Microsoft’s Nonprofit arm had an unusual problem: they couldn’t give away their product for free. It didn’t matter whether it was Office, Azure, GitHub or Dynamics, users that wanted the product simply couldn’t get it, despite both sides best efforts. To make matters more complicated, a few weeks into the contract COVID arrived, adding a real sense of urgency to getting these tools into the hands of nonprofits.
When we joined the project, the registration process had an extremely high abandonment rate. When a user was able to complete the process, the average completion time was an eye-popping 45 minutes.
In addition, the process had to support localization into 238 different countries including right-to-left languages. It was also desktop-only, required a lawyer to complete, and took 2-4 weeks to receive an approval.
Client
Microsoft Nonprofit
Responsibilities
Planning, Discovery, Competitive Analysis, UX Design, Front-End Development, Usability Testing
Approach
We conducted several series of interviews with users worldwide so that we could better understand the challenges they faced with both the existing process and their day-to-day responsibilities that could impact us. From those conversations we developed three thin personas, a journey map and a healthy backlog of requirements.
Armed with our user’s needs and an understanding of the pain points, we performed a competitive analysis to discover how the existing process faired. This proved to be pretty eye-opening for the MS team, effectively driving home the point that to increase registrations they not only had to make the process easier to use, but had to remove as much friction from the process as possible.
Trimming the Fat
From there we set an internal goal: remove half the content from the process, then go through and attempt to remove half of it again. This required partnering with additional teams (Marketing, Legal, Development, PM) across Microsoft to ensure we didn’t run afoul of any internal policies or regional laws.
Execution
With the registration process as lean as possible, we began redesigning it from the ground up. Without adding bloat, we also incorporated ideas from the comparative study while addressing feedback and concerns from the user interviews. We utilized a mobile-first approach, producing mockups in four languages concurrently to ensure user satisfaction was high regardless of the device they chose or language they spoke. The languages were English, German, Arabic (for right-to-left testing) and Japanese. As this was a project for Microsoft, we used their existing design language, Fluent.
Reception
We performed four rounds of moderated usability tests with seven users each, with all of the personas represented in each round. While there were no consistent friction points during testing, we did notice after the second round that one of our personas (Project Managers) generally had more friction points than the others.
To address it we conducted another series of interviews to understand their issues and address them. The third round of tests went smoothly, and we moved on to telemetry. Users generally found the process quick and painless, on both mobile and desktop. We were later told the localization team gave similar feedback after the different languages were implemented.
