AI Development ROI: The Real Economics of AI-Powered Software Development in 2025

Every dollar spent on generative AI yields $3.71 in return. In financial services, that number jumps to 4.2x ¹. Yet despite these impressive figures, most companies report modest cost savings under 10% and revenue gains below 5% ². This paradox lies at the heart of understanding AI development economics—the returns are real, but they're not always where you expect them, which is why platforms like RepoBird.ai focus on delivering measurable, practical value.

As we navigate 2025, the conversation around AI in software development has shifted from "if" to "how much." CFOs are demanding hard numbers, development teams are pushing for better tools, and executives are caught between AI hype and budget reality. The truth, as always, lies in the data—and the data tells a fascinating story about how AI is fundamentally reshaping the economics of building software.

Whether you're a startup founder calculating runway, an enterprise architect justifying tool investments, or a developer curious about your market value, understanding the real economics of AI development has never been more critical. This comprehensive analysis breaks down the numbers, reveals the hidden costs, and provides frameworks for calculating your own AI ROI.

Understanding AI Development ROI: Beyond Simple Metrics

The New ROI Equation

Traditional software development ROI focused on straightforward metrics: time to market, bug rates, and developer productivity. AI development introduces a more nuanced equation that captures both direct and indirect value creation.

The modern AI ROI formula considers multiple dimensions of value. Direct cost savings come from reduced developer hours and faster deployment cycles. Productivity gains manifest as increased feature velocity and improved code quality. But the real transformation happens in the indirect benefits: enhanced innovation capacity, improved developer satisfaction, and competitive differentiation.

Consider how a major e-commerce platform calculated their AI development ROI. They initially focused solely on time savings—their AI tools reduced coding time by 35%. But when they expanded their analysis to include reduced bug rates (down 42%), faster onboarding for new developers (60% reduction), and improved feature experimentation velocity (3x increase), their actual ROI jumped from 150% to over 400%.

Productivity vs Profitability: The Metric Shift

The most significant change in 2025 is how organizations measure AI success. According to recent studies, productivity has overtaken profitability as the primary ROI yardstick ³. This shift reflects a deeper understanding of how AI creates value in software development.

Productivity gains compound over time. When developers spend less time on routine tasks, they don't just save hours—they enter flow states more often, tackle more complex problems, and produce higher-quality solutions. One financial services firm found that while their immediate cost savings were modest, the productivity improvements led to launching new products 40% faster, ultimately driving revenue growth that dwarfed the initial savings.

This productivity-first approach also addresses executive impatience around AI investments. Rather than waiting for long-term profitability metrics, teams can demonstrate immediate productivity improvements that justify continued investment while building toward larger financial returns.

Real-World AI Development Cost Savings

Time Savings That Transform Economics

The numbers from the field paint a compelling picture of AI's impact on development economics. Organizations implementing AI development tools report consistent patterns of time savings across different activities.

Code generation and boilerplate creation show the most dramatic improvements, with 70-80% time reductions being common. A healthcare technology company reduced their API endpoint creation time from 4 hours to 45 minutes per endpoint. Across their 200+ annual endpoints, this single improvement saved over $150,000 in developer costs annually.

Testing and quality assurance represent another major area of savings. Automated test generation reduces test creation time by 60%, while AI-powered code review catches 45% more bugs before human review. One enterprise reported that their bug fix costs dropped from $5,000 per critical bug to $1,500, primarily due to earlier detection and more comprehensive automated fixes.

Documentation, often the most neglected aspect of development, sees remarkable improvements with AI assistance. Teams report 65% time savings on documentation tasks, with the added benefit of more consistent and comprehensive documentation. This improvement has downstream effects on onboarding efficiency and maintenance costs.

Breaking Down Cost Structures

Understanding where AI creates cost savings requires analyzing the full development lifecycle. Traditional development cost structures allocate roughly 40% to initial development, 60% to maintenance and updates. AI fundamentally shifts this equation.

Initial development costs can actually increase slightly as teams learn new tools and establish AI-friendly architectures. However, maintenance costs plummet. AI-assisted codebases tend to be more consistent, better documented, and easier to modify. One study found that maintenance costs for AI-assisted projects were 45% lower over a three-year period.

