2020 State of the API Report

2020 State Of The API Survey

Tooling for APIs and Development

API tools

When respondents were asked which API tools they use, Postman came out on top, garnering mentions from 94.5% of responses. SwaggerHub was the only other tool earning above single digits, with 23.3%. SmartBear (8.8%), Mulesoft (7.6%), and Insomnia (7.6%) rounded out the top five.

API-first leaders were more likely to mention Postman and SwaggerHub than others. Respondents with 6+ years of API development experience were more likely to mention Postman, SwaggerHub, and SmartBear than those with 0–5 years of experience.

Postman: 94.5%
SwaggerHub: 23.3%
SmartBear: 8.8%
Mulesoft: 7.6%
Insomnia: 7.6%
Readme.io: 5.9%
Oracle/Apiary: 5.8%
Kong: 3.9%
API Fortress: 2.7%
APIMatic: 2.1%
Talend: 2.1%
Paw: 1.9%
Stoplight: 1.7%
Restcase: 1.6%
Rest United: 1.3%

Multiple responses allowed

Download icon

Platform vs separate tools

We asked respondents whether they prefer a single platform or a mix of tools to design, document, test, and deliver APIs. A combination of both was the most popular answer, garnering more than 50% of the responses.

Respondents with 6+ years of API development experience were more likely to report using a combination than those with 0–5 years of experience.

I/we use a single platform: 33.3%
I/we use a separate tool for each function required: 15%
I/we use a combination of a platform for several functions and other separate tools: 51.7%

Download icon

Types of tooling for producing APIs

We asked survey-takers about the types of tools they use to produce APIs. Coding/programming tools (87.6%) received twice as many mentions as the next most popular, automated testing (43.1%). Code review tools were the third most popular at 38.7%.

The remainder of the tools mentioned were primarily API-specific tools. Interestingly, respondents with 6+ years of API development experience were more likely to report using coding/programming tools to produce APIs than those with 0–5 years of experience.

Coding/programming: 87.6%
Automated testing: 43.1%
Code review: 38.7%
API documentation: 34.2%
API debugging/manual testing: 31.6%
API designers/API mocking: 26.6%
API monitoring: 23.4%
API performance / load-testing: 21.2%
API developer portals: 17.1%
API management: 17%
API publishing: 14.6%
Security testing: 14.4%
Data visualization: 11.7%

Multiple responses allowed

Download icon

Types of tooling for consuming APIs

We also asked survey-takers about the types of tools they use to consume APIs. Mentions of coding/programming tools dwarfed all other API consumption tools, with 78%, compared to 30.3% for automated testing, the second-highest. Respondents with 6+ years of API development experience were also more likely to report using coding/programming tools to consume APIs than those with 0–5 years of experience.

Coding / programming: 78%
Automated testing: 30.3%
API debugging / manual testing: 26.7%
API documentation: 24.8%
Code review: 23.3%
API developer portals: 19.5%
API designers / API mocking: 18.4%
API monitoring: 15.5%
API management: 15.2%
API performance / load-testing: 13.6%
Security testing: 10.4%
API development collaboration: 10.4%
Data visualization: 9.5%
API publishing: 7.1%

Multiple responses allowed

Download icon

DevOps tooling

DevOps practitioners rely on a number of tools, with Jenkins standing out and leading the way at 41.3%. AWS DevOps (30.7%) and Azure DevOps (26.1%) registered second and third, respectively. GitHub Actions, GitLab Pipelines, and Bitbucket Pipelines round out the remaining tools that received responses totaling 15% or higher.

API-first leaders and respondents with 6+ years of API development experience were also more likely to mention Jenkins.

Jenkins: 41.3%
AWS DevOps: 30.7%
Azure DevOps: 26.1%
GitHub Actions: 25.9%
GitLab Pipelines: 19.6%
Bitbucket Pipelines: 15.7%
CircleCI: 9.6%
Google CI/CD: 8.9%
Hashicorp: 7.3%
TeamCity: 6.6%

Multiple responses allowed

Download icon

Deploying APIs

Respondents deploying APIs reported using a number of different approaches. CI/CD pipelines were the most popular, at 53.4%, followed by deploying APIs in the cloud.

API-first leaders were more likely to use all of the approaches offered when deploying APIs than others; similarly, respondents with 6+ years of API development experience were more likely to use all of the approaches offered when deploying APIs than those with 0–5 years of experience.

CI/CD pipelines: 53.4%
In the cloud: 44.4%
Frameworks: 35.3%
Custom-built: 33.6%
On-premise: 25.7%
Serverless: 23.2%
Gateways: 21.3%
Code generation: 19.1%

Multiple responses allowed

Download icon