Research Foundation
VEMO is built on peer-reviewed research from 8 major industry studies
The Velocity-Complexity Divergence
AI tools boost initial velocity while creating downstream quality, maintenance, and cost issues that traditional metrics fail to capture. Teams feel productive but deliver slower.
1. GitClear Study (2024): The Code Churn Crisis
"Coding on Copilot: 2023 Data Suggests Downward Pressure on Code Quality"
Key Findings:
- • 153M lines of code analyzed (2020-2023)
- • 41% increase in code churn with AI
- • Churn projected to double by 2024
- • Logic errors increase while syntax errors decrease
VEMO Impact:
Foundation for Code Churn Cost (CCC) formula. 41% more code gets rewritten/deleted within 2 weeks, representing massive wasted developer time.
2. DORA State of DevOps Report 2024: The AI Paradox
DevOps Research and Assessment (Google Cloud)
Key Findings:
- • 75% of professionals use AI daily
- • +2.1% productivity reported
- • -1.5% delivery throughput
- • -7.2% delivery stability
- • Root cause: Batch size increases
VEMO Impact:
Foundation for Rework Cost (RWC) and Quality Incident Cost (QIC). Larger changesets = more testing, integration issues, and defects.
3. Qodo Research (2025): The Code Review Crisis
Analysis of code review patterns with AI-generated code
Key Findings:
- • 67% of AI code requires significant modification
- • Code review time increases 15-25%
- • Reviewers must check for AI-specific issues
- • Hallucinations and outdated patterns common
VEMO Impact:
Foundation for Code Review Overhead (CRO) and Rework Cycle Time (RCT). Extra scrutiny and multiple fix cycles add significant effort waste.
4. Harness AI Engineering Report (2025)
Enterprise adoption and impact analysis
Key Findings:
- • 78% of teams using AI coding tools
- • 30-50% increase in context switching
- • Integration overhead often underestimated
- • Tool fragmentation creates workflow friction
VEMO Impact:
Foundation for Context Switching Overhead (CSO). 23 minutes average to regain focus after interruption, multiplied by 4+ extra switches per day.
5. McKinsey: AI and Technical Debt
Enterprise cost analysis of AI adoption
Key Findings:
- • AI accelerates tech debt 2-3x without governance
- • Hidden costs often exceed tool savings
- • Opportunity cost of wasted capacity
- • Long-term maintenance burden increases
VEMO Impact:
Foundation for Technical Debt Cost (TDC) and Opportunity Cost (OPC). Debt compounds at 1.5x annually without governance.
6. Gartner: Technical Debt Impact
Industry benchmarks for tech debt costs
Key Findings:
- • Tech debt costs 20-40% of dev capacity
- • Average organization: $5M+ annual impact
- • Maintenance burden grows exponentially
VEMO Impact:
Benchmark for Technical Debt Cost (TDC) baseline calculations and future cost factor multipliers.
What Research Shows
What AI Tools Deliver:
- • 55% faster code writing (GitClear)
- • 2.1% productivity boost (DORA)
- • 7.5% better documentation (DORA)
- • Reduced syntax errors
Hidden Costs Created:
- • 41% increase in code churn (GitClear)
- • 1.5% slower delivery (DORA)
- • 7.2% reduced stability (DORA)
- • 67% of AI code needs rework (Qodo)
- • 2-3x faster tech debt accumulation (McKinsey)
- • 30-50% more context switching (Harness)