The hidden cost savings often prove most valuable. Reduced context switching saves an average of 1.5 hours per developer per day. Fewer production incidents mean less emergency overtime. Better code quality reduces technical debt accumulation. When aggregated, these "soft" savings often exceed the direct time savings.

Enterprise vs Startup Economics

The economic impact of AI development varies dramatically by organization size, creating distinct patterns for enterprises and startups.

Large enterprises typically see ROI through scale effects. A 20% productivity improvement across 500 developers creates massive value. One Fortune 500 company calculated that their AI development platform investment of $2.5 million annually was offset by productivity gains equivalent to hiring 100 additional developers—a $15 million value at their fully loaded cost per developer.

Startups experience a different but equally compelling economic story. For them, AI levels the playing field. A five-person startup using AI tools can now deliver products that previously required 20-person teams. This isn't just about cost savings—it's about achieving previously impossible velocity. One YC startup reported shipping their MVP in 6 weeks instead of the planned 6 months, allowing them to secure funding before their runway expired.

Mid-size companies often see the best of both worlds: enough scale for significant absolute savings, but still agile enough to rapidly adopt and optimize AI tools. They typically report the highest ROI percentages, often exceeding 300% within the first year.

Measuring AI Investment Return: A Practical Framework

Key Performance Indicators for AI Development

Measuring AI development ROI requires a comprehensive set of KPIs that capture both immediate and long-term value. The most effective organizations track a balanced scorecard of metrics across multiple dimensions.

Velocity metrics form the foundation. These include feature delivery speed, time from commit to production, and story points completed per sprint. AI-assisted teams typically show 30-50% improvements in velocity metrics within the first quarter of adoption. But velocity alone can be misleading—it must be balanced with quality indicators.

Quality metrics reveal AI's true impact on development economics. Bug density, code review turnaround time, test coverage, and production incident rates all improve with AI assistance. One particularly telling metric is the "defect escape rate"—bugs that make it to production. AI-assisted development typically reduces this rate by 40-60%, preventing costly production fixes.

Developer experience metrics often get overlooked but prove crucial for long-term ROI. Developer satisfaction scores, time spent in flow state, and retention rates all improve with AI tool adoption. Given that replacing a senior developer can cost $50,000-$100,000, even modest improvements in retention deliver substantial ROI.

Cost-Benefit Analysis Models

Building an effective cost-benefit model for AI development requires understanding both immediate costs and long-term value streams. The most successful models incorporate multiple time horizons and value categories.

Start with direct cost inputs: licensing fees, training costs, integration expenses, and any infrastructure upgrades. For a 50-person development team, typical first-year costs range from $50,000 to $250,000 depending on the platform and implementation scope.

Next, model the direct savings: reduced developer hours, decreased bug fix costs, and lower infrastructure needs due to more efficient code. These savings typically offset initial investments within 6-12 months.

The real value emerges in indirect benefits modeling. Faster time-to-market can be valued using competitive analysis—what's the value of beating competitors by three months? Improved developer retention saves recruitment costs. Higher code quality reduces long-term maintenance burden. These indirect benefits often represent 60-70% of total AI development ROI.

ROI Calculation Examples

Let's walk through a detailed ROI calculation for a typical mid-size software company implementing AI development tools.

TechCorp Case Study (100 developers):

Initial Investment:

AI platform licensing: $150,000/year
Training and onboarding: $50,000 (one-time)
Integration and setup: $25,000 (one-time)
Total Year 1 Cost: $225,000

Year 1 Direct Savings:

Developer time saved (25% average): 52,000 hours
Value at $100/hour fully loaded: $5,200,000
Bug fix cost reduction (40%): $800,000
Documentation time saved: $300,000
Total Direct Savings: $6,300,000

Year 1 Indirect Benefits:

Faster feature delivery (2 major features): $2,000,000 revenue
Improved retention (5 developers retained): $250,000 saved
Reduced technical debt: $500,000 future savings
Total Indirect Benefits: $2,750,000

Year 1 ROI: ((6,300,000 + 2,750,000 - 225,000) / 225,000) × 100 = 3,922%

This exceptional ROI reflects the reality many companies experience—when properly implemented, AI development tools deliver returns that dwarf their costs.

Hidden Costs and Considerations

Integration Complexity and Technical Debt

While the benefits of AI development are substantial, organizations must account for hidden costs that can impact overall ROI. Integration complexity tops the list of unexpected expenses.

