Solving problems with triangles
How data science, UX design and research work together to create valuable products
By Jason Bressler, Subhasree Chatterjee, Archana Shah and Sanket Shukl
When it comes to product development, there is no shortage of tools, methodologies, and frameworks available to produce data needed to inform decisions. The challenge is figuring out how to take the data from these various sources and put it together in a way that allows you to articulate and defend design decisions or, in some cases, convince stakeholders that a product isn’t even worth building. In other words, the data needs to tell a story.
We, the authors of this article, are a self-formed group of UX designers, UX researchers, and product analysts with a common goal — to create the best possible user experience for our customers. Our company has become much more data driven over the past several years, which is obviously a good thing. However, just having more data doesn’t necessarily equate to better products.
In the first post of this series, we will outline the process that we have developed (including a link to a downloadable checklist for you to use). In future posts, we will provide actual case studies showing how our process has helped us to make better products.
Becoming data driven — the early days
At the start of our journey into being a truly data driven organization, we were thrilled to have access to much more data than ever before. Our company had invested heavily in analytics and data science and we were now able to supplement our qualitative data with more quantitative data. However, early on we found ourselves working in a way that wasn’t harnessing the true power of the data. Questions were either routed to the product analytics team or the UX research team, and the answers lived separately.
Enter the triangle…
As our team evolved and we began working more closely with one another, we soon discovered the power of storytelling through data triangulation. Connections and insights that may have been missed before were now being surfaced.
We have now evolved to co-designing experiments and co-sponsoring innovative research projects! (Note: watch this blog for more stories as we have them). By triangulating research and behavioral data at each phase of a project, we derive insights from both qualitative and quantitative data that enables us to see the whole picture. We get the answers of what the users are doing in the product from behavioral analytics data, and why they are doing it from the research/survey data. Using both sets of data has helped us to parse out and focus on actual user problems instead of wasting time solving the wrong problems.
Though we are continuing to evolve, we have taken our learnings and developed a process that we try to apply to every project.
Our research process
Research questions often depend on the maturity of an idea, as do the definitions of success metrics. It is okay for these questions to evolve as time goes by and value is assessed. The image below lays out how the quantitative and qualitative data is combined at each stage of product development. A lot depends on a well written hypothesis!
How to make this process work for you
- Get involved at the beginning. Data (both qualitative and quantitative) is very important at all stages of the product development lifecycle, but arguably the most important at the beginning, while assessing the feature itself.
- Ask the right questions.
- Is there really a problem to solve?
- How big is the problem?
- What change are we trying to bring about and why?
- How do we drive the necessary change in behavior?
3. Eyes on the prize. How do we measure success or failure?
- Create metrics to define success. Just one composite metric or some vanity metrics won’t help. Have specific sets of metrics for a specific chunk of a problem.
- Make sure to have proper tracking mechanisms to measure and have a visual setup to track the movement of those metrics.
4. There is no such thing as no data, even when there is time pressure
- Use usage data, customer support call data, nps data, heuristic reviews, etc. to gain a better understanding of how many users utilize the feature and what their pain points may be.
5. Collaboration is everything! Having Product, UX and data science consistently talking to each other is important both to avoid wrinkles going forward and work efficiently.
We have created the following checklist that we think will help you ensure you are getting the ‘right’ data or answering the ‘right’ questions for your product development.
You can download the checklist here.
Stay tuned…
Watch this blog for detailed success stories on how we have combined qualitative and quantitative data to not only build better products, but also save hundreds of thousands by not building features!
