For years, digital marketers and web design agencies developed tools to build aesthetic experiences for customers. And they were successful. The quality of today’s online environments is infinitely higher than it was in the past, and it’s all thanks to the drive to make the digital world more beautiful.
However, visuals are only part of the story. While they are important, they’re not the whole show. Digital era designers have to move beyond mere artistic motifs and incorporate data into their design choices to maximize business relevance. Pure aesthetics is okay for art galleries, but it just doesn’t cut it in the commercial world.
Why Care About Data-Driven Design?
Data-driven design is a self-explanatory concept: businesses use the information they collect to concoct digital experiences that encourage their audiences to convert.
Primarily, that’s why data-driven design is so compelling. Companies that use data in their design decision-making see improvements in metrics across the board.
Research from MIT futurists and authors Andrew McAfee and Erik Brynjolfsson suggest that firms that use data were around 6 percent more profitable than their competitors and five percent more productive.
But what does it actually mean to be “data-driven?”
According to Designing with Data: Improving the User Experience with A/B Testing by Rochelle King, Elizabeth Churchill, and Caitlin Tan, a company achieves data-driven status (as opposed to “information informed” status) when it uses quantitative data to inform all design decisions. Thus, a genuine data-driven design strategy kills all instinct and subjects all design decisions to scientific inquiry.
Data-Driven Design Principles
Building a design strategy around data requires adopting a specific set of principles to ensure that you use the available information correctly.
Here’s what to do:
Set Up Your Experiment
Before you begin using data, you need to set up a hypothesis – as if you were conducting a scientific experiment. Ideally, you want to create a “testable statement” – something you could potentially refute, were you to find contradictory data.
Here’s an example:
Red call-to-action buttons are better than green ones for clicks.
You can also construct hypotheses as “if X, then Y statements.”
If red call-to-action buttons are more enticing than green ones, then they should receive more clicks.
Here you have a statement you can test. If more people click your green CTRs than your red ones, then you can refute the hypothetical statement (and use green buttons instead).
Confounding variables are extraneous features that cloud the purity of your experimental setup.
In the above example, the only thing that should change is the color of the button. Everything else must stay the same. If it doesn’t, you don’t know whether the secondary variable variation is causing the difference in click rates or whether it is the color of the button itself.
Incorrect sampling, such as splitting your audience between red and green CTRs along demographic lines, could skew the results. Ideally, you should randomize visitors when A/B testing to wash out any idiosyncratic characteristics.
Data Collection Techniques
There are several places you can acquire the data you need to implement data-driven design:
- Interviews: This technique relies on sending out questionnaires and surveys to people on your contact list. You can either do this via email or call people up on the phone.
- Competitor analysis: In many cases, there’s no need to reinvent the UX wheel. Competitors often implement designs that have gone through significant testing and already identified various strengths and weaknesses in their approach. Using SEO and traffic data can often help identify patterns that will also work for your brand.
- User flow: Lastly, you can collect data from your website and other consumer channels. A/B testing and heat map data are essential tools to determine your designs’ overall impact and success.
Data-Driven Design In Practice
Mabbly used data-driven design in its work with healthcare services provider Symphony Care Network. The company needed a strategic plan to better reach their target audience and be the first brand patients considered when searching for post-acute care. Using quantitative research methods, Mabbly first identified Sympony’s target market and then created a data-driven go-to-market plan which focused on enhancing customer touchpoints throughout their journey.
Mabbly overhauled the UX and UI features of the website and used data insights to make it easier for customers to interact with the company. The results were spectacular. Symphony benefited from a first-of-its-kind Airbnb experience for customers in the elderly care category, radically transforming the entire experience.