Legacy systems present particular challenges. One enterprise spent six months and $500,000 adapting their 15-year-old codebase to work effectively with AI tools. The investment paid off—they now save $2 million annually—but the upfront cost surprised executives expecting plug-and-play solutions.

Technical debt can actually increase initially as teams adapt to AI-generated code patterns. The key is distinguishing between transitional debt (which decreases as teams adapt) and structural debt (which requires architectural changes). Smart organizations budget 15-20% of their AI investment for addressing integration challenges.

Training and Change Management

The human side of AI adoption often generates the highest hidden costs. Effective AI development requires more than just tool deployment—it demands fundamental shifts in how developers think and work.

Training costs extend beyond initial workshops. Developers need ongoing education as AI capabilities evolve. Leading organizations budget for quarterly training updates and maintain internal AI champions who share best practices. One company found that developers with comprehensive AI training were 3x more productive than those with basic training—making the investment clearly worthwhile.

Change management proves equally critical. Resistance to AI tools can severely limit ROI. Successful organizations invest in clear communication about how AI enhances rather than threatens developer roles. They celebrate early wins, share success stories, and create psychological safety for experimentation. This cultural work requires time and resources but ultimately determines adoption success.

Ongoing Subscription and Maintenance

The subscription economy of AI development tools creates ongoing cost considerations that impact long-term ROI calculations. Unlike traditional development tools with one-time purchases, AI platforms require continuous investment.

Subscription costs typically scale with usage and team size. While entry-level tiers seem affordable, costs can escalate quickly. One startup found their AI tool costs grew from $500/month to $5,000/month as they scaled from 5 to 50 developers. However, their productivity gains scaled even faster, maintaining positive ROI.

Model maintenance represents another ongoing cost. AI models require periodic retraining, especially for specialized domains. Organizations using custom-trained models budget $50,000-$100,000 annually for model updates. Generic models require less maintenance but may provide less value for specialized use cases.

Platform lock-in risk also factors into long-term economics. Switching AI platforms can be costly, so initial platform selection becomes crucial. Smart organizations maintain some platform independence by using standard interfaces and avoiding proprietary features where possible.

Sector-Specific ROI Analysis

Financial Services: Leading the Charge

Financial services organizations are seeing the highest AI development ROI, with average returns of 4.2x investment ¹. This sector's success stems from several factors that create ideal conditions for AI value creation.

Regulatory compliance drives much of the ROI in financial services. AI tools that automatically ensure code meets compliance standards save enormous amounts of manual review time. One investment bank reduced their compliance review cycle from two weeks to two days, saving $3 million annually just in accelerated deployment.

High-frequency trading systems benefit dramatically from AI-optimized code. Microsecond improvements in execution time translate to millions in additional revenue. AI development tools that optimize performance while maintaining correctness provide immediate, measurable value.

Risk management applications showcase AI's ability to handle complex, interconnected systems. Traditional development of risk models required months of careful coding and validation. AI-assisted development reduces this to weeks while improving model accuracy and documentation quality.

Healthcare Technology: Quality and Compliance

Healthcare technology companies face unique challenges that make AI development particularly valuable. The combination of strict regulatory requirements, life-critical quality needs, and rapid innovation demands creates compelling ROI opportunities.

HIPAA compliance checking automated through AI saves healthcare companies average of $500,000 annually in compliance costs. More importantly, it reduces compliance violations that could result in multi-million dollar fines.

Clinical decision support systems benefit from AI's ability to generate highly reliable, well-tested code. One electronic health record company reduced their bug rate by 75% using AI-assisted development, potentially saving lives while reducing liability.

Interoperability challenges in healthcare create another AI opportunity. AI tools that can understand and implement complex healthcare data standards (HL7, FHIR) reduce integration time by 60%, enabling faster deployment of critical healthcare innovations.

E-commerce and Retail Tech

E-commerce platforms demonstrate how AI development ROI extends beyond cost savings to revenue generation. The ability to rapidly experiment and deploy features directly impacts bottom-line growth.

Personalization engines built with AI assistance deploy 3x faster than traditional development. One major retailer launched 15 personalization experiments in the time previously required for 5, leading to a 23% increase in conversion rates.

Inventory management systems benefit from AI's ability to handle complex, multi-variate optimization problems. AI-assisted development of these systems reduces implementation time by 50% while improving accuracy by 30%.

