Finding The Right Balance in Human-Centred Design Research

Human-centred design is a structured process for solving problems through collaboration and creativity. As the name suggests, it’s also a process that should start and end with people — the people who will ultimately use the product, service or policy we are designing as well as those that will administer it. At its heart, human-centred design is about understanding the needs of those end-users in order to provide solutions which really work for them — solutions which people want to use, rather than have to be persuaded to use. One of the most widely adopted ways of doing this is by following a four-stage design approach consisting of discovery (research), definition (of the problem to be solved based on research), development (of solutions which address the opportunity presented by the problem, including testing and iteration) and delivery (of a solution and a plan for taking it forward). At each stage of the design process we get the opportunity to engage with the people for whom we are designing, which means that by the delivery stage, there should be no big bangs or unwanted surprises.

During the discovery stage of the design process, we use divergent thinking to help us understand more about our hypothesis. This means gathering both qualitative and quantitative data which we can later analyse and use to synthesise a set of design insights and arrive at a clear problem definition.

• Qualitative Research: Usually relies on smaller sample sizes and consists of observations derived from the 5 senses. This is useful if we want to understand how or why something happens but aren’t too hung up on representative samples.
• Quantitative Research: Usually relies on large sample sizes and consists of numerical, discrete and continuous measurements. This is helpful if we want to know how much something happens, or how many of something there are but not that helpful when it comes to understanding why something happens.

It is important that we strike the right balance between both qualitative data and quantitative data in our discovery research. For example, we often use quantitative datasets to help surface potential problems and qualitative research to help us understand more about the opportunities those problems present by deeply understanding user needs, aspirations, and pain points.

During my career, I’ve been involved in a great number of discussions around the way we should conduct research during the discovery stage of a project. In traditional organisations leaders can often be used to making ‘straight to pilot’ or even ‘straight to scale’ decisions based on key performance indicators and data dashboards informed by datasets derived from a few hundred people and the expert knowledge of staff — it feels familiar, comfortable and large enough to be representative. Trying to sell the benefits of starting off with an iterative, collaborative design process which starts with talking to 15 to 20 customers and doesn’t have a defined outcome has sometimes been a struggle; to some leaders, it can feel unnecessary, time-consuming and unrepresentative. Arguably, traditional organisations, particularly public sector organisations, haven’t always seen the need for qualitative research and have risked falling foul of the field of dreams trap (if we build it, they will come), because:

• They think they know what their customers want because they are their customers after all;
• They’ve hit lucky in the past, so why the need to spend time and money on qualitative research now;
• They’ve been in the market for many years so they must be doing something right, why change;
• They don’t think they have time for qualitative research because they are too busy serving customers, that is their business after all.

However, more recently, traditional organisations and the public sector have begun to see the potential of human-centred design approaches and are starting to feel the benefits of understanding customers in greater detail by taking an inch-wide and mile-deep approach to problem-solving, rather than the other way around.

Ultimately, datasets don’t tell human stories and during the discovery stage, stories are key to understanding user need and generating buy-in and sponsorship, however, as we progress through the design process and on to testing it’s often hard data which helps validate our concepts and ensure that buy-in and sponsorship remain when the initial excitement generated during the discovery stage dies down and budgets associated with scaling a solution come into focus.

It is, therefore, important that project teams understand the benefits and limitations of both types of research and find ways of balancing them at each stage of the design process. When planning discovery activity it is essential to think about what we want to achieve and ask ourselves — how will our research help us understand our hunch and what are the best ways of getting what we need to progress?

Some useful things to consider when choosing how to progress discovery activity include:

• Tools: Interviews, surveys, and observations which complement each other — if one research tool leaves a gap in understanding, another can help to fill it;
• Data types: Text, imagery, video and personal artefacts;
• Participants: Customers, employees, stakeholders;
• Researchers: Specialists, generalists, colleagues;
• Research locations: Contextual environments such as homes, workplaces, public transport.

Ultimately the goal of discovery should be to surface enough good quality data to enable us to synthesise rich insights, which in turn, will help us to better define opportunities and provide a springboard for idea generation, prototypes, tests and pilots.

If you get the discovery right and keep the people who will ultimately use and administer the product, service or policy you are designing at the heart throughout the design process, the outcomes will speak for themselves.

Originally published at http://simonpenny.wordpress.com on August 6, 2020.