By Dr. Gleb Tsipursky

On a quarterly spend report, $10,000 that used to flow to a freelancer marketplace now shows up as a few hundred dollars of model usage tied to an expense platform dataset and a handful of vendor names that every CFO now recognizes. The shift looks small in a chart until you translate it into unit economics. Then it feels like a pricing shock that lands in the middle of knowledge work.

Ryan Stevens uses payments data from thousands of firms to track spending from Q3 2021 through Q3 2025, then treats the October 2022 release of ChatGPT as an adoption shock in a difference-in-differences design detailed in a Ramp research paper preprint on arXiv. The result delivers a hard ratio that leaders can apply to real budgets.

Twenty-Five Times Cheaper Changes Everything

Among the most exposed firms, each $1 decline in online labor marketplace spending lines up with about $0.03 of additional spending on AI model providers by Q3 2025, which implies roughly 20–25x cost savings when firms swap outsourced task labor for model usage. That single number deserves more attention than any headline about prompts replacing jobs because it captures the real mechanism that drives adoption inside companies: unit cost.

A 25x gap forces a new kind of budgeting conversation. Contract work typically scales linearly. More output requires more hours, more invoices, and more coordination time. Model usage scales through throughput. A manager pays for tokens and tooling, then pushes volume through a workflow that blends generation, review, and deployment. The spend looks like software spend, yet it often replaces labor spend in categories like copy drafts, first-pass research, support macros, lightweight coding scaffolds, and structured summaries.

Stevens’s dataset shows the spending mix changing alongside that ratio. The share of spending on online labor marketplaces falls sharply over time, while spending on AI model providers rises, reaching 2.85% by Q3 2025. Ramp also summarized the same pattern for a broader audience in its spending analysis, which helps connect the econometrics to what finance teams see in their own general ledger.

The ratio also explains why substitution looks gradual and then speeds up. Early usage often sits in pilots, sandboxes, and individual experimentation. Once a team builds a reliable loop for quality control, the economics pull the work toward AI like gravity. A workflow that turns a $2,000 contractor assignment into $80 of model usage will spread across departments fast. That same math becomes even more compelling when teams pair models with retrieval, templates, and evaluation, because the marginal cost stays low while quality rises.

The bigger point for leaders sits behind the decimal. Three cents on the dollar changes pricing power in every market where language and analysis drive value. It changes how agencies price retainers. It changes how startups staff early functions. It changes how enterprises think about shared services. The labor market reacts to wages and unemployment with a lag. Accounts payable reacts the moment a manager reroutes spend.

The Iceberg Below The Spend Data

The paper captures the visible tip of the iceberg because it measures what firms pay externally through online labor marketplaces and AI model providers in a payments platform. The deeper mass sits underwater inside payroll budgets, internal teams, and embedded contractors that never show up as Upwork or Fiverr line items. In my work with companies adopting generative AI, I keep seeing the same cost curve play out in those internal budgets, where leaders redirect effort rather than cancel a marketplace contract. The savings show up as slower hiring, smaller backfills, shorter project timelines, and fewer outsourced hours in categories that procurement never labeled as “freelance marketplace spend.”

This matters for interpretation. A $1 to $0.03 substitution ratio in a narrow spend category signals a broader capability shift that reaches far beyond the measured slice. The ratio captures direct replacement of purchased task labor. The iceberg includes internal task compression, where one analyst finishes in an afternoon what used to take a week because the model handles the first pass and the human focuses on judgment.

Research on task exposure helps explain why the underwater portion grows quickly. The OpenAI and University of Pennsylvania team behind a recent research paper estimates that about 80% of the U.S. workforce has at least 10% of tasks exposed to LLM capabilities, and about 19% has at least 50% exposure, which frames how wide the potential surface area is. When exposure spreads across roles, substitution can occur through workflow redesign even when the company keeps headcount steady.