Mobile commerce applications particularly benefit from AI development. The ability to rapidly iterate on user experience, optimize performance, and maintain consistency across platforms drives measurable revenue improvements. Companies report 40% faster mobile feature deployment with 25% fewer platform-specific bugs.

Enterprise SaaS: Scalability Economics

Enterprise SaaS companies experience unique ROI patterns from AI development, driven by their need for scalable, maintainable codebases that serve diverse customer needs.

Multi-tenant architecture development sees dramatic improvements with AI assistance. Complex isolation requirements, performance optimization, and security considerations that traditionally required senior architect involvement can now be handled by mid-level developers with AI support. This democratization of expertise saves an average of $2 million annually for growing SaaS companies.

Feature flagging and configuration management benefit from AI's ability to handle combinatorial complexity. One SaaS platform reduced configuration-related bugs by 80% while increasing the rate of feature experiments by 5x.

API development and documentation represent another high-ROI area. AI tools that generate consistent, well-documented APIs reduce integration time for customers by 60%, directly impacting customer satisfaction and retention rates.

Break-Even Analysis and Payback Periods

Typical Payback Timelines by Company Size

Understanding when AI investments pay for themselves helps organizations plan budgets and set expectations. Payback periods vary significantly based on company size, implementation approach, and industry sector.

Startups typically achieve the fastest payback periods, often within 2-4 months. Their small size enables rapid adoption, and even modest productivity gains translate to significant relative impact. A 10-person startup saving 20% of development time effectively gains two additional developers worth of capacity.

Small to mid-size companies (10-100 developers) usually see payback within 4-8 months. They benefit from enough scale to justify comprehensive platforms while remaining agile enough for quick adoption. The sweet spot appears to be around 30-50 developers, where payback often occurs within 6 months.

Large enterprises face longer payback periods, typically 8-14 months, due to integration complexity and change management needs. However, their absolute returns dwarf smaller organizations. A 12-month payback on a $2 million investment that generates $10 million in annual value remains highly attractive.

Factors Accelerating ROI

Several factors can dramatically accelerate AI development ROI, turning marginal investments into transformative successes.

Strong executive sponsorship cuts payback periods by 30-40%. When leadership actively champions AI adoption, removes obstacles, and celebrates successes, adoption accelerates and value realization follows.

Choosing the right initial use cases proves crucial. Organizations that start with high-impact, low-risk applications see faster returns. Bug fixing, test generation, and documentation typically provide quick wins that build momentum for broader adoption.

Developer champions make an enormous difference. Teams with enthusiastic early adopters who share knowledge and success stories see 2x faster adoption rates. Investing in identifying and supporting these champions pays dividends.

Technical readiness accelerates ROI. Organizations with modern CI/CD pipelines, good test coverage, and clean architectures see faster AI value realization. Those with technical debt may need to invest in modernization first, extending payback periods but ultimately achieving higher returns.

Long-Term Value Accumulation

The true power of AI development investment emerges over time through compounding effects that create sustained competitive advantages.

Knowledge accumulation represents a key long-term value driver. AI systems learn from every interaction, becoming more valuable over time. Organizations report that their second-year productivity gains often exceed first-year gains as both AI systems and developers optimize their collaboration.

Technical debt reduction compounds annually. AI-assisted development tends to produce cleaner, more maintainable code. Over a five-year period, this can reduce maintenance costs by 60%, freeing resources for innovation.

Developer skill enhancement creates lasting value. Developers working with AI tools report accelerated learning curves, exposure to best practices, and improved problem-solving abilities. This human capital improvement persists even if organizations change tools.

Market positioning effects multiply other benefits. Organizations known for AI adoption attract better talent, win more contracts, and command premium valuations. One startup credited their AI-forward development approach with securing a 40% higher Series B valuation.

Building Your AI Development ROI Case

Framework for Executives

Presenting a compelling AI development ROI case to executives requires balancing hard numbers with strategic vision. The most successful presentations follow a structured approach that addresses both immediate concerns and long-term opportunities.

Start with the burning platform: what competitive pressures or operational challenges make AI adoption urgent? Executives respond to clear problem statements backed by market data. Show how competitors are using AI to gain advantages or how current development velocity threatens market position.

