2024_final_dora_report

DORA · pdf · 2024-10-25 Reader 🖥️

AI as Amplifier Burnout & Recovery Developer Experience & Flow DORA AI Capabilities Model (Team Profiles & Capabilities)DORA & Engineering Metrics Enterprise AI Adoption & Readiness Golden Path Measuring AI Impact Platform Engineering

Technology investments — generative AI adoption and internal developer platforms alike — reliably raise individual, team, and organizational performance only when paired with the underlying sociotechnical fundamentals (small-batch delivery discipline, developer independence, transformational leadership, a genuine user-centric and documentation culture, and stable priorities); absent those fundamentals, the same investments degrade software delivery throughput and stability and can increase burnout, so AI and platforms amplify whatever organizational system already exists rather than fixing it.

practical · mixed — A decade-ten, 39,000-respondent empirical survey report (expository, statistical) that is explicitly in service of practical organizational action: it closes with a recommended continuous-improvement method and named interventions (stabilize priorities, be user-centric, become data-informed, commit to cloud or stay out), so the findings are presented as inputs to 'what should a team or org do.'

Structure

Part	Role
Executive summary and key findings	context
Software delivery performance chapter — splits the historic 'four keys' into two statistically distinct factors, throughput and stability, and introduces 'rework rate' as a second stability metric	evidence / redefinition of measurement
Artificial intelligence: adoption and attitudes — survey data on who uses AI, for what tasks, and how much they trust it	evidence
Exploring the downstream impact of AI — AI's effects on process measures (doc/code quality, review speed), on software delivery throughput/stability, and on organizational/team/product performance	evidence / core finding
Platform engineering — productivity and performance gains from internal developer platforms, developer independence, and the throughput/stability trade-off	evidence / parallel case
Developer experience — user-centricity, documentation culture, and the burnout effects of unstable priorities	evidence
Leading transformations — transformational leadership, user-centricity, data-informed decision making, and cloud commitment as the four practical levers	conclusion / recommendation

Key terms

Software delivery throughput
A statistically distinct factor composed of change lead time, deployment frequency, and failed deployment recovery time — the speed component, newly separated from stability this year because the data showed independence between the two.
Software delivery stability
A factor composed of change failure rate plus a newly added 'rework rate' metric (unplanned deployments to fix a user-facing bug) — DORA's hypothesis that change failure rate functions as a proxy for rework.
Developer independence
Existing users of an internal platform being able to perform their tasks for the entire application lifecycle without relying on an enabling team.
Golden paths
Highly automated, self-service workflows a platform's users follow when provisioning or operating application infrastructure, designed to abstract away delivery complexity so the developer only worries about code.
Transformational leadership
A leadership model scored across five specific dimensions — vision, inspirational communication, intellectual stimulation, supportive leadership, personal recognition — used as a validated survey construct, not a vague honorific.