Once leaders view the iceberg clearly, the operational priorities sharpen. Finance teams track spend share, yet they also track throughput metrics that reveal internal task compression. HR teams redesign entry roles toward evaluation and domain context. Procurement teams negotiate usage governance and data handling terms with model vendors. Public-sector guidance increasingly treats these capabilities as general-purpose technologies that require management capacity, as emphasized in the OECD’s work on AI adoption in firms.

The iceberg metaphor also clarifies why public debate often feels behind. A marketplace invoice disappearing creates a clean story. A team that quietly ships twice as much with the same headcount creates a subtle story that still drives real labor demand shifts downstream.

A Labor Market Repriced In Real Time

A 20–25x unit cost advantage pushes a repricing of work that touches every layer of the labor market, from gig platforms to entry-level pipelines to professional services. Work that maps cleanly to predictable language output faces direct demand pressure, while complementary work that involves evaluation, domain nuance, and integration gains value.

The most immediate effects show up where tasks are modular and buyers already treat labor as on-demand. A recent platform strategy working paper reports a meaningful decline in job posts for automation-prone categories, including a 21% drop in job posts for automation-prone work in the analysis presented in platform demand shifts. Those movements align with the Stevens finding because they share the same driver: buyers can buy output quality at a far lower unit price.

The next layer hits early-career roles that historically served as training grounds for higher-skill judgment. Stanford researchers using ADP payroll data report that early-career workers ages 22–25 in highly exposed occupations experienced a 16% relative employment decline after the widespread adoption of generative AI, even after controlling for firm-level shocks, as described in their research. That finding fits the iceberg dynamic. Firms gain the first-pass productivity from models, then reserve the remaining human work for people who already hold domain context, trust, and accountability.

Over time, the labor market will adapt through new task bundles. Companies will hire fewer people for pure drafting and more people for evaluation, customer nuance, and system building. Education and training will shift toward model supervision, data stewardship, and applied domain reasoning. Wage premia will flow toward roles that combine judgment with tool mastery, and toward roles that create proprietary feedback loops that raise quality. The same exposure research that highlights breadth also hints at productivity upside when tasks run faster at similar quality.

The Stevens ratio brings discipline to all of this. Leaders do not need to guess whether substitution exists. A three-cent-on-the-dollar signature in real payments data shows it already happening at scale in a measurable slice of the economy. The iceberg view suggests the larger change sits inside internal workflows, where the ledger records outcomes in slower hiring and higher throughput rather than clean vendor swaps.

Conclusion

The most important labor-market statistic in the generative AI era may be a ratio hidden in payments data: $1 of contracted online labor giving way to about $0.03 of model spend among the most exposed firms by Q3 2025 in Stevens’s estimate. That ratio explains why adoption persists even when quality debates continue, because buyers follow unit economics when they can protect output standards through review and governance.

The broader implications extend beyond freelancer marketplaces. The spending shift in the paper captures a visible sliver of substitution. The larger iceberg includes internal task compression, delayed hiring, and redesigned junior roles that reshape career ladders. Companies that treat this as a workforce redesign opportunity, with clear governance and strong training, will convert three-cent inputs into premium outputs. Everyone else will watch their cost structure get repriced by competitors who already did the math.

About the Author

By Dr. Gleb Tsipursky

Generative AI (Gen AI) promises transformative possibilities for businesses, but without clear goals and expectations, its potential challenges and risks become an expensive experiment in disconnected innovation. Leaders must approach Gen AI projects with purpose and precision, aligning them with organizational priorities while encouraging an environment where creative experimentation thrives.

Setting a clear roadmap ensures that teams remain focused, their efforts meaningful, and their results impactful.

Defining Purpose: Business Challenges to Prevent Gen AI Failure

Every Gen AI initiative must begin with a clear articulation of the business problem it aims to address. This clarity anchors the project in reality, ensuring it solves pressing challenges rather than chasing theoretical advancements.

For instance, an e-commerce company striving to boost customer retention could prioritize Gen AI projects that personalize shopping experiences or streamline service interactions. In contrast, a logistics firm might deploy Gen AI to optimize supply chain management through predictive analytics or real-time demand forecasting.