Present a phased investment approach that demonstrates early value. Rather than requesting large upfront investments, propose pilot programs with clear success metrics. A typical phased approach might include: Month 1-3 pilot with 10 developers, Month 4-6 expansion to 50 developers, Month 7-12 full rollout based on proven ROI.

Address risk explicitly. Executives worry about failed technology investments. Acknowledge potential challenges while showing mitigation strategies. Include contingency plans and exit strategies that minimize downside risk while preserving upside potential.

Connect AI investment to strategic objectives. Whether the goal is faster innovation, improved quality, or cost reduction, show how AI development directly supports corporate strategy. Use industry benchmarks and competitor analysis to validate projections.

Quick Wins Strategy

Achieving early victories builds momentum and justifies continued investment. The most successful AI development initiatives focus on quick wins that demonstrate value while building toward larger transformations.

Bug fix automation provides ideal quick wins. Within weeks, teams can show reduced bug resolution time and improved code quality. One company demonstrated 50% faster bug fixes within their first month, securing executive support for broader implementation.

Documentation generation offers visible, immediate value. Developers appreciate comprehensive documentation, and AI can generate it in minutes rather than hours. This quick win addresses a universal pain point while demonstrating AI's practical benefits.

Test generation and code review assistance provide measurable improvements quickly. Teams can track metrics like test coverage and review turnaround time, showing clear before-and-after improvements that justify expansion.

Performance optimization through AI analysis delivers quick wins that impact user experience. One e-commerce platform used AI to identify and fix performance bottlenecks, improving page load times by 30% within two weeks.

Scaling Based on Results

Success with AI development requires thoughtful scaling based on proven results rather than aggressive expansion. Organizations that scale strategically see sustained ROI growth while avoiding common pitfalls.

Establish clear scaling triggers based on metrics, not timelines. For example: expand to the next team when the pilot achieves 25% productivity improvement, roll out department-wide when bug rates drop 40%, go enterprise-wide when ROI exceeds 200%.

Build infrastructure ahead of scaling needs. Nothing kills momentum like performance problems or integration issues during expansion. Invest in robust platforms, comprehensive training materials, and support systems before scaling.

Create centers of excellence that support expansion. These groups of experienced AI users provide training, best practices, and troubleshooting support. They multiply the value of AI investment by ensuring effective adoption across teams.

Maintain feedback loops throughout scaling. Regular surveys, metrics reviews, and retrospectives identify challenges early. Organizations that actively gather and act on feedback during scaling see 50% higher ultimate adoption rates.

Future-Proofing Your AI Investment

Emerging Trends in AI Development Economics

The economics of AI development continue to evolve rapidly, with several trends reshaping ROI calculations and investment strategies.

AI model commoditization is driving costs down while capabilities increase. Open-source models and competitive pressure mean that raw AI capabilities become cheaper annually. Organizations that built competitive advantages on exclusive AI access must shift to focusing on implementation and integration excellence.

Specialized AI agents are emerging for specific development tasks. Rather than general-purpose coding assistants, we're seeing agents optimized for security analysis, performance optimization, or specific frameworks. This specialization drives higher ROI for targeted use cases.

Edge AI development is creating new economic models. As AI moves closer to where code executes, development costs shift from cloud subscriptions to edge infrastructure. Organizations preparing for this shift position themselves for next-generation advantages.

Collaborative AI networks enable smaller organizations to access enterprise-grade capabilities. Shared learning across organizations (while maintaining privacy) means that even startups can benefit from patterns learned across thousands of codebases.

Strategic Planning for Maximum ROI

Maximizing long-term AI development ROI requires strategic thinking beyond immediate tool selection. Organizations that plan holistically see sustained competitive advantages.

Invest in AI-friendly architectures from the start. Microservices, clear API boundaries, and comprehensive testing enable AI tools to provide maximum value. The cost of retrofitting architectures later far exceeds upfront investment in AI-ready designs.

Build versus buy decisions become more complex with AI. While AI platforms provide immediate capabilities, organizations with unique needs may benefit from custom training or specialized tools. The key is understanding where generic AI provides sufficient value versus where differentiation justifies custom investment.

Talent strategy must evolve alongside tool strategy. Hiring developers comfortable with AI collaboration, investing in continuous learning, and creating career paths that value AI orchestration skills ensures sustained ROI from human capital.

