Postman’s Authorization

Sending successful API requests posed a significant challenge for developers on Postman. The main obstacle was the complex and time-consuming nature of the authorization process. The documentation provided by API producers were unclear, creating a barrier to efficient API usage. We decided to create a MVP focused on simplifying and expediting the authorization process. I led the design of MVP and also conducted usability testing to validate the idea. As a result, successful API request percentage increased by 21.2%

DURATION

2 months

MY ROLE

UX Researcher and Product Designer

METHODS

Semi-structured User Interviews, Online Data Scraping, Moderated Usability Testing, Wireframing, Rapid Prototyping, High-Fidelity Prototyping

MyMandi

Background

At Postman, I primarily worked as a Product Designer but also took on User Research responsibilities at various stages of this project. My goal was to design the MVP for a new assistive authorization and authentication experience, making the process smoother for developers. In this case study, I’ll highlight how my user research contributed to the project’s success.

Problem Context

One of the key steps in sending a successful API request is authentication and authorization, which varies depending on the type of request. However, developers frequently struggle with this step.

Discovery Research

Given the constraints of time and budget, conducting user interviews with Postman’s end users wasn’t feasible. Recruiting participants and scheduling interviews would have delayed the process. However, identifying pain points was crucial to defining what this MVP should address.

To gain firsthand insights, I tested the top 20 APIs on Postman, focusing specifically on the authorization and authentication experience. To further validate my observations, I analyzed qualitative content from Reddit discussions, where developers shared their struggles with API authentication.

Additionally, I conducted 1:1 sessions with the Chief Engineering Architect to discuss and refine the identified pain points. Through this triangulation of data, we defined the core problems that needed to be solved in the MVP.

Final Problem

This process helped me refine the problem statement, which I then structured using the Jobs-To-Be-Done (JTBD) framework to ensure clarity, user-centered focus, and business alignment. It also helped in bringing all stakeholders onto the same page.

Proposed System

Before diving into design, it was essential to define how the proposed system should function. Drawing inspiration from 'Thinking in System's' by Donella Meadows, I collaborated with the Chief Engineering Architect to craft an initial framework. Feedback from stakeholders helped iteratively refine this foundation before starting the design process.

Proposed Design

After establishing a solid foundation, I transitioned back into my role as a designer to craft intuitive interaction flows. Through iterative refinement, guided by continuous feedback and reflection, I honed the designs to directly address user pain points and business needs. Below is the design used for usability testing.

Usability Testing

With no dedicated research team, I again took on the researcher role. During the design phase, I formed two hypotheses that shaped my decisions:

Due to budget and time constraints, recruiting end users was not feasible. Instead, I tested the feature with Postman employees who were developers and closely aligned with our target audience but had no prior exposure to the new system. This allowed me to validate the hypotheses effectively and identify areas for improvement.

Findings and Design Improvements

Issue 1: Users using shortcut key ignored the Contextual Nudge
Issue 2: Users using shortcut key ignored the Contextual Nudge

Impact

We received positive feedback from the producers on how these summary emails are beneficial for them to understand and measure their consumer’s interaction with their APIs.

Reflections

  • During this project, I learned a lot and made a significant impact. I realized the value of using hypotheses when making design decisions.
  • When creating the MVP (Minimum Viable Product), it's crucial to keep the feature set minimal while maintaining high quality. This allows us to effectively test the viability of our idea.
  • Figma prototyping fell short because the API authorization process was complex. To truly test our solution, I took the beta version built by developers and conducted user testing within the company.
  • User testing was eye-opening as it revealed flaws in our initial hypotheses, guiding us toward iterative improvements in the user experience.