Why demographics are unhelpful in conversion optimisation
Demographic-based personas which describe characteristics such as gender, age and income have long been used in marketing.
Demographics help marketers segment and target customers, but if you’re trying to change the way people behave on your website so more people convert, demographic information alone is not going to help.
Consider the following scenario. You’re trying to improve your website and so you’ve undertaken some user research. You’ve discovered the following about your customers:
- They are mostly men
- They are mostly between the ages of 50 and 70 with many being born in the late 1940s
- Many of them own homes in the South East of England
- Many are divorced and are on their second marriage
- They are wealthy with most being in the top 1%
This is a fairly comprehensive list of characteristics and you’re happy to discover so much about your visitors. You start tweaking the website to improve the experience and increase sales based on these findings.
You change the colours to make them feel more ‘premium’. You invest in a CSR scheme donating to a local children’s charity and promote the scheme in your website footer. You alter the imagery on the website to try to align it to what you feel your customers probably look like.
Nothing works. Your A/B tests continually make no difference.
Then one day you discover that your two best customers (who incidentally, fit your customer profile perfectly) are Ozzy Osbourne and Prince Charles; 2 people with extremely different perspectives, drivers, priorities, behaviours and attitudes.*
The value of behavioural personas
It’s now clear that whilst your customer demographic profile might help with your marketing, it’s not going to help you improve website conversions.
When we do user research, discovering who your visitors and customers are is useful as long as it’s combined with behavioural characteristics. We want to know things like:
- How they think and what they value
- How they behave both before and after buying our product or service
- How they behave on your website in its current form
- What’s important to them when they’re considering your product or service
- Other websites they use when looking for your product or service
By recognising patterns in our users’ behaviour, drivers, thought processes, and then forming personas which are representative of those patterns, we’re able to understand and empathise with the human beings we want to serve.
We can then create behavioural personas which are based on the patterns of behaviour and drivers we’ve observed.
Here’s an example of a behavioural persona:
John is retired and struggles with mobility so he shops for clothing online. He wants the shopping experience to be as easy as possible. He’s able to see well but struggles with colour differentiation. He’s able to navigate and search quite well but does find filling forms in hard and often can’t recall passwords. He’s quite an anxious online shopper and would prefer to go to a physical store to touch and feel the products before purchasing. He shops online because of his lack of mobility, rather than because he’s short of time.
This example gives us a lot of information beyond John’s age, gender and income. It tells us what he needs and what drives him. With this information, we can start to see how the website needs to work to serve John.
Testable hypotheses could be:
- If we allow users to checkout without logging in, even when we detect an account associated with their email address, then more people will complete the checkout process.
- If we give users clear steps so they understand what will happen after they fill in the enquiry form, then more users will submit the form.
- If we give users clear indication of the day and time their product will be delivered, then more people will complete the checkout.
By combining the insight from user research with insight from the various sources of analytics data we have available, we can start to solve real website problems users are facing.
When we solve real problems and provide an experience which is closely aligned to users’ needs, we can change the way people behave so that more people buy or enquire.
*I’ve stolen this example from a speaker I saw a few years ago. Unfortunately, I can’t remember who it was. If it was you, please do get in touch and I’ll make sure you get the credit.
7th December 2018