Partnership strategies multiply AI value. Organizations that share learnings (while protecting competitive advantages), contribute to open-source AI projects, and collaborate with AI vendors see faster capability improvements and better long-term economics.

Quick Takeaways

Every $1 spent on generative AI yields $3.71 average return, with financial services seeing 4.2x returns
Productivity has overtaken profitability as the primary metric for AI development ROI in 2025
Startups see fastest payback (2-4 months) while enterprises take longer but achieve larger absolute returns
Hidden costs include integration complexity and change management but rarely exceed 20% of total investment
Sector-specific benefits vary dramatically, with regulated industries seeing highest ROI from compliance automation
Long-term value compounds through knowledge accumulation, technical debt reduction, and developer skill enhancement
Success requires phased implementation with quick wins building momentum for broader adoption

Conclusion: The Economic Imperative of AI Development

The economics of AI development in 2025 present a clear verdict: organizations that fail to adopt AI development tools risk severe competitive disadvantage. With proven ROI ranging from 150% to over 400%, payback periods measured in months not years, and compounding benefits that accelerate over time, the question isn't whether to invest in AI development but how quickly you can implement it effectively.

The shift from profitability to productivity as the primary success metric reflects a maturing understanding of AI's true value. While immediate cost savings justify investment, the real returns come from sustained productivity improvements, enhanced innovation capacity, and competitive advantages that compound over time.

Success requires more than just tool selection. Organizations must invest in change management, choose appropriate use cases, and scale based on proven results. Hidden costs exist but pale in comparison to documented benefits. Sector-specific considerations matter, but every industry shows significant positive ROI from AI development adoption.

As we look toward the future, AI development economics will continue to improve. Costs are falling while capabilities rise. Early adopters are building insurmountable advantages through accumulated knowledge and optimized processes. The window for achieving first-mover advantages is closing rapidly.

The time for pilot programs and cautious experimentation has passed. Organizations serious about competing in modern software development must embrace AI tools, measure their impact rigorously, and scale based on the inevitably positive results. The economics are clear, the benefits are proven, and the future belongs to those who act decisively today.

Frequently Asked Questions

What's the average ROI timeline for AI development tools in 2025?

Most organizations see positive ROI within 6-12 months, with startups achieving payback in 2-4 months and enterprises in 8-14 months. The timeline depends on company size, implementation approach, and chosen use cases. Organizations focusing on quick wins like bug fixing and documentation see faster returns than those attempting comprehensive transformations.

How do I calculate the true cost of AI development implementation?

Include direct costs (licensing, training, integration) plus hidden costs (change management, technical debt adaptation, ongoing maintenance). Typical hidden costs add 15-20% to visible expenses. For a 50-person team, budget $50,000-$250,000 for first-year total costs, including all hidden expenses.

Which development tasks show the highest ROI from AI automation?

Bug fixing (50% time reduction), test generation (60% time reduction), and documentation (65% time reduction) consistently show the highest ROI. Code review and boilerplate generation also provide significant returns. Focus initial efforts on these high-impact areas for fastest payback.

How does company size affect AI development economics?

Smaller companies see faster payback but lower absolute returns. Enterprises face longer implementation times but achieve massive scale benefits. The sweet spot for ROI percentage is often mid-size companies (30-100 developers) who combine agility with meaningful scale. All sizes see positive ROI with proper implementation.

What are the biggest hidden costs in AI development adoption?

Integration complexity with legacy systems, comprehensive change management, ongoing model maintenance, and potential technical debt from initial adoption represent the main hidden costs. These typically add 15-20% to visible costs but are offset by benefits. Planning for these costs prevents budget surprises.

Start Measuring Your AI Development ROI Today

The data is clear: AI development tools deliver exceptional ROI when properly implemented. Whether you're a startup racing against runway or an enterprise optimizing thousands of developers, the economics favor immediate action.

Take action now: Start with a pilot program using RepoBird on your most painful development bottlenecks. Measure time savings, quality improvements, and developer satisfaction. Use our ROI framework to build your business case. Within weeks, you'll have data proving what thousands of organizations already know—AI development isn't just the future, it's the profitable present.

Start building with RepoBird and join the economic revolution in software development.

What's your biggest development bottleneck? Share your challenges and learn how others are using AI to transform their economics. How much could your team save with 30% productivity improvement?