By Adam Mills
AI Editor
0
Share
Newsweek is a Trust Project member
See more of our trusted coverage when you search.Prefer Newsweek on Googleto see more of our trusted coverage when you search.
The way IT services firms make money is starting to drift out of sync with the way they deliver work.
For years, companies have priced software development and related services according to how many people are assigned to a project and how long they spend on it, because effort and results tended to move together. More engineers meant more code, more progress and, in turn, more value.
That relationship is beginning to break down. As AI begins to take on more of the execution itself, the amount a small team can produce has expanded sharply. Work that once depended on large teams can now be handled by far smaller groups working alongside automated systems, compressing timelines and changing how work gets done.
Read More on AI
“The model no longer reflects how value is created,” Abby Kearns, CEO of ActiveState, a software supply chain security company, told Newsweek. AI, she added, is reshaping the relationship between effort and output in ways that traditional pricing approaches struggle to capture.
IT services firms have long explored moving toward output- or outcome-based pricing, but adoption has been uneven, constrained by internal structures and customer expectations that still revolve around headcount.
AI is widening the gap between how services are priced and how they are delivered, making that mismatch more visible to clients.
“Customers are now aware that the cost of inputs is dropping,” Seth Robinson, vice president of industry research at CompTIA, an IT industry trade association, told Newsweek. “That means defining value requires more transparency into how their productivity is impacted.”
As that awareness grows, buyers are more willing to question pricing that no longer aligns with how work is produced.
The Breakdown of a Linear Model
Both sides are starting to confront how far apart those business assumptions have become. Buyers are increasingly familiar with software-driven delivery and are questioning why pricing remains tied to time as production becomes more automated.
Providers, meanwhile, are still operating within models built on billable hours and staffing levels—an approach that presumes a direct relationship between labor and output.
That relationship is starting to break.
...
“A headcount-based model assumes that value scales with effort, but AI fundamentally challenges that premise,” Rajesh Nambiar, president of NASSCOM, the Indian IT industry trade association, told Newsweek. Nambiar said the shift is pushing firms away from labor arbitrage toward models that reflect how AI is shaping delivery.
“It assumes a linear relationship between people and production, and AI removes the linearity entirely,” Kearns said. Suddenly, contracts built around hours or headcount lose credibility when the same outcomes can be achieved with fewer resources.
A recent Newsweek “AI Impact Forum” discussion highlighted how the industry has long aimed to move from input-based pricing toward outputs and, eventually, outcomes. Former Cognizant CEO Frank D'Souza described that progression as a logical evolution for the sector.
What has changed is the speed at which the underlying economics are moving. The ambition is not new. The urgency is.
Why This Is Hard to Execute
Moving to output- or outcome-based approaches requires more than rewriting contracts. It changes how value is defined, measured and shared between provider and client.
In a traditional engagement, providers are accountable for activity, delivering a defined scope of work over a set period of time. In an outcome-based model, providers are held accountable for results, requiring a clearer, shared understanding of what success looks like.
“An outcome-based engagement looks more like a shared definition of success, and less like a time-and-materials contract,” Kearns said.
That change introduces new demands on both sides. Providers must rethink how they deliver work and measure performance, while organizations on the client side face new expectations around defining success more precisely**.**
“For clients, the biggest obstacle is having clear internal calculations around expected benefits,” Robinson said. Without that foundation, it becomes difficult to structure pricing models that tie services to outcomes rather than effort.
The friction is not just technical—it's organizational.
...
Many services firms still rely on revenue models tied to utilization, shaping everything from staffing decisions to financial forecasting. Clients, meanwhile, are accustomed to buying capacity rather than defining outcomes in advance.
Both sides are operating within systems built for a different way of working, even as AI ushers in a new era.
“It can be very difficult to change quickly,” Kearns said. That inertia helps explain why progress has been slow, even as the limitations of the current approach become more apparent.
From Transition to Viability
As AI expands what smaller teams can deliver, the gap between perceived value and billed effort is becoming more visible, particularly for buyers who are increasingly attuned to how software is produced.
In the near term, a mix of old and new pricing models is likely to persist, with approaches tied to hours and staffing gradually giving way to models that reflect outputs and business results.
The underlying question is no longer whether the model changes, but how quickly firms can adjust to it.
“Firms that do not adapt will find themselves in a steadily narrowing position,” Kearns said.
At a certain point, she added, “it is not a transition problem, it is a viability problem.”
For more on how AI is reshaping the economics of IT services, watch Newsweek’s recent “AI Impact Forum” webinar discussion with Frank D'Souza above. You can also sign up for the next "AI Impact Forum" webinar, taking place on April 30 and featuring Tiger Tyagarajan, the former CEO of the advanced technology solutions company Genpact.
View 0 Comments
Submit Correction
View Editorial & AI Guidelines