Design by Numbers?

Design is often thought of as sitting at a notional half-way point between the disciplines of Art and Science. It can be described as Craft in the same way that a fine piece of handmade furniture is a balance of engineering an artistry – aesthetically pleasing yes, but also deeply functional. As our industry has matured it has become more process-led. As technology has evolved and the impact of our work become more measurable website design has become increasingly data-driven.

To some designers this is seen as a slow erosion of creative scope. Decisions are reduced to a series of tests where short-term measurable outcomes are favoured over broader artistic goals. In a world in which Google tests 40 shades of blue to see which performs best and Facebook massages its users’ news feeds to see if it can manipulate their emotional state, perhaps the pendulum has swung too far towards the science end of the spectrum?

Data benefits

The rationale behind a data-led approach is solid. The purpose of most design work is to achieve specific goals that match up to business aims and objectives with measurable outcomes. If an A/B test can tell us that one design converts visitors to customers better than another then surely that is something to be embraced? Is it just ego if as designs we don’t feel our work should always need to be validated in this way?

At The Pixel Parlour we spend a lot of time considering the impact of design decisions on the user experience. Often this involves testing, be it face-to-face usability testing or automated multivariant design testing. But we also value that serendipitous moment of inspiration, that unexpected spark of creativity.

We believe that truly great design has the ability to see the bigger picture and understand that a website can be more than just the sum of its individual parts.

Finding a balance

I think most web professionals from both sides of the fence would agree that there is a balance to be struck. Where that balancing point is will be heavily dependent on the nature of the particular design challenge being faced. If you are Amazon embarking on a redesign of your checkout system then it is probably appropriate to test every detail because of the potential returns from even small improvements.

As designers we must recognise that the world has changed. There are circumstances where testing should dictate design decisions and the outcomes be treated as absolutes. In others we need to trust the instincts of experienced designers to make the right call and see testing, research and data as tools to support the creative process rather than replace it. But to help clients feel confident doing that we also need to ensure that we are doing the best job we can articulating the rationale behind our design decisions.