A founder of the design firm Fjord wrote an insightful article for Fortune about design helping companies make sense of data. From the article:
Designers know how to take complex or disparate information and make it tangible, understandable, and importantly, more human. ...
Technologists on the other hand, tend to try to structure data in a flexible way. They analyze it, dashboard it, and cross-reference it to make it more efficient and intelligent.
Both approaches are needed. The data needs to be flexible enough to enable users to manipulate it to find the insights they seek.
Design of the data display and controls to manipulate data need to be clear enough to enable people to work with what they see. From the article:
Here are some ways that design can help companies make sense of their data and identify opportunities to turn it into a benefit for their customers:
- Use data to guide, not dictate: Use data to build a backdrop of understanding and learning, but don't let data trump creative inspiration or minimize the importance of leaps of faith.
- Obsess with your customers: Extend the data insights and value from your organization to your customers.
- Mess up your data: We live in a messy world. How can you make your pure data fit a messy world? Use design to make it personal and emotional, and tell stories.
- Instead of understanding data only, make sure you understand people: You're planning to hire another data scientist to make sense of the growing data mountain you've got? How about hiring a psychologist instead?
- Major on elegance and simplicity: The science of data can be complex, abstract, and ugly. Design will help you simplify and bring clarity. Good design can also present data in elegant and beautiful ways, drawing people in where they were previously turned off.
- Take creative leaps of faith: A creative process that uses intuition and imagination can give you innovations that data alone can't provide. This is crucial if you want to reimagine your business or invent something new.
This list speaks to the fact that data, even well-designed data, needs to have an editorial or curatorial layer that either guides people to the stories that are relevant to them, or otherwise helps to highlight the significance. Getting this right involves deep understanding of what the data can show, knowing who is likely to be interested in what, and using interface and information design to make sense of the display.