API testing is widely used for integration and functional checks, while contract testing remains less common. Postman survey results from 2021 through 2025 also show how performance, security, workflow, and load testing fit into teams’ API quality practices.
Contents
- The latest API testing adoption figures
- How API test coverage is distributed
- What changed between 2024 and 2025
- API testing maturity and advanced-practice gaps
- API change failure rates and recovery
- The 2023 practice mix
- Longer-term patterns from 2021 and 2022
The latest API testing adoption figures
Postman’s 2025 survey measured the testing practices reported by respondents. Integration testing and functional testing were tied as the most frequently selected practices, each reported by 67% of respondents. These figures describe respondents to the survey, not every API team worldwide.
Performance testing was reported by 57% of respondents in the 2025 survey. Acceptance testing followed at 44%, while security testing was selected by 42%. The results indicate that teams commonly combine behavior-focused testing with checks for throughput and security, although the survey does not show how often each practice was performed or how extensively it was automated.
The remaining 2025 results show a wider gap between common and specialized practices:
- Load testing: 35% of respondents.
- Workflow testing: 32%.
- Contract testing: 17%.
- None of the listed API testing practices: 5%.
Postman also reported that 75% of respondents used CI/CD pipelines for API deployment in 2025. This is an automation-related delivery measure rather than a direct measure of test coverage. A CI/CD pipeline can provide a place to run API tests, but the survey figure alone does not establish which tests ran in those pipelines or whether they blocked releases.
All 2025 figures in this section come from 2025 State of the API Report | Postman. The practice question allowed multiple selections, so the percentages are not mutually exclusive and should not be added together.
How API test coverage is distributed
The 2025 data places integration and functional testing at the center of API quality work. Both reached 67%, 10 percentage points ahead of performance testing at 57%. Acceptance and security testing occupied a middle tier, at 44% and 42%, respectively.
| API testing practice | Respondents in 2025 |
|---|---|
| Integration testing | 67% |
| Functional testing | 67% |
| Performance testing | 57% |
| Acceptance testing | 44% |
| Security testing | 42% |
| Load testing | 35% |
| Workflow testing | 32% |
| Contract testing | 17% |
| None listed | 5% |
The lower reported use of contract testing is notable because contract tests address compatibility expectations between API providers and consumers. However, the available statistic only records whether respondents selected the practice. It does not measure the quality of a contract, the number of services covered, or the consequences of omitting contract tests.
The distinction between performance testing and load testing also matters. Postman’s survey lists them separately: 57% reported performance testing and 35% reported load testing in 2025. These should be treated as separate survey categories, not as interchangeable labels or as percentages that can be combined.
What changed between 2024 and 2025
The 2025 survey reports higher percentages than the 2024 survey for every directly comparable practice supplied here, but the figures come from different report years and should not be interpreted as a controlled year-over-year experiment. Survey composition and question wording may affect comparisons.
Integration testing rose from 61% in 2024 to 67% in 2025, and functional testing rose from 61% to 67%. Performance testing increased from 45% to 57%. Acceptance testing moved from 34% to 44%, while security testing moved from 37% to 42%.
The supplied 2024 practice list also includes categories that are not all listed in the supplied 2025 figures. In 2024, 37% reported end-to-end testing, 34% reported CI/CD testing, and 29% reported regression testing. Workflow testing was reported by 29%, load testing by 29%, and contract testing by 13%.
For the categories present in both supplied years, the comparison is:
| Practice | 2024 | 2025 |
|---|---|---|
| Integration testing | 61% | 67% |
| Functional testing | 61% | 67% |
| Performance testing | 45% | 57% |
| Security testing | 37% | 42% |
| Acceptance testing | 34% | 44% |
| Workflow testing | 29% | 32% |
| Load testing | 29% | 35% |
| Contract testing | 13% | 17% |
The source for the 2024 values is the 2024 State of the API Report. These percentages describe respondents’ selections and are not a global census of API teams.
API testing maturity and advanced-practice gaps
The 2024 results show a familiar concentration around core test types. Integration and functional testing each reached 61%, the highest values in that survey’s supplied practice list. Performance testing reached 45%, followed by security and end-to-end testing at 37% each.
