Due to the influence of artificial intelligence (AI) on pricing, Indian IT companies are beginning to record a slight increase in the number of outcome-based deals. These new agreements increasingly link payment to achievable business metrics, marking a shift away from traditional Time and Materials (T&M) and fixed-price payment models.
The outcome-based pricing model represents a fundamental shift from long-standing traditional approaches, as it moves the focus from expended resources to measurable results. Under this format, clients pay exclusively for actual usage or achieved outcomes. In contrast, standard models involve paying for the service provider's effort, which is typically calculated based on daily hours or total project time.
Tata Consultancy Services (TCS), India's largest IT firm, reported that its client engagement models cover various horizons. These include outcome-based models, end-result models, fixed prices, and T&M. Aarti Subramanian, COO of TCS, told analysts last month that while T&M requirements still come in, they imply responsibility for achieving results. However, a significant transition towards outcome-based commitments is observed in three considered models, especially in agency GBS, this quarter.
For Coforge, outcome-based contracts account for about 6-7 percent of the company's total revenue in current revenue mode. Meanwhile, Cognizant concludes 45 percent of its Business Process Outsourcing (BPO) agreements under outcome-based commercial models.
Nevertheless, such contracts remain a minor part of IT companies' revenues, as most business is still focused on traditional T&M and fixed-price models. Most companies have not disclosed details of these contracts but note that some AI-driven deals are already outcome-based, even while they work on creating a corresponding pricing strategy.
The outcome-based pricing model is well-suited for small and short-term, AI-managed deals aimed at measurable results. A significant reason for this is that enterprises are scrutinizing their investments much more closely, as AI is transforming both the service delivery economy and client expectations regarding what they should pay for.
Saurbh Gupta, President of HfS Research, noted that while large contracts continue to be viewed as a way to reduce costs, new deals often feature a hybrid pricing model that includes subscription, consumption, and specific outcomes. He added that if 80 percent of the deal is executed by a platform or tool, there will be a subscription cost, followed by a consumption cost, which can be measured by week or AI agent time. Week time is measured in hours, and AI time in tokens, with the third component being performance.
Tech Mahindra, for instance, recently closed a healthcare sector deal where the commercial model is tied to measurable indicators: a reduction in calls by approximately 40 percent, a decrease in average resolution time by 20 percent, a reduction in technical debt by 30-35 percent, and a substantial increase in productivity during the contract term.
However, Infosys has not yet seen an increase in such deals. Despite the company's clients showing greater interest and holding more discussions, Infosys CEO Salil Parekh stated that 'it has not become such a big part of our business.'
Mayank Verma, Global Head of Data and AI at Xebia, believes the world needs to move from staff augmentation to a 'division' model that combines agents and humans. He noted that people are currently experimenting with this, and it may take a couple of years, but market share will increase as the company is in a transitional phase.