Until late 2015, JSTOR had two services with two separate ecommerce checkouts; a newer one with a more effective design, and a considerably older one. They did not share any infrastructure and the older checkout process had a lower conversion rate and was prone to technical problems. We decided to merge the two workflows by modifying the newer checkout process to accommodate the older service.
We kicked off the project with a workshop where I lead the team in an activity to map out how the new workflow should work. This activity exposed assumptions and places where design was needed.
With the help of another UX resource on the project, I created wireframes and documentation, iterating in response to new information, and supporting development and QA as they built the new workflow.
Adapting an existing interface to incorporate new elements constrained the design space in a way that designing a new interface from scratch would not. However, this approach let me build upon the research that had gone into designing and optimizing the more recently designed checkout interface.
In the months following the release of the new ecommerce workflow, revenue increased year over year by 22%, indicating that the new workflow was an improvement over the old one.