Designed By Data

How data is helping with product design


The concept of design is one of the oldest ideas in the world. Planning how to do something before embarking on it is at the basis of everything we do, from the most basic, such as cooking food, to the most complex, like building vast civil engineering projects.

Designs have largely fallen into two broad categories; those designed for aesthetics and those designed for practicality. Those companies who have managed to merge the two perfectly are the ones that have had the most success, and it is largely considered to be down to the genius of a particular designer or leader.

However, with the amount of information that we have today, this may not necessarily need be the case.

A strong design should be able to come from the data that you can collect based on the use, or potential use, of the product. This could be from the ways in which people use it, to the basics of design and stress testing.

There are several examples of how this can be used:

Smart Car Creation

Despite being the first major car company over a century ago, Ford has managed to create an innovative and reactive company that has utilized significant technological developments to improve the design of their cars.

For instance, the Ford F-150 has been designed with data to reduce weight and carbon emissions. They have managed to combine this with a sustainable and profitable business model by not simply creating a fleet of hybrid or electric cars, preferring instead to concentrate on the elements of their traditional cars that could lead to reduced emissions.

Through combining user data and sales data, they have managed to reduce weight and increase sustainability without affecting their bottom line, making these changes sustainable in both the environmental and business sense.

A/B Testing Real Products

A/B testing has been a popular method of creating effective marketing choices, but has been underused in other areas.

A quick explanation of the process is that two (or more) variations of something are given to people and the numbers who click or react to each is monitored with the variation with the most clicks being the ‘winner’. It is a very simple concept, but due to the constraints on manufacturing, something that many have not utilized when it comes to products themselves.

With the change in how people are buying products started by companies like Kickstarter, people are now more amenable to looking at products that are not necessarily available or which may be in an experimental phase.

Companies have found that by putting variations of products to the public vote, they can use the data gathered to make decisions about which products to create based on public interest or which changes to make to an existing product through general consensus.

Architectural Design And Testing

As well as needing to build modern structures that look good, one of the keys to the success of a building is how it can stand up over time or during natural disasters.

Testing of this kind could be done on a broad spectrum previously, but with increasingly complex algorithms being used in CAD design programmes it is becoming even more accurate.

With the ability to design down to where the tiniest bolt will fit through the smallest beam, the structural integrity of a building can be created in increasingly complex models. Different variations can be included and tested to establish the strength and quality of a structure in several different scenarios, from natural disasters through to potential degrading of materials over time.

As each individual bolt, beam and screw is included in the design, the ability to test through millions of different scenarios becomes possible, and with faster computing power and complex algorithms, different circumstances can be created to test the overall integrity of any structure designed in this way. 


Read next:

Working At The Boundaries Of Aesthetics And Inference