Propositions

explicitdevelopedAI adoption is now broad and fairly uniform across industries (75.9% of respondents rely on it for at least one daily task), and it measurably improves individual-level outcomes (flow, productivity, job satisfaction) and process measures (code quality, internal documentation quality, code review speed, approval speed) and team/organizational performance.conf: high
extends ←The Pragmatic Engineer 2025 Survey: What's in your tech stack? · p1 — DORA's 2024 report found 75.9% daily-task AI adoption; this survey a little over a year later finds 85% report using at least one AI tool — consistent continued-adoption trajectory from a different measurement method and population.
explicit⚡ disputed & corroboratedDespite those process-level and individual-level gains, a 25% increase in AI adoption is associated with a 1.5% decrease in software delivery throughput and, more sharply, a 7.2% decrease in software delivery stability.conf: high
supportsAI-Assisted Software Development: Team Profiles and Capabilities for Putting Research into Action · p5 — This 2024 finding (AI adoption reduces delivery stability) is the earlier-year evidence that DORA's 2025 AI Capabilities Model report corroborates with its own instability finding.
supports ←How AI will change software engineering · p2 — Restates the same 2024 DORA finding (1.5% throughput / 7.2% stability decline per 25% AI-adoption increase) as independent corroborating evidence in a different outlet.
contradicts ←Is Agile Dead? How AI and Extreme Agile Redefine Work · p1 — This article's claimed 30-100x productivity multiplier from AI-augmented Agile is an order of magnitude beyond — and in direction opposite to — DORA's 2024 measured finding of a 1.5% throughput decrease per 25% AI-adoption increase.
implicitcorroboratedDORA hypothesizes the delivery-performance harm occurs because AI lets developers produce more code per unit time, which likely grows changelist (batch) size — and DORA's decade of research has consistently shown larger changes are slower and more prone to instability; improving upstream process measures does not automatically improve delivery unless small-batch discipline and robust testing are maintained.conf: medium
supportsAI-Assisted Software Development: Team Profiles and Capabilities for Putting Research into Action · p7 — Both reports converge on the same small-batch-size mechanism for why AI pressures delivery stability — the 2024 report states it as a fresh hypothesis, the 2025 report restates it as an established pressure point.
supports ←Technology Radar: An opinionated guide to today's technology landscape (Volume 30) · p3 — This April 2024 practitioner observation (AI throughput -> larger PRs -> strained review) predates and independently corroborates the same mechanism DORA's October 2024 report could only infer from survey data.
explicitAI adoption benefits organizational performance (+2.3% per 25% increase) and team performance (+1.4%) but shows no clear association with product performance, because team/org performance depend on communication, knowledge-sharing and decision-making bottlenecks that AI eases, while product success depends more on taste, empathy, and user-experience judgment that AI does not yet substitute for.conf: high
explicitInternal developer platforms raise individual productivity (+8%), team performance (+10%), and organizational performance (+6%), but reduce throughput (-8%) and change stability (-14%) relative to not using a platform, and platform-driven instability combined with the platform itself is specifically linked to higher burnout.conf: high
explicitdevelopedDeveloper independence (completing the full application lifecycle without an enabling team) is the platform-engineering lever with the clearest productivity payoff — a 5% gain at both individual and team level — larger in per-capita terms than having a dedicated platform team, which produced a 6% team-level gain but a negligible individual-level one.conf: high
supportsPresentation: Platform Engineering in the Age of AI · p2 — Both sources treat platform-engineering fundamentals (here, developer independence; there, service intent/APIs) as unchanged in kind by AI — AI shifts magnitude of effects, not the underlying discipline.
extends ←Architecture Modernization: Aligning software, strategy, and structure • Nick Tune • CodeCrafts 2024 · p11 — DORA 2024 finds developer independence without an enabling team to be the stronger steady-state productivity lever; Tune's enabling team is explicitly scoped as a temporary transition structure for the modernization push itself, not a permanent capability dependency — a scope qualification on when enabling teams help versus hurt.
explicitRequiring teams to use a platform exclusively for the entire application lifecycle (removing the option to break out of it) costs throughput (-6%) — mandated, inflexible platform adoption undercuts the independence that otherwise drives the platform's benefits.conf: high
explicitTransformational leadership (vision, inspirational communication, intellectual stimulation, supportive leadership, personal recognition) raises productivity (+9% per 25% increase), job satisfaction, team/product/organizational performance, and reduces burnout, but functions only as an enabler of technical and product practices, not as a standalone driver of performance — it works through delegating authority, supplying metrics/business intelligence, and aligning incentives to value delivery over feature delivery.conf: high
explicitUnstable organizational priorities produce small productivity decreases but substantial burnout increases, and this burnout effect is resistant to otherwise-protective factors — strong leadership, good internal documentation, and a user-centered approach do not counteract it; only stabilizing priorities (and shielding employees' day-to-day from priority churn) helps, and the chronic-uncertainty / loss-of-control mechanism behind this mirrors general chronic-stress research.conf: high
explicitQuality internal documentation has no meaningful effect on predicted product performance by itself, but amplifies the positive effect of user-centricity on product performance — documentation is the channel through which user signals propagate across a team into the product.conf: high

Arguments

AI's harm to software delivery stability stems from an erosion of small-batch delivery discipline, not from AI-generated code being inherently worse. strength: medium

Explicit premises:

AI improves upstream process measures historically correlated with better delivery performance (doc quality, code quality, review speed, approval speed, reduced complexity).
Despite those improvements, AI adoption is associated with worse delivery throughput and especially worse delivery stability.
DORA's prior decade of research shows larger changesets are slower and more unstable.

Implicit premises:

AI's productivity gains manifest partly as developers producing and committing more code per unit time.
Teams are not compensating for this increased code volume by deliberately shrinking batch size.
No other confound (e.g., riskier work being attempted because AI makes it feel feasible) fully explains the gap.

Inferences:

Without explicit discipline to keep changesets small, raw AI-driven throughput gains convert into integration and stability costs rather than net delivery improvement.
The fix is organizational/process (batch-size and testing discipline), not a wait for better models.

Platform engineering's net value depends on preserving user-centeredness and developer independence rather than mandating platform use, because the same mechanism that creates value (self-service) also explains why forced, non-independent use erodes it. strength: medium