The Challenge: There is a growing need to help people find the answers to their questions in their data. However, for those who are not data analyst specialists, this is a hard and time consuming task with a high learning curve.
The Solution: Through intuitive design DataHero gave users control and access to their data, no matter the source, in one singular location.
The Outcome: In three years we grew DataHero’s platform into a robust data analytics solution. Resulting in a successful exit. DataHero was purchased at the end of 2015 by Cloudability.
Designing a data analytics tool that was intuitive and understandable to users who weren’t necessarily data experts was the definition of my role at DataHero. There was no other tool in the market to compare it to and it meant doing deep user research into the target audiences behavioral patterns, understanding how they processed data and their goals. We kicked off the process by interviewing users with a range of skills in data analysis, from the masters, to the novices. We then prototyped simple interactions with pen and paper to get some visual cues on their behavioral patterns. With that information we then started prototyping mvps which we’d test and iterate on rapidly. This system allowed us to theorize and solidify conclusions quickly without the need of developer resources. Our first MVP was built on a design language whose cues came from that initial research period and which would guide us throughout the products life span.
As logical and straightforward as data may be, the way it is analyzed, absorbed and parsed through is very personal. We learned early on that our target market had very strong opinions on what they wanted to do with their data to do for them, but that those opinions were based on very individual needs. This stemmed from the users relationship with their data. They’d ask us for specific features to take specific actions to find insight in a specific piece of data. These requests would often be influenced by their interactions with Excel and would often be items they didn’t necessarily need in their day to day data analysis. This meant that as we iterated rapidly on our product, we were also iterating on our interview and user testing structure to get to the insights we needed to create a successful product.
Through our research phase two personas who’s needs and wants overlapped in the product emerged. The first persona was a power user who wanted to fine tune his data search parameters in a series of data sets and have access to the information in real time. The second user, was a novice that wanted guidance on how to best leverage their data and gain insights they felt they may be missing out of. Both of these personas for different reasons needed to feel that the tool was empowering them. Our original product flow had the user start with a blank slate from which they could pull in data, visualize it and manipulate it. Though intuitive on what to do next, this set up made the user feel like they had to put in extra work before feeling successful. We had catered to the users need but ignored the power of emotional engagement. After all, our users felt some level of ownership over their data. Once we acknowledged the importance of giving our users an “aha!” moment, our flow changed completely. Instead of leading our user to our data visualization tool to get started - we created a dashboard and populated it with charts created from their data. The charts themselves drew from the simplest parameters of metric to dimension. However the automation of this one detail gave the power users a sense of strength of our analysis tool and the novice user a starting point from which to engage and understand their data.
As we built out the product and added new features, tools and connectors to our platform we took in a lot of feedback from our users that indicated that they wanted to have full access to all of the editing, viewing and tool options available. The logic was that they would not know what to do if they couldn’t see it. This rapidly lead to a state where the product was visually cluttered. We turned to UI/UX research focus to better understand what were the most common actions our users took on a daily basis. By this point we had data coming in from our analytics to help us create a good testing strategy. Through it we managed to parse through the actions that users needed to take in the product, actions users wanted to take and align those with the actions that we needed the user to take to be successful. This data is what set the parameters for what would be the final update on the interface structure and the visual design language. By setting a strong hierarchy of information and a set of rules on each section of the product, through design, we could visually guide our user to successfully engage with their data.
The result was an intuitive and powerful data visualization and analytics tool accessible to anyone.
“Datahero, represents how the market is segmenting for data visualization and data-analytics tools based upon level of skill, complexity and overall richness and sophistication of the technology.” Techcrunch in an article about the product.