to the point works better or if more playful and verbose text helps you get more of your brand identity across. 

Testing by cohort

While traditional A/B testing typically tests two or more variations side by side to a randomly selected audience, testing by cohort puts customers into a group. This group or cohort then gets shown the same variant throughout the journey. 

This enhances your results. Some of the many factors that influence people’s responses to messaging are:

• The time of year they signed up 
• Where they live
• The economy

As you can see, there are a lot of variables at play, so being able to control for all of these makes your results much more reliable. Once you have controlled for external factors, you can test multiple versions of your offer (or even completely different offers) to different groups of customers at the same time. 

This approach allows you to learn:

• What genuinely resonates best with your customers
• If aspects of your messaging work better on some groups than on others.

This second point is important when thinking about how you can get your A/B testing to serve your overall customer experience. You can boost conversion and customer retention by intelligently integrating A/B testing into your wider marketing strategy. 

Incorporating A/B testing into your segmentation strategy

A 'cohort' refers to a defined group of customers who share similar characteristics. Some of these we have already talked about, but there are other things more directly relevant to your business that you might want to consider when A/B testing:

• Repeat customers
• VIP customers
• Whether customers were referred by a friend
• Whether they themselves are frequent referrers

You might have already segmented your customers into groups like these. By A/B testing on them, you can more accurately learn what resonates with each segment and thereby optimise performance according to those specific customer behaviours and preferences. 

You can test on both existing segments and create new ones:

• Creating a cohort for an A/B test: If you have a specific hypothesis that involves a group defined by certain characteristics (for example, customers who have purchased in the last month), you might create a cohort specifically for that A/B test.

• Using pre-existing segments for an A/B Test: If you already have defined segments in your customer base, you can use these pre-existing segments as cohorts for your A/B tests.

As long as the group used for the A/B test is defined and consistent, you can gather significant and representative results that allow you to optimise for different segments, maximising customer retention and optimisation across your customer base.

A word of caution is in order here: Segmenting your customers splits your traffic, so creating several different cohorts can extend the time required to achieve statistical significance. While this may not pose a significant issue for businesses with a large monthly customer base, it's an important factor for smaller companies to weigh into the equation.

A/B testing case studies

Several leading brands have optimised their referral journeys with A/B testing, allowing them to develop combinations of copy, images and offers that maximise conversions and develop personalised campaigns.  

Bloom & Wild

Rigorous testing on a wide range of cohorts made referrals one of their primary acquisition channels for the flower brand. 

• Over 30% of new customers were gained through referrals.
• In the program's first year, referrals contributed to over 15% of new customer acquisitions.
• A notable 45% of referred customers became referrers themselves, illustrating the program's impact on customer retention and expansion.

Biscuiteers

By running A/B tests on their sweet-toothed fans, Biscuiteers doubled the number of new customers coming from referrals when they offered “20% off” compared with “£6 off £30 spend”.

Kurt Geiger

The fashion brand experimented with different creative to see what resonated the most:

• Descriptive copy proved to be more appealing, with 21% more shares and 98% more new customers than shorter messaging. They also tested different images and found pictures of friends socialising and sharing to be the most persuasive. 
• Their willingness to learn from their customers paid off in the form of retention: 24% of revenue from referred customers comes from repeat purchases. 

A/B testing as a cornerstone of your customer retention and conversion strategy

With A/B testing, Mention Me clients typically see a 4X improvement in their referral campaigns by month 6. It takes a bit of time to build up the data that lets you confidently implement optimisations, but over a relatively short period, the results are impressive and make a real impact on your ability to retain customers and optimise your referral journey. 

If you're already a Mention Me client and you’d like to speak to us about building your own A/B testing strategy to drive more value from your customer base, speak to your CSM. For clients without a CSM, get in touch here.

If you're not yet a client but you want to get serious about engaging your biggest brand fans and optimising your customer retention strategy, Mention Me’s Customer Advocacy Intelligence Platform gives provides the insights you need to create sophisticated yet simple to implement A/B testing strategies that convert: Request a demo.