In the years that I have spent working on data projects, I’ve found those that deliver the most value and have longevity in an organisation are those where the end-users of the data delivery have trust, ownership and pride in the end product.
In order to make this happen, there needs to be true empathy and understanding between the technical developers and the end-users – a typically difficult ask, as most developers focus purely on the technical elements of the data projects (data structure, modelling, security, governance), and most end-users are focused on the business benefits of the data.
In addition, business intelligence and data requirements are constantly changing as new insights come to light, and data projects, once delivered, therefore require continuous enhancement in order to support the business – again, a divergence from a typical developer’s view of a fixed start and end to a project.
Rachel Woods, in her paper on “A design thinking mindset for data science”, talks about a “unicorn data scientist – as one who excels in the technical, business strategy and communication aspects of data science”.
We are fortunate enough in our business to have quite a few of those, but they are few and far between in our industry, so our challenge has shifted away from finding those unicorns, to growing them, focusing on developing not just the technical expertise of our consultants, but also cultivating a keen interest in the people, processes and organisations they are consulting for.
Following a discussion with a friend who heads up a creative school, and learning about various leading organisations (Apple, Netflix and UberEats, to name a few) who have utilised design thinking to develop effective solutions to challenges, I started looking into design thinking as a framework that our consultants could use to develop business- and user-centricity in their work.
In this article, I discuss what I’ve discovered about design thinking thus far, and how it could be practically applied to data projects.
Design thinking is a construct that looks beyond the traditional processes, procedures and practices of logic-based problem-solving, to rather address challenges from the perspective of the people who will use the product or service.
We’ve needed to unlearn some behaviours in order to take things to the next level.
Unlike more conventional problem-solving methods that emphasise logic and data, design thinking encourages designers to consider the desired outcomes, their own personal biases, and the potential obstacles that users might encounter along the way.
Designers who employ design thinking are less concerned with the how and more interested in the why. They don’t just want to know how to achieve a certain outcome; they want to also know why they must cut a path in a certain direction. Design thinking encourages designers to examine all the different options available to them, identify the pitfalls that may lie ahead, and then pick a path forward.
Data teams who use design thinking in data projects ensure they are aligned with the business strategy – focusing on the “why” from a business perspective, rather than the “how” from a purely technical perspective.
Building a data project that is focused purely on technical delivery is like building a house without first understanding the people who will inhabit it. If the architect doesn’t care about those people or understand what they need and want from the project, he or she probably won’t build a house that’s suitable for them. The house will simply fulfil a list of stipulated requirements – three bedrooms, two bathrooms, open plan kitchen, etc.
The variety of designs that will tick these boxes is endless, but to ensure the occupants truly benefit from the project, their unstated needs and wants need to be considered – and the architect needs to use his intuition and experience to advise and guide the occupants on choosing the best possible options for them.
Doing the same in a data project will inform your decisions throughout the project lifecycle, ensuring you don’t just build something that works, but something that works for the right people at the right time.
Design thinking involves five phases:
- Empathise – understand your users and context.
- Define – define the problem you need to solve.
- Ideate – produce many ideas and learn from them.
- Prototype – create quick build and learn (not necessarily build and burn) solutions that can prove the effectiveness of the ideas.
- Test – confirm the idea and decide on a course of action.
It is important to note these are non-linear. For example, prototyping might yield a deeper understanding of the processes, which in turn will lead to more ideas.
Rikke Friis Dam and Teo Yu Siang, in their article on “What is design thinking and why is it so popular” state the phases can be performed in parallel, repeated and circled back to, with the goal of working in a dynamic way to develop and launch innovative ideas.
It is a departure from traditional project management, but there is room in the data project space for a hybrid approach incorporating the principles and phases of design thinking in each step of the project; in essence, merging the two. For this to work, you need to keep your users close to the process – move away from delivering in a big bang approach, and keep them involved at each step of the project.
Don Norman, the father of user interface design, explains that designers should take the original problem as a suggestion, not as a final statement, and should resist the temptation to jump immediately to a solution to the stated problem – something I’ve certainly been guilty of, especially when the data and/or the tech is really exciting! We’ve needed to unlearn some behaviours in order to take things to the next level.
Our traditional approach to developing projects:
- Business analysis
- Technical analysis
- Data extraction
- Data modelling
- Data integrity testing
- Data apps development
- User workshops
- User sign-off
Nothing changes with the steps of the development cycle, but within each phase, we are working towards using the principles of design thinking. Certain phases may have a larger component allocated to the design thinking phases (eg, business analysis vs technical analysis), but at each step of the way, the challenge is to keep the usefulness of the solution; ie, the end-user experience, in mind.
In a departure from the traditional project approach, this often requires returning to earlier steps – for example, when developing the front-end data app, we might in the ideation phase discover a new way of looking at the data that would add more value, but this, in turn, may require returning to the data modelling phase to change the data model.
Fortunately, we live in an era which has a plethora of easy to use business intelligence tools that effortlessly lend themselves to this incremental, design-thinking based approach, allowing us to quickly churn out prototypes and play with various approaches to answering the business questions.
In summary, applying design thinking to data projects helps the developer to connect with their end-users, empathising and understanding their true challenges, and using this clarity to develop solutions that are truly value-adding, and key partnerships that are long-lasting and symbiotic.