Design an on-boarding experience that made sense to developers and let them see their data in New Relic faster.
New Relic is a product for developers to observe the health of their system across all technology stacks. New Relic’s leader created the first tool ever for application monitoring and built the New Relic company to continue leading the industry.
For many years New Relic focused on product-led growth; they used developers' love of the software to increase user acquisition. However, years went by and the focus on onboarding became difficult to keep up with multiple products and the ever changing developer landscape. Fast-forward to 2020, and New Relic was losing over half their users from the signup to deployment process.
4 months with multiple releases
(1) Lead designer
(1) Engineering scrum
(1) Lead PM/GM
Interaction & visual design
We were losing half of our customers at every stage of the product journey in the onboarding funnel.
Our largest customers had higher conversion rates because they were getting a “white glove” service from sales and the customer service group.
Product led growth makes it easy for all customers to figure out how to use the product themselves and lowers the toil for the sales and customer service group teams.
We were a 2 person team (GM and Lead designer) to start and we needed to hire our team as well as bring the product organization up to speed on getting back to our roots of product-led growth. We put together some conceptual wireframes as a way for the product org to envision what we could do. Fully knowing we would still need to research and learn more from our users, we called out our hypotheses along the way for future testing.
Research and experimentation
There were some areas we could release right away that were based on best practices for product-led growth as well as baseline data we already knew. For the sign up page, our data already showed for every form field we added, the completion rate decreased by 2.5%.
Our first test reduced the number of fields from 15+ lead generation fields to 4 form fields. It also restated the value proposition of the product. The results from the A/B test showed better conversion and usability testing confirmed we were matching users expectations. The team stopped the test and pointed 100% of the traffic to the new experience.
While we were running and learning from usability testing on the existing flow, we created new variations of the flow that were testing an assumption about users not being able to understand which New Relic product to choose when starting to set up the source data. One model kept an uncategorized view and the other a categorization on data/product type.
Original product selection
A combination of uncategorized and categorized data sources satisfied the needs of most of our users.
This led to a layout with:
Logos with names
Users were able to find the data sources they wanted to use for observing their technology stacks.
Users added more data sources than they originally planned on adding because they could see all the options.
We got 3x number of sign ups during launch.
"Much better than what I expected. I can actually find out what my services are."
— Kalagee, developer