Microsoft — Engineering Performance
Avg. perf / dev / mo (ETV)
+21.2%
2.51 → 3.04
Active contributors
−12.9%
171.0 → 149.0
Growth
−0.3pp
43.7% → 43.4%
Fixes
−6.4pp
22.1% → 15.7%
Microsoft vs. 500 OSS Performance Index
Per-engineer ETV for Microsoft plotted against the pooled 500 OSS Performance Index. Both series are 90-day trailing rolling averages scaled to a 30-day month, so the curves sit on the same scale (ETV / dev / mo) and can be compared point-for-point. Latest reading: Microsoft is 55% above the index (3.04 vs 1.97 ETV/dev/mo). Baseline gap was 29% above.
Monthly reports
Highlights
- Enhanced *AgentHost* and *Sessions* features with new *browser-style chat tab keybindings* [e7bf9e83 · Sandeep Somavarapu], improved *Agent Host Copilot* integration [4eb9bf2b · Anthony Kim], and accurate *languageModelToolInvoked telemetry* [f18c6b4f · Tyler James Leonhardt].
- Introduced new capabilities for *session management*, including persistence of renamed chat tabs [ece376a4 · Sandeep Somavarapu], a dedicated 'Session Files' section in the Changes view [63790479 · Benjamin Christopher Simmonds], automatic unpinning of sessions in groups [c567c6d0 · Benjamin Christopher Simmonds], and a 'terminals' pill in the session header [35f6ba07 · Benjamin Christopher Simmonds].
- Improved *Playwright* and browser platform stability with a fix for *CLI daemon browser garbage collection* [07b5acd9 · Pavel Feldman], better *error handling* for page load timeouts [1f01c15f · Kyle Cutler], enhanced *browser compatibility* with a `crypto` stub [e8439069 · Dmitry Gozman], and support for `AbortSignal` to *cancel pending Playwright operations* [eb9eb488 · Simon Knott].
- Delivered significant *UI/UX refinements*, including a bug fix for *image carousel attachments* and *chat state persistence* [8c3bcf84 · Justin Chen], refined *browser permission request handling* [6ca57e42 · Kyle Cutler], immediate display of *open-PR icons* in the sessions list [98ea0fe1 · Benjamin Christopher Simmonds], and improved *sash handle visibility* [9f395453 · mrleemurray], [b307393c · mrleemurray].
- Modernized *pnpm configuration* across *server* [f29c68db · Craig Macomber (Microsoft)] and *tools* packages [4c5c9a12 · Craig Macomber (Microsoft)], and introduced a new *Azure DevOps pipeline* for *Docker image publishing* [9e613269 · Dmitry Gozman], enhancing developer experience and CI/CD processes.
Observations
- Commit volume decreased by 27% (1618 this month vs 2211 5-month average), indicating a notable reduction in overall development activity compared to previous periods.
- The maintenance score increased by 21% (177 this month vs 146 5-month average), suggesting a slightly elevated focus on system upkeep and refactoring, though not a significant change (>30%).
- A consistent and high volume of development activity, including new features, enhancements, and bug fixes, centered around the *AgentHost* and *Sessions* features, indicating ongoing investment and refinement in AI agent integration and multi-chat session management.
- There was a recurring pattern of commits addressing *UI layout*, *floating panels*, and *sash handle* styling issues, with multiple commits ([6d8b7045 · mrleemurray], [c1d3c121 · Lee Murray], [9f395453 · mrleemurray], [b307393c · mrleemurray], [e601f5e2 · mrleemurray], [534d1e04 · mrleemurray], [8914f64d · mrleemurray], [b30acd1b · mrleemurray]) suggesting a complexity hotspot in achieving consistent visual presentation across dynamic UI components. Several of these were classified as waste, indicating rework.
- Several commits classified as 'waste' (e.g., [07b5acd9 · Pavel Feldman], [1f01c15f · Kyle Cutler], [75758121 · Ladislau Szomoru], [e58bd4bd · Danyal Ahmed], [b168fa7e · Benjamin Christopher Simmonds], [9a8648f0 · mrleemurray], [321f6bcd · Benjamin Christopher Simmonds], [8914f64d · mrleemurray], [b30acd1b · mrleemurray]) were in fact critical bug fixes or usability improvements, highlighting that the 'waste' metric can sometimes capture necessary corrective work rather than purely inefficient activity.
Repositories
Active repositories ranked by average performance per developer per month (over the last 90 days). The chart shows monthly performance composition — each repo as a stacked layer, with the top of the stack representing total org performance per month. Top 9 repos shown; the remainder is aggregated as “Other”.
