Wysa is a mental health & wellness app with over 2.5 million users. During the COVID-19 pandemic, it's gotten significant traction with employers for its B2B offering. Our goal design an analytics product that employers could use to track employee wellbeing, quantify impact and engagement. This was designed in Tableau, using Wysa brand colours.

Our problem statement:

How might we... showcase analytics our customers want in a straightforward and impactful way that demonstrates: a) the value Wysa provides to employees, and b) insights on areas of employee wellbeing that need attention?

‍

Research

Analysing Current Data

The first step we took was to understand the current situation that we were working to improve. We looked at and analyzed a number of Excel sheets that business clients had requested tor insights. This was to get a better idea of the metrics we were working with at present, and the way it told the story of Wysa's value to them.

‍

Research

interviewing Stakeholders

In a number of interviews with both the product team and the client users, we established that the metrics that we had been using up until now were metrics the clients had asked for exactly. But in reality, it wasn't a rigid ask - at its core, they just wanted data that answers the question "how is Wysa doing?", and they expect a cohesive, detailed picture of that answer. I got a better image of client concerns of whether Wysa is not only successful at creating better mental health support, but also scalable - how possible is it to implement parts of Wysa's platform at a larger scale, would doing so involve a lot of friction, and does that lead to scalable value? Those two ideas - success and scalability - really mattered. Meanwhile, the product team was concerned with whether we would be reflecting the value that Wysa has to offer. I got a better idea of the power of how data is represented and sliced, in the way it portrays both useful information and a real story of value.

‍

Research

Competitive Benchmarking

Once I had insights from the people at the heart of the problem space, I broadened my scope to how other products are addressing a similar need. I took inspirations from different analytics products - we looked at the industry standard in Google Firebase, Facebook Analytics and branch.io,  where we then extracted insights about each product's strengths and any weaknesses.

‍

Wireframing

Rough Storyboarding

We created a Google Slides storyboard (as non-designers were familiar with Slides) with a long list of data fields, which were populated with real data we were currently using to understand how useful they would be in real life. But going through the storyboard, we saw the story they were portraying as inconsistent with the goals we were trying to fulfil, and that story wasn't reflective enough of either point of the problem statement: Wysa's value, or insights that actually need attention. A number of metrics were working - but others just weren't.

‍

Wireframing

A High-Fidelity Pass

Over multiple iterations, we ideated on new lenses to look at any and all of our available data (not just the data we were already using). We ended up creating some new fields and letting go of others.  With the final set of data, we started wireframing it, with three broad sections: Takeup, to contextualize the first touchpoints between Wysa and its users; Engagement, to assess how users use the platform in real life; and Helpfulness, to demonstrate outcomes in the lens that matters most - how helpful Wysa has been to them.

‍

Design

The Product

The style guide I constructed for these dashboards is centered around a dark theme UI. As the client would already be parsing a lot of information on a data-dense dashboard, I was drawn to the sleekness, sensory softness and low eye strain of a dark theme - aspects that make it such a popular design trend. I played around with Wysa branding colors for the primary palette, with variations on shade to give a calming progression - even the background palette, rather than shades of grey, varies on dark purple in keeping with Wysa branding. I picked DM Sans as the font for its modern and unique character.

We built and designed the dashboards in Tableau directly, working within the layout constraints of the software (of which there were many) to deliver a polished, aesthetically pleasing UI. Based on the metrics we were looking for, we set up data connections and created the fields, and I then added the styles and constructed the dashboard to tell an engaging informational narrative in each one. Here's one dashboard (not permitted to show you more), with numerical metrics blurred out for obvious reasons:

‍

Outcomes

These dashboards have gone live with the client (can't tell you who, but it's a large health insurer), and the testimonials have been outstanding. Given feedback on their evolving needs, the dashboards continue to be iterated on to keep addressing them.

This project taught me a whole new way to think about data and the narrative it tells. The same data can be perceived under multiple lenses and establish different contexts for this data. I'm glad that I got a taste of incorporating conscious decisions about data into my designs, because the market is starting to orient itself around being able to leverage data to create stories. Design has always been about storytelling for me, and I just learned a new medium to do so. I'd never designed for a business client before this either, so this project was full of firsts that grew my range as a designer that can adapt to different environments to produce results. As data's becoming increasingly depended on in just about every market, understanding how the simple representation of it can tailor the context it creates is a critical skill as I go into my future career.