Data-Driven UX Design and Innovation

There are many convincing reasons for not basing design decisions on intuition, somebody’s opinion or personal experience – it is against the core principles of user-centred design and can lead to many poor decisions (nobody reads users’ minds). When so many tools for collecting data are available, it is hardly worth the risk of getting it wrong, thus data-driven design is getting more and more popularity.

Data can help you make better decisions

Even though there is a lack of a widely accepted definition or agreement on what “data-driven” means, in essence it is basing design decisions on actual data (qualitative and quantitative). Depending on the type and stage of the project, it could involve focus group and usability test to understand user’s needs, using analytics and A/B testing to see what performs the best, as well as various tests and benchmarks – decisions at each stage of a project could be based on clear data.

There are some other benefits too, for example it could remove some tensions within the design team – nobody gets offended that others do not like his ideas, the opinion of the highest person in the company’s hierarchy is not taken unquestionably – if the data shows that some idea is better than the others, it is better. It is also easier to convince your clients that the new design is worthwhile – you can just show them the data, the statistics instead of relying on highly subjective measures such as “more beautiful and contemporary” and sounding expert enough to make them believe you know what you are saying.

Data itself does not drive innovation

Even though these benefits are real and substantial, there is no such a thing as a magic bullet in design that is suitable for all situations and challenges, thus it is important not to get overexcited about data-driving everything.

Besides the risk of making a wrong analysis with a wrong data and confusing correlation with causation, which could be avoided by letting only a qualified person do the analysis, there is a risk of becoming too data-focused and disregarding the fact that innovation and creativity come from a designer, not data. Requiring a designer to back up even small design decisions with clear data will discourage thinking outside of the box, which is essential in creating a fascinating product.

Of course, this is not a problem with data-driven design, rather with it being used too religiously in a hope that following it will solve all the challenges and create what users really need and want. Statistics never suggest anything innovative, neither do users – there is a famous quote by (reportedly) Henry Ford that summarises this idea well: “If I had asked people what they wanted, they would have said faster horses”. Data can inspire and inform, but it can never drive something innovative, it can make the product great, but more is needed to make it awesome.

As everywhere, a balance needs to be found

As everywhere in the real world, there needs to be a balance. An expert designer should be able to know when to get creative and when to validate the ideas using appropriate methods of data collection, thus there is no need for forcing data-focus at every stage or letting creativity go wild without user testing (that would go against the principles of user-centred design anyway).

Mercedes Benz: Left Brain – Right Brain, Paint

Also, much depends on your needs – if you need just a good design (e.g. a decent company website) for your small business in a non-creative industry, there is less need for innovation and creativity than for creating (and fine-tuning) something that needs to be really awesome and stand out from the crowd.

Innovation is about embracing the unknowns of the creative process and the risk of getting it wrong at the first or even eighth iteration.

It is up to you to decide how much of it is required for your project. It is just important not to have unrealistic expectations that by following some recipe and analysing the data correctly you can easily create something outstanding or that the creative process is quick and predictable.

For further reading

• Data informed design, not data-driven design (UX for the masses)
• Data-driven vs. data-informed design in enterprise products (Medium)
• The User Experience: Why Data – Not Just Design – Hits the Sweet Spot (Wharton)
• Instinct Isn’t Enough: Great Design Demands Data (fastcodesign)
• Six Myths about Data-Driven Design (UX Mag)
• Data-driven Design and Generative Design (Toptal)