Case study

Rapid’s API Hub

Standardising and simplifying the API development workflow, so developers can design, test, monitor and publish APIs in one place.

Visit Rapid’s API Hub
Company
  • Rapid
  • March 2022 - March 2023
Outcome
  • 30% increase in conversion
  • 100k active users
  • Top VS Code extension
Contribution
  • Research & testing
  • User experience
  • Design systems
  • User interface
two user interfaces

The brief

Challenge
Designing, testing and monetising APIs meant juggling multiple platforms. Developers faced a clunky development process and a slow time to value.
Objective
Standardise and simplify the API development workflow at every stage, so developers and organisations deliver faster and code more efficiently.

My role

I owned the developer tools experience at Rapid, from conception to release. I worked with product and engineering squads across three time zones to make the tools that developers use every day simpler and more effective.

Two very different users

Most of Rapid’s paying users weren’t technical, but a small group of experts drove half the revenue.

Non‑technical users

70%

of paying users

Seasoned API developers

5%

of users, generating around half of revenue

Any change had to make APIs approachable for newcomers without slowing down the experts.

Understanding the existing product architecture

API development covers a vast technical landscape with a steep learning curve, especially for people without a computer science background. Rapid’s tools already covered most of the workflow, but the architecture grouped features by what they were rather than what users were trying to do.

We ran remote tests with each user group and I turned the findings into empathy maps. They showed that moving between tasks in a journey was the hardest part. That led to our central hypothesis.

Our central hypothesis

A goal-oriented architecture would help users finish tasks faster, and give us clearer sight of our conversion funnel.

The Customer Journey empathy map for Sarah Cox, a developer new to Rapid: how she felt at each stage of building an API, from her first project to testing, and the opportunities we found along the way.

Core design principles

Four principles guided every decision, and each one shaped a specific part of the product.

Onboarding and shared understanding
Introduce core concepts and terms early, so users can navigate the product with confidence.
Guided progression
Support users from start to finish with clear steps, reducing friction and uncertainty at each stage.
Clarity over complexity
Use simple, familiar patterns and plain language to make functionality accessible without technical knowledge.
Adaptive experience
Offer depth for experts while keeping the core experience approachable for everyone else.

Making APIs accessible to everyone

Onboarding and shared understanding. Curated API discovery reduced the need for prior knowledge, while search-led entry supported people who already knew what they wanted.

To meet developers where they work, the hub also had to live inside their IDE. Designing for Visual Studio Code meant balancing powerful features against an already busy interface, while respecting each user’s theme and interaction habits.

Two Visual Studio Code windows showing the Rapid API Hub extension: API discovery and the NBA API’s endpoints

The Rapid API Hub inside Visual Studio Code: browsing recommended and popular APIs, then opening the NBA API’s endpoints, all without leaving the editor.

Rapid API Hub discovery in VS Code (deck slides 7–8)
resources/images/rapid/discover.jpg

Discovering APIs in the Hub, with curated collections for newcomers and search for people who know what they need.

Design documentation for the API Hub in VS Code: the builder flow for unsigned, registered and paying users, endpoint screens for each, and responsive breakpoints. It let us build the extension with engineers, for engineers.

Supporting users from discovery to first call

Guided progression. Step-by-step flows reduced uncertainty, and pre-configured requests sped up setup, so users could get from finding an API to making their first call without getting stuck.

Making a first API request (deck slide 9)
resources/images/rapid/first-call.jpg

Pre-configured requests take people from finding an API to making their first call.

Reducing complexity through familiar patterns

Clarity over complexity. Standardised endpoints built familiarity across APIs, and building requests inside the product simplified setup.

API testing in VS Code (deck slide 10)
resources/images/rapid/testing.jpg

Building and running API tests without leaving Visual Studio Code.

Designed for beginners and experts alike

Adaptive experience. API linting offered simple entry points with advanced control when needed, and a consistent experience across the web, tools and integrations.

API linting on the web and in VS Code (deck slides 11–12)
resources/images/rapid/linting.jpg

Linting rules and results, consistent across the web and Visual Studio Code.

Iterating on a live product

The empathy maps gave me what I needed to make the case to business leaders for redesigning the architecture, in staged releases that fitted a lean, agile way of working.

They also helped other product managers prioritise their own work. We fixed small issues in existing workflows while planning the bigger changes, so Engineering could keep shipping while Product and Design worked out the best path.

Handoff documentation for exporting logs as CSV, annotated screen by screen: the logs table, processing, log details and downloading exports. Built from existing design system components, work like this took days rather than weeks.

The current journey from testing locally in VS Code to becoming a subscriber, mapped across discovery, account creation and acquisition. Markers flag feature and strategic issues, which aligned stakeholders on where to focus next.

Squad coordination across three geographies

Our team was spread across San Francisco, Central Europe and Tel Aviv. Working well together took careful scheduling, the right collaboration tools and clear communication habits.

Working across so many perspectives and cultures made me more adaptable, and taught me to be deliberate about how and when a distributed team makes decisions.

Rapid team members infront of a large billboard saying Rapid

The whole team in one time zone for once, at our 2023 kickoff in Marrakech.

Outcomes

After a year of iterative design, conversion rose 30% across every point in the workflow, from onboarding to monetisation, and across both core user segments.

Our IDE extensions had the highest daily active users of any of our platforms, reaching 100k users in their first week. In the Visual Studio Code marketplace, Rapid’s extension became the most popular that month.

30%
increase in conversion across the whole workflow
100k
IDE extension users in the first week
#1
most popular VS Code extension that month

Learnings

Be lean, and prioritise impact

Limited resources meant a short discovery period, so I focused on hypotheses with the biggest impact for the business. Accepting that I couldn’t master every technical detail at once helped me focus where design could add the most value.

Balance urgency with rigour

Urgency and rigour were both core values at Rapid. Knowing when to move fast and when to slow down was what let us keep delivering value without cutting corners.

What I’d do differently

Get closer to a complex domain sooner

More dogfooding and more time with technical peers would have shortened my ramp-up.

Share design patterns across the team

Docs, dashboard and SDK examples each grew their own conventions. A shared pattern library would have made the product feel coherent sooner.

Align business units across regions

Regional teams served customers with different levels of API maturity. Closer collaboration would have caught more edge cases earlier.

Tighten feedback loops with developers

Usability issues often reached us through support tickets. More frequent sessions with real customers would have caught friction sooner.

Next case study

Times RadioUX lead, 2019–20

A new audio platform for live and on-demand radio

Led UX for a new national radio station, from testing early concepts to launch.

600k
weekly listeners
3.96m
weekly listening hours
View case study
Times Radio app and website showing the breakfast show