Acceptance testing and CI/CD testing were each reported by 34% of respondents. Workflow, load, and regression testing each reached 29%. Contract testing was reported by 13%, the lowest named practice in the 2024 list.
This distribution suggests that adoption is uneven across test purposes. The figures support a description of relative reporting frequency, but they do not prove that a team with more selected practices has higher maturity. The survey does not provide, in these statistics, a common scoring model for depth, coverage, test maintenance, defect detection, or release governance.
The 2024 report also found that API-first leaders were 83% more likely than others to say they could restore a failed API in under an hour. This is a comparative finding about reported recovery capability. It is not a measured restoration-time average, and it does not establish that any particular testing practice caused the difference.
API change failure rates and recovery
Postman’s 2024 survey asked respondents about the share of their API changes that failed. The reported bands were distributed as follows:
- Less than 5% of API changes failed: 56% of respondents.
- Between 5% and 10% failed: 25%.
- Between 11% and 25% failed: 13%.
- More than 25% failed: 5%.
These are respondents’ reported API-change failure-rate ranges. They are not a single overall failure rate, and the ranges should not be converted into an average without the underlying respondent-level data and a defined treatment for the open-ended categories.
The results also show why recovery capability belongs beside prevention practices in API testing discussions. The same 2024 report found that API-first leaders were 83% more likely than others to report restoring a failed API in under an hour. That statistic describes relative likelihood between groups, not the percentage of all respondents who restored an API within an hour.
Together, the findings separate two operational questions: how often respondents said API changes failed, and how quickly one comparison group said it could restore a failed API. Neither result alone measures end-user impact, incident duration, or the effectiveness of a particular test suite.
The 2023 practice mix
Postman’s 2023 survey reported functional testing as the most frequently selected API testing practice at 66.6%, followed by integration testing at 64.1%. Performance testing reached 51.1%, and security testing reached 45.9%.
Acceptance testing was reported by 40.7% of respondents. Load testing reached 34.4%, while workflow testing reached 32.5%. Contract testing was selected by 16.1%.
The survey also included respondents who said their teams did not test their APIs: 4.2%. Another 0.8% selected other API testing practices. Because respondents could select multiple practices, these percentages are not mutually exclusive. The no-testing response is a separate answer category and should not be treated as the complement of the listed practice percentages.
The detailed 2023 practice figures are available in API testing | State of the API 2023 | Postman API Network. The values retain their published precision, including one decimal place, rather than being presented as rounded whole-number estimates.
Longer-term patterns from 2021 and 2022
The 2022 survey reported functional API testing at 68% and integration API testing at 67%. Performance testing reached 54%, security testing 46%, and acceptance testing 42%. Load testing was reported by 36%, workflow testing by 35%, and contract testing by 17%. Four percent of respondents said their teams did not test their APIs.
In the 2021 survey, functional testing was also reported by 68% of respondents, while integration testing reached 66%. Performance testing stood at 52% and security testing at 45%. Acceptance testing reached 41%, load testing 34%, and workflow testing 32%. Contract testing was reported by 14%, and 4% said their teams did not test their APIs.
| Practice | 2021 | 2022 |
|---|---|---|
| Functional testing | 68% | 68% |
| Integration testing | 66% | 67% |
| Performance testing | 52% | 54% |
| Security testing | 45% | 46% |
| Acceptance testing | 41% | 42% |
| Load testing | 34% | 36% |
| Workflow testing | 32% | 35% |
| Contract testing | 14% | 17% |
| Teams reporting no API testing | 4% | 4% |
The 2022 figures are from the 2022 State of the API Report, and the 2021 figures are from Executing on APIs. These historical report years are retained as published; they should not be relabeled as 2026 measurements or treated as independently reverified legacy data.
Across the supplied 2021–2025 results, functional and integration testing remain the most frequently reported categories. Contract testing is consistently lower than the other named practices, ranging from 13% in the 2024 survey to 17% in the 2022 and 2025 survey figures supplied here. The data supports that pattern in respondent selections, while leaving the depth, frequency, and effectiveness of each practice unspecified.