3 Great A/B Tests To Help Drive First Transactions

There’s a strong argument nothing is more important than the first transaction on mobile. A big claim, so probably worth breaking down a little.

Let’s start with the fact that the mobile app is by now (it’s nearly 2018 remember) the place where the consumer / brand relationship lives. We use our phone to book tables, order cabs, log exercise, buy stuff and everything in between. So if your brand isn’t picking up that mobile business where relevant, chances are someone else is eating your lunch.

And what’s the single best way to get customers using your app? Getting them to a first transaction. Don’t forget there’s a big difference between someone who is just playing around with your app and someone ‘using’ it in the sense of turning to it every day or even hour to help them with the key tasks they need to get done. The former are nice to have but about as much use as an underwater hairdryer when it comes to the bottom line. The latter are where it’s at.

So why are first transactions so important? Because they are the moment at which most ‘friction’ in any given app has been surmounted. Once a first transaction has taken place, the customer has supplied payment details, registered (or at least shared personal information), and perhaps most importantly got over the mental hurdles associated with ‘becoming a customer’ of your organization.

After all that - the second one is easy! And the third is easier again!

So for any mobile (or partially mobile) business, getting to that all important first transaction is important. And you don’t necessarily have a huge amount of time. Research consistently shows that first transactions tend to happen fast. And if users don’t get to that transaction, they are less likely to stick around and keep trying.

With that in mind, we believe it is incredibly important to run A/B tests on the core user experience that stands along the critical path from first session start to first transaction. Making that journey as smooth as possible is the key to success, and to that end here’s three examples of the sort of tests we’re talking about - try them and start delivering more first transactions than ever before!

1. Registration: Up Front or At Purchase?

Many services require registration to be complete before a first transaction or purchase can take place. But they disagree about when users should be asked to register. Two common alternatives are almost immediately on first session start and during the actual purchase flow. There are pros and cons to both approaches, simplistically ‘stopping users from exploring and experiencing value’ and ‘placing too much burden at checkout’ respectively. But the way to settle the argument is simple: test.

Whenever you ask for registration, keep the process as simple as possible and provide all the contextual help and reassurance your users might want. Without it you’ll reduce successful registration whenever you ask the question.

2. The Checkout Screen and Process

When any transaction takes place, you’re going to have a checkout screen. What exact form it takes is up to you, but given that anything up to 80% of purchases don’t get completed after items are placed in the cart, it’s an obvious place to look. Here’s a few decisions you need to make in this area that you might want to fold into a testing program:

  • How many screens the process takes to complete. The mobile screen is small and fiddly. Putting everything on one screen can confuse users and reduce completion rates. At the same time, too many screens can (by definition!) be frustrating. A/B testing can identify the sweet spot here.
  • How much information you really need. There are benefits to collecting lots of data when you get the chance. But ask for too much, and you’ll start to experience some serious drop-off. As always, whenever you’re in a trade off situation like this, the answer is almost certainly an A/B test.
  • The layout of each screen. Where do the buttons sit? What language is used for key actions? What format for data entry is used? Again, these are all great candidates for some serious testing en route to a smooth, optimized process.

3. App Store Or Stand-Alone?

We come across this frequently, and again there is not necessarily any ‘right’ answer. Using Google or Apple Stores certainly helps when it comes to reducing friction and getting more users over the line. They just have an irritating habit of taking 30% of your revenues. This is usually a big decision - particularly for media organizations - but a well-designed test can help establish whether the improved conversion rate makes up for the loss of revenue. Just remember to factor in the reality that if a first transaction is through the store, future ones will be as well.