Conversion rate optimization should consist of more than throwing up random ideas into A/B tests to see which performs better or a way to resolve arguments. That’s not conversion rate optimization, that’s gambling with a low likelihood of payoff. I’ve seen conversion rates double on sites that implement a sound CRO process but I’ve never seen it happen by chance. The optimizers had (1) a firm grasp of the theories behind online marketing and user experience (2) were able to create valuable hypothesis to test and (3) a plan to test the highest value theories first and iterate on learnings.
To give teams just getting started on testing and optimization, I put together a presentation I use in workshops. This is intended to be an accessible platform to start a CRO program that should then continue to expand and grow into more advanced practices.
I’ve put my presentation from these workshops into a blog format below so that you can use it to start your process and avoid pitfalls many teams go through when starting from scratch. This is part one on defining your audience.
Define Your Audience
The first step to creating an optimized experience is to define who you are targeting. Too many teams skip this step, assuming that they already know the answer. Going through this step of the process give you an opportunity to dive in and discover audience attributed the team wasn’t aware of and to identify opportunities that may not have been pursued. For example, you may know that the majority of your visitors are women, but you can take this opportunity to look at why. Is it that women are more interested in your product than men? Or is it that most of your promotions are done on Facebook which skews female? Is there a difference in conversion rate or purchase behavior for different age ranges? If so, are you targeting the right age? Is your flow optimized for this target audience? Identifying your audience can include pulling information from current customer data, website usage (such as Google Analytics), segmenting customers and visitors, revisiting business goals, and unexpected trends.
Once you have identified your audience or audiences create personas around the demographics you notice. This attributes should be specific, not a range. A persona is not “Michael Mechanic” age 25 to 35. A persona misses its purpose unless it is relatable. I was originally introduced to personas in agile development and they were used to give our development team someone to empathize with. We talked about the personas as if they were a real person and made key decisions based off of what our persona would want or do. A key feature was the specificity of the demographics which allowed us to end conversations about what “a user” would want. “A user” may or may not want to customize their report profiles, but “Caroline Bale” wouldn’t because she was too busy to screw around with settings and wasn’t as technically savvy as our development team.
Marketing teams should be using personas the same way. You should be designing your site and writing content for your persona to avoid confusing paths and generic content. “A visitor” may be equally likely to click on link a, b, or c but you should know which one is most important for your persona. If there are multiple pathways or intentions then you should create multiple personas (within reason).
Personas should be based off of data and research and then validated through user interviews and updated as your team learns more about them. Even if you’ve been working to attract the same visitors for the past 10 years I guarantee there are things you will learn once you start taking another look and questioning assumptions.
Personas should also have pain points and goals that allow you to identify opportunities. An excellent framework to evaluate these is “Jobs to be Done.” I pulled the chart below from Alan Klement’s article on Jobs to be Done which I recommend reading through.
Once you’ve identified Jobs to be Done that relate to your product or service, you can experiment with ways to appeal to them in regards to that goal.