| Repository | ||||
|---|---|---|---|---|
| vscode | 73 | 1005 | 4.6 | +310%since Q2 2025 |
| playwright | 8 | 99 | 4.1 | +52%since Q2 2025 |
| Agents-for-net | 3 | 32 | 3.6 | −3%since Q2 2025 |
| fluentui | 14 | 78 | 1.8 | −61%since Q2 2025 |
| PowerToys | 10 | 48 | 1.6 | +224%since Q2 2025 |
| FluidFramework | 27 | 68 | 0.8 | −29%since Q2 2025 |
| semantic-kernel | 6 | 14 | 0.8 | −72%since Q2 2025 |
| terminal | 3 | 6 | 0.7 | +72%since Q2 2025 |
| DeepSpeed | 5 | 10 | 0.7 | +27%since Q2 2025 |
| TypeScript | 2 | 0 | 0.1 | −22%since Q2 2025 |
| markitdown | 1 | 0 | 0.0 | −65%since Q2 2025 |
Company total11 repositories | 149unique devs | 1360ETV total | 3.04ETV / dev / mo | +115%since Q2 2025 |
| Performance (ETV) is the sum of every repository above. Active devs at the company level counts unique contributors across all repos, so a contributor working in multiple repos is counted once here but appears in each repo's row (the per-repo column will sum higher). ETV / dev / mo = Company ETV ÷ unique devs ÷ 3 mo. The "Since start" column compares each repo's Q1 2026 quarterly performance to the first quarter it had any activity — for repos that existed in Q2 2025 (when this index began), that's Q2 2025; for younger repos it's the quarter they actually started. The company row uses Q2 2025 as the baseline since the index itself began then. | ||||
Performance Growth vs Active Contributors
Shows how engineering performance scales relative to team growth. Left axis shows total performance score, right axis shows active contributor count. The gap between curves represents productivity gains — more delivered per person, not just more people. Unit: Engineering Throughput Value (ETV).
Cost per Performance Unit
−63%
If performance per engineer more than doubled, each unit of engineering performance now costs approximately 63% less than at the baseline 90-day window (ending 2025-06-29). This is a directional estimate — the exact figure depends on fully-loaded engineer cost, but the direction is unambiguous.
Effective Capacity Added
+251 engineers
At today's productivity, the current 149-person team delivers the performance equivalent of 400 engineers at the baseline 90-day rolling window (ending 2025-06-29). That's roughly 251 engineers worth of capacity added through productivity gains, not hiring.
Performance Composition
Stacked bars show total complexity performance split into Growth (new value), Maintenance (sustaining systems), and Fixes (rework). The yellow line overlays performance per contributor — rising line means each engineer is delivering more, regardless of team size changes. Unit: Engineering Throughput Value (ETV).
CapEx vs OpEx
Monthly CapEx vs OpEx split. CapEx (capitalizable investment) is Growth — new features and capabilities. OpEx (operating expense) is Maintenance plus Fixes — keeping the lights on and reworking what's already shipped. The yellow line is the CapEx share, a quick read on how much of the month went into building new vs sustaining existing. Unit: Engineering Throughput Value (ETV).
Hours per Repository
Trailing 90-day window (64 working days). Org-level capacity is allocated to each repo by its share of org performance, then split CapEx / OpEx by that repo's own Growth vs Maintenance + Fixes mix.
| terminal | 68.4% | 31.6% |
| vscode | 45.6% | 54.4% |
| fluentui | 43.3% | 56.7% |
| PowerToys | 40.6% | 59.4% |
| playwright | 39.1% | 60.9% |
| DeepSpeed | 33.2% | 66.8% |
| Agents-for-net | 32.7% | 67.3% |
| FluidFramework | 28.5% | 71.5% |
| semantic-kernel | 24.7% | 75.3% |
| TypeScript | 7.3% | 92.7% |
| markitdown | 0.0% | 0.0% |
| Total | 43.4% | 56.6% |
Fix Burden Distribution
Monthly rework volume broken down by who did it. Top contributors carry their named slice; everyone else is rolled into Others. Use this to spot whether fix work is concentrated on a small group (bus-factor risk) or distributed across the team.
Quarterly Summary
The raw numbers behind the charts: commits analyzed, active contributors, total performance, performance per developer, and the Growth / Maintenance / Fixes split for each quarter.
| Quarter | ||||||||
|---|---|---|---|---|---|---|---|---|
| Q2'25 | 4,641 | 160 | 556.91 | 1.2 | 37.7% | 47.1% | 15.2% | — |
| Q3'25 | 5,279 | 166 | 552.86 | 1.1 | 40.3% | 43.1% | 16.7% | −1% |
| Q4'25 | 6,137 | 164 | 595.17 | 1.2 | 33.2% | 44.8% | 22.1% | +8% |
| Q1'26 | 9,278 | 169 | 1,197.49 | 2.4 | 44.2% | 32.4% | 23.4% | +101% |
Top Contributors
Contributors ranked by performance per month (Growth + Maintenance + Fixes), over the last 90 days normalized to a 30-day calendar month.
The best way to measure AI efficiency
SampleA preview of the Navigara engine running on a sample organization. The numbers below are illustrative, not part of the OSS500 benchmark above.
Measure
Score every commit by depth
GitHub commits are weighted by what it took to write them, not by lines of code. The result is ETV per developer per month.
SourceGitHub
Spend
Tie ETV to cost
AI token bills and seat costs are pulled per team and divided by the ETV produced. The result is your true cost per unit of work.
SourceToken usage + finance
Map
Tie work to objectives
Each ETV is mapped to your Jira epics and labels, so you can see what's key-aligned, aligned, or unmapped capacity.
SourceJira
Performance
ETV delivered per developer / month
9.4ETV / dev / month
5.6 below target- Non-AI 5.8
- AI 3.6
AI Efficiency
AI spend per ETV unit delivered
$4.20
$0.60 over target- Cost / ETV $4.20
Objective Alignment
Share of work mapped to key objectives
51%
24 pts below target- Key-aligned 28%
- Aligned 23%