This was an innovation project developed for the biggest logistic and brewery company in Brazil. Our team helped to create a visual representation of analytical with a dashboard for their sales and support team.
Along with a header of design and with help from other teammates, my main tasks were UX research, Wireframing, User interviews, Workshops, Mobile and desktop interface design.
The client contacted us with an internal innovation request for designing two dashboards to translate rough complex data into an intuitive and useful visual representation. The objective was to assist them in their investment and negotiation strategies, bringing an overview of the performance of the retail and wholesale chains and their stores during a certain period of time.
Before and after.
The client also told us that there were two distinct user profiles: Key Account Managers and their assistants, both were from all around the country and had different levels of skills.
They worked with different data, so we had to design two dashboards.
The main initial challenges that demanded a lot of flexibility were:
No initial time for user interviews as the business deadlines required previous validation, so we proceeded with desk research and validated with stakeholders before having time to talk to users;
An initial deadline of three weeks to delivery;
The complexity of data and terms: There were dozens of lines and columns in the spreadsheet that the data provider sent and other specific terminology used by the client;
There were also specific tech requirements as the interface had to be replicable inside Power BI tables.
We used the design thinking double diamond methodology to support our process.
Research and definition
After talking to stakeholders and analyzing the tables and spreadsheets, we clustered data separating by market, price variation, products and brands performance, and investment suggestions. The wireframes were created and there was a first-round of validation with stakeholders and developers.
With their approval, we proceeded to a high-fidelity prototype and user interviews.
Users interviews and refining
We discovered some users had a deep analytical profile and inside their teams already created weekly visual reports inside Excel.
We tried different representations for market, brands and product numbers, price relativity, and investment suggestions. During the interviews, users discarded some of them and suggested others. At this point as we realized some visualizations had to be changed, as labels and detailed numbers were very important.
With their expertise, we could get even more precise representations of how to show store performance, what interdependent data existed and what would help them to negotiate and trace better strategies.
There was also an insight about the dark mode, once considered the best choice for saving battery as the users were always visiting clients, it was not their main choice because they wanted to export these reports and send them by email, preferring a white background as it looked better attached to the message body.
Screens after the interviews: at left two screens showing a broader view for Key Account Managers with chain data and list of investment suggestions and at right the visualization for the assistants with store data details.
We updated the interface and ran a final validation workshop with a group of 15 users, where we presented the screens, gave them more visualization choices and the opportunity to manage data inside the dashboard.
After their feedback, we understood that besides their different profiles, the same screen with minor alterations would fit the needs of both users as they worked closely together and accessed both chain and store data.
We also realized it would be necessary to add filters to personalize their experiences as there were different degrees of experience and knowledge in data analysis.
The final design for mobile and desktop with added filters to provide personalized visualization and exporting options. It is composed of highlights with the main performance data above and the data granularity increases as the user scrolls down and interacts with filters.
All data was labeled and tooltips were explaining the calculation memory used in each session and explaining the specific terms.
We also later assisted in online training, as users would have the time to learn how to replicate the visual data from dynamic tables developed previously inside power BI.
"After 3 months using it, I consider the dashboard a reference tool that brought value and agility to my routine, helping me with strategies and teamwork."
- Pedro, Key Account Manager.
Final thoughts and conclusion
It was a complex project due to the amount of data, analytical context, and a short timeline allied to strict business and technical requirements.
I learned a lot about the user's context and by the end of the second week, I could already manage to understand the terms they used and calculation needed, this close communication with stakeholders and users increased our participation promoting the extension of the project term. It also made it possible to plan other dashboards for other profiles within the company, extending the life of the whole project both for us and the stakeholder.
I believe that if we had more time to talk to users we could have learned faster and talked to more people. I also wish we had the opportunity to keep up with user's feedback and metrics after some months, improving the design as needed.
The project was considered successful by our stakeholders and when we had the opportunity to interview some users from the same company for another project that involved data transparency, they responded spontaneously that they considered the dashboard a reference that brought value and agility to their professional life.