API Developer Hub
Broken up into three phases, my role included:
Discovery phase
Analytics
Gather requirements and constraints
Stakeholder interviews
Content audit and inventory
Set goals and objectives
Exploration phase
Competitive analysis
Build personas
User journeys and task flows
Card sorting
Tree testing
Surveys
User interviews and synthesis
Information architecture
Category/sub-category labeling
Testing phase
Qualitative usability testing of high fidelity wireframes and prototypes
Design iteration
Discover
Analytics | Gather requirements | Identify constraints | Stakeholder interviews | Content audit and inventory | Set goals and objectives
Looking at qualitative data we could glean where users were dropping off, which pages yielded high and low traffic, where users were missing intended flows. With this information we began to set goals and objectives, and make initial guesses as to where user needs and business needs align and diverge.
With customer and business needs in mind, as well as our design approach, our ultimate goals were to:
Encourage users to explore and engage without API offerings
Allow customers to leverage our expertise while creating seamless products for their customers
Forge scalability partnerships and set expectations for the process ahead
Explore
Competitive analysis | Personas | User journeys/task flows | Card sorting | Tree testing | User interviews | Information architecture
From tree testing we gained a number of key insights regarding site content - specifically that labelling was inconsistent, pathways were convoluted when users attempted to create a new app, retrieve keys and copy sample code. Users also had trouble figuring out the necessary steps for Authorization. This provided opportunities to re-organized and re-label content. This also revealed a clear need to provide more educational and instructional content across the site.
Test
Qualitative usability testing of high fidelity wireframes and prototypes | Design iteration
We hosted 3 rounds of testing for each page in the site. Based on the feedback received we then went back to the drawing board to adjust key task flows accordingly. With each new design iteration there were multiple rounds of 90-minute task-based remote user testing sessions. Representing 3 out of 4 core personas, our developer pool spanned the globe. Test subjects were located in Singapore, Malaysia, Taiwan, Australia, Hong Kong, Mexico City, Miami, San Francisco, Indonesia, Thailand and the Philippines.
The new Documentation designs were highly regarded by all users for its clean structure and thorough display of all the necessary information. Users were also able to find specific information easily when given a task.
In the previous design, users were forced to select a market before they could view any API documentation. This cumbersome process was frustrating for users, but unfortunately this was a system limitation. Working within the system's constraints users preferred our proposed solution of a "default market selection," rather than pogo-sticking back and forth between navigation levels. Additionally users are given the option to adjust their market or API selection contextually, which was greatly appreciated by developers who worked across multiple markets.