Case in point: a retail leader, faced with declining customer loyalty, identified their primary challenge as understanding customer preferences in real time. By leveraging Gen AI for dynamic product recommendations, the company increased repeat purchases by 15% within six months.

This success was rooted in their ability to connect the Gen AI initiative directly to a business objective: enhancing the customer journey.

When leaders pinpoint such strategic challenges, they create a clear purpose for Gen AI exploration. This alignment not only guides technical teams but also ensures that the resources and energy invested produce tangible outcomes.

Avoid Gen AI Failure By Balancing Structure and Flexibility

While clarity is vital, prescribing every detail of a Gen AI project can stifle innovation. Gen AI thrives on iterative experimentation. Leadership should focus on defining outcomes, not micromanaging methodologies. This balance allows teams to explore various AI models, algorithms, and tools to discover the most effective solution.

For example, consider a telecom company that set a goal to reduce customer service response times by 25%. Rather than dictating specific tools or processes, the leadership outlined the desired result and empowered teams to experiment. Some explored AI-driven chatbots, while others developed sentiment analysis models to triage customer tickets.

This flexibility led to a hybrid solution that reduced average response times by 30% and improved customer satisfaction scores by 12%.

This approach drives results, as well as enhances creativity. Teams feel ownership over their experiments and are motivated to uncover innovative solutions, knowing they have the freedom to adapt as they learn.

Accountability and Adaptability: Staying on Track

Clear goals establish benchmarks for accountability throughout a Gen AI project’s lifecycle. Teams can periodically assess progress against these objectives, ensuring efforts remain focused and aligned with business priorities. Dashboards tracking key performance indicators (KPIs) make it easier to evaluate success and make data-driven adjustments as necessary.

Take the case of an automotive manufacturer exploring AI-driven quality control systems. Initial experiments showed promising results in detecting defects on production lines, but further analysis revealed opportunities to refine the system for predictive maintenance.

By revisiting their original objectives and incorporating new insights, the company expanded the project’s scope to include early identification of equipment failures, saving millions in unplanned downtime.

The iterative nature of Gen AI means projects often uncover unanticipated opportunities. Leaders must remain agile, ready to refine objectives as new data or insights emerge. This adaptability ensures that projects continue delivering value even as circumstances evolve.

Cultivating Alignment and Transparency

Setting clear goals also cultivates transparency and engagement across the organization. When employees understand how their work contributes to broader business priorities, they are more likely to feel motivated and invested in the project’s success.

Consider a multinational financial institution that launched a Gen AI initiative aimed at detecting fraud. By clearly communicating how this effort aligned with the company’s mission to enhance trust and security, the leadership inspired teams across departments to collaborate. Data scientists, compliance officers, and IT specialists worked together to develop an AI model that reduced fraudulent transactions by 40% in its first year of deployment.

The sense of shared purpose was a key driver of this success.

By linking individual efforts to organizational goals, leaders create a culture of accountability and pride. Employees are more likely to see the value of their contributions, creating a commitment that transcends technical challenges.

Measuring Success: From Objectives to Outcomes

The ultimate test of any Gen AI initiative lies in its outcomes. Predefined goals provide a yardstick for evaluating success, enabling leaders to assess ROI and identify lessons for future projects.

For example, a consumer goods company set a goal to improve demand forecasting accuracy by 20%. The initiative surpassed expectations, achieving a 25% improvement and significantly reducing excess inventory costs. With this success, the company scaled the solution across its global supply chain, reaping even greater benefits.

Measuring outcomes also builds a foundation for continuous improvement. Leadership can analyze what worked, what didn’t, and how to refine strategies for future Gen AI endeavors. These insights ensure that each project contributes not just immediate value but also long-term organizational learning.

Client Case Study: Enhancing Customer Service Efficiency in a Mid-Sized Retail Bank

As a consultant, I collaborated with a mid-sized retail bank aiming to overhaul its customer service operations using Gen AI. The bank was grappling with rising customer expectations and increased competition, which highlighted the need to improve service responsiveness and satisfaction.

