PowerShell merges 6x faster than C in 2025. 25 languages ranked by cycle time.
Speed Gap
PowerShell (4h) vs C (24h)
Based on 802,979 merged PRs | GitHub Archive / BigQuery | October 2025
All 25 programming languages ranked by median PR merge time. The fastest languages merge in hours; the slowest take nearly a full day.
| Rank | Language | Median (h) | Avg (h) | PRs Analyzed | Avg PR Size |
|---|---|---|---|---|---|
| #1 | PowerShell | 4h | 33.2h | 1,292 | 234 lines |
| #2 | Dockerfile | 11h | 51.6h | 1,174 | 110 lines |
| #3 | Shell | 14h | 64.4h | 5,230 | 148 lines |
| #4 | Nix | 14h | 61.1h | 1,366 | 77 lines |
| #5 | TypeScript | 15h | 59.5h | 43,177 | 503 lines |
| #6 | JavaScript | 15h | 61.6h | 16,904 | 461 lines |
| #7 | Ruby | 15h | 66.8h | 3,534 | 174 lines |
| #8 | HCL | 15h | 56.5h | 1,492 | 133 lines |
| #9 | YAML | 15h | 74.7h | 1,416 | 29 lines |
| #10 | C# | 16h | 62.2h | 6,762 | 463 lines |
"PowerShell PRs merge 6x faster than C PRs: 4 hours vs 24 hours median."
At the other end of the spectrum, C takes 24 hours median to merge. Swift (23h), C++ (21h), and Java/PHP (20h) also sit in the slow lane.
13,726 PRs analyzed
8,374 PRs analyzed
4,478 PRs analyzed
1,759 PRs analyzed
1,105 PRs analyzed
14,940 PRs analyzed
These languages aren't slow because they're bad—they're slow because they're used in domains that demand careful review. Systems programming, safety-critical applications, and legacy codebases require more scrutiny.
The 6x speed gap between PowerShell and C isn't about language syntax. It's about ecosystems, use cases, and review culture.
TypeScript leads with 43K+ PRs. Larger communities mean more active reviewers and faster turnaround. Smaller communities (Swift: 1.1K) have fewer available reviewers.
Dart PRs average 844 lines; Shell PRs average 148 lines. Infrastructure code (PowerShell, Dockerfile, YAML) tends to be smaller and more straightforward.
C/C++ dominates embedded systems and kernel development where bugs can be catastrophic. More thorough review is a feature, not a bug.
DevOps languages (PowerShell, HCL, Dockerfile) have strong CI/CD integration. Automated tests provide confidence, speeding up human review.
"TypeScript and JavaScript tie at 15 hours median merge time—the web stack moves fast."
Larger communities don't automatically mean faster reviews. TypeScript has 43K PRs and 15h median; C has 2.8K PRs and 24h median. The relationship is more nuanced.
9 languages
PowerShell, Dockerfile, Shell, Nix, TypeScript...
11 languages
C#, HTML, Python, Vue, Scala, CSS, Go, Rust, Kotlin, Jupyter Notebook, Dart
5 languages
Java, PHP, C++, Swift, C
Comparing to our 2024 study (3.4M PRs), the hierarchy is largely stable. PowerShell remains the fastest, C remains the slowest. But the gap has widened from 4x to 6x.
From 4x (2024) to 6x (2025). Fast languages got faster; slow languages stayed slow.
PowerShell dropped from 6h to 4h median—a 33% improvement in cycle time.
Correlation, not causation
Languages don't determine review speed—teams, domains, and cultures do. C is slow because it's used in cautious environments, not because of syntax.
Public repos only
GitHub Archive captures public repositories. Private enterprise repos may show different patterns—especially for languages like C# that dominate in enterprise.
Sample size varies
TypeScript has 43K PRs; Swift has 1.1K. Languages with smaller samples may be less representative of true median values.
Language matters less than team practices.
A 6x speed difference sounds dramatic, but it reflects ecosystem cultures more than technical constraints. Your TypeScript team can be slower than a well-organized C team. The real question isn't "what language should we use?" but "what review culture are we building?"
The takeaway: Don't blame your language. Benchmark against peers in your domain, set appropriate SLAs, and focus on the practices that actually move the needle—PR size, reviewer availability, and automated testing.
This analysis is based on 802,979 merged pull requests from GitHub Archive / BigQuery during October 2025. Language is determined by GitHub's primary language detection for each repository. Merge time is calculated from PR creation to merge event. Languages with fewer than 1,000 PRs were excluded for statistical reliability. For full methodology, see the complete study.
CodePulse shows you merge times by language, repository, and developer.