Role
WordPress.com is a hosted version of the popular content management system that powers over 34% of sites on the internet. From mid 2018 to late 2019, I served as the Design Director for WordPress.com’s onboarding flow.
My responsibilities included partnering with an engineering lead to manage a team of designers and developers, setting the onboarding product vision and strategy, doing quality assurance on design and development, running usability tests, conducting competitor analysis, performing quantitative and qualitative analysis, and reporting.
Problem
When we started working on WordPress.com’s onboarding, millions of people were visiting WordPress.com every month to create a site and yet only a fraction of them succeeded. A small percentage of those visitors became paying customers that quickly refunded within their first couple days.
To get a better understanding of what was happening, we kicked things off with a round of usability tests of our existing onboarding experience. I lead the efforts to interview eight participants over the span of a week and synthesized the results. We learned a lot but our biggest takeaways were that it was not clear to our participants what they were signing up for and they didn’t know what to do after their site had been created.
Our bets
Working with my engineering partner, we brought our teams together in a design sprint fashion to design an ideal end state. We started with a high level flow diagram of what we wanted to create and worked through quick iterative cycles to produce a prototype that made the following bets:
- Introducing a preview of the site in our signup flow will help people better understand what they’re working towards and therefore lead to more sites being created.
- Pre-filling our new sites with relevant starter content would help people get up and running quicker.
- Guiding people with next steps after they signed up will keep them engaged, teach them how to use the product, and help them achieve their ideal end state.
Approach
We had a lot of work to do and so I worked with my engineering partner to break things down into smaller chunks that we could implement and measure at a regular cadence. I also took some time upfront to figure out what kind of data we needed to collect and worked with our data team to design the reports that I published on a weekly basis.
While we started implementing our first milestone, I ran another usability test on our ideal end state to see how it compared to our existing experience. It was a crude prototype but it got the point across, highlighted some changes we needed to make, and validated the direction we were heading in.
As we worked through each milestone, I worked really closely with the engineering team helping them prioritize their work and testing new features before we launched them — there were a number of times I’d get right in there and fix some bugs myself. I also continued to run regular usability studies after each major milestone which helped us understand the numbers we were seeing and course correct our design for a better outcome. While we brought our initial vision to life, I continued to work with the design team to think ahead and experiment with different approaches to see what we could learn. There were times that paid off and we incorporated some of thinking into our existing work.
Outcome
Over the course of a year we built and tested a lot of changes to our onboarding experience and we had our shares of ups and downs. The biggest takeaway during this time for me came from our usability testing and working closely with our customers. The regular contact allowed us to make changes on the fly and showed in the impact we made by increasing our purchase rate, reducing our churn, and reducing the number of customer support requests related to setting up a website.