Leadership identified a clear goal: reduce average response times for customer inquiries by 30% within nine months, tying this initiative directly to their broader strategic objective of boosting customer retention and loyalty.

Identifying the Pain Points

The bank faced a range of customer service challenges. The average response time for inquiries was around 12 minutes, a figure that was increasingly at odds with customer expectations for quick and seamless service. Additionally, front-line customer service agents were overwhelmed by routine queries, such as account balances, loan eligibility requirements, and branch hours. This limited their ability to address more complex customer needs effectively.

The bank’s leadership saw an opportunity to leverage Gen AI to tackle these issues. They envisioned a solution that could handle repetitive inquiries efficiently while empowering human agents to focus on higher-value interactions.

Designing and Implementing the Gen AI Solution

To address these challenges, we developed and deployed a Gen AI-powered virtual assistant tailored to the bank’s specific needs. The assistant was built using advanced natural language processing (NLP) capabilities, allowing it to understand and respond accurately to a wide range of customer questions.

The implementation process began with a thorough analysis of the bank’s customer inquiry data. By examining patterns in historical call and chat logs, we identified the most common questions customers asked. These insights guided the design of the virtual assistant’s initial response templates.

Next, we integrated the virtual assistant into the bank’s existing communication channels, including their website, mobile app, and phone system. The assistant could interact with customers through text-based chat or voice, offering a seamless omnichannel experience.

To ensure accuracy and reliability, the virtual assistant was trained on a curated dataset of the bank’s terminology, policies, and product offerings. We also incorporated sentiment analysis capabilities, enabling the system to recognize when a customer was frustrated or dissatisfied and escalate the interaction to a human agent when necessary.

The Results

The implementation of the Gen AI-powered assistant exceeded expectations. Within the nine-month target period, the average response time for customer inquiries dropped by 35%, from 12 minutes to just under 8 minutes. This significant improvement directly contributed to a 20% increase in customer satisfaction scores, as measured through post-interaction surveys.

Moreover, the virtual assistant handled approximately 65% of all customer inquiries independently, freeing human agents to focus on resolving complex or sensitive issues. This not only improved operational efficiency but also enhanced the quality of service for customers with more nuanced needs.

The efficiency gains translated into substantial cost savings. By reducing the workload on human agents, the bank was able to reallocate resources and optimize staffing levels, resulting in annual operational savings of approximately $500,000.

Lessons Learned and Future Directions

Throughout the project, several key lessons emerged. First, the importance of clear, measurable objectives cannot be overstated. The leadership’s decision to set a specific target — a 30% reduction in response times — provided a clear focus for the team and a benchmark for success.

Second, flexibility during implementation was critical. Early in the deployment phase, we discovered that customers frequently asked questions about topics not included in the assistant’s initial training data, such as loan repayment deferrals during the pandemic. By quickly updating the system to address these new topics, we ensured its relevance and effectiveness.

Finally, ongoing monitoring and refinement proved essential. Post-launch, we implemented regular performance reviews to evaluate the assistant’s accuracy and effectiveness. These reviews allowed us to make iterative improvements, ensuring the system continued to meet customer needs as they evolved.

Looking ahead, the bank plans to expand the use of Gen AI beyond customer service. Inspired by the success of this initiative, leadership is exploring applications in areas such as fraud detection and personalized marketing, aiming to leverage Gen AI’s capabilities to drive further business growth and innovation.

Conclusion

Setting clear goals and expectations is the cornerstone of any successful Gen AI initiative. It provides direction, supports innovation, ensures accountability, and aligns efforts with strategic priorities.

Leaders who strike the right balance between structure and flexibility create an environment where experimentation flourishes without losing sight of tangible business outcomes. By cultivating alignment and transparency, organizations can harness the full potential of Gen AI, transforming it from a buzzword into a driver of sustainable growth.

As businesses navigate the complexities of Gen AI, the lesson is clear: clarity and purpose are not constraints; they are enablers. When goals are well-defined, Gen AI ceases to be a speculative investment and becomes a transformative force, delivering measurable value and driving innovation.

About the Author