Our March 2026 SSON article, "AI-Driven Outsourcing in 2026: New Ways to Transform with Intelligent Automation," explained why companies should rethink outsourcing in light of Artificial Intelligence (AI). This article looks more closely at how AI changes the economics of a Global Business Services (GBS) organization and its service providers – and what that means for pricing outsourced services. A follow-up article will explore how AI may also affect whether, and how, GBS organizations price and charge internal customers for services.
AI is Changing the Economic Structure of Outsourcing
The March 2026- article identified several ways AI is changing the economics of outsourced service–shifts that also apply to captive GBS organizations:
- Technology replaces labor as the primary cost driver. Historically, more than 70% of service delivery costs [BB1] came from labor rather than technology, infrastructure, or overhead. AI has the potential to reverse that balance.
- Costs become less sensitive to volume changes. AI creates non-linear cost dynamics, making service costs less responsive to volume shifts within a reasonable range. It is far easier to process incremental volume through an AI agent operating around the clock than to ask a human worker to do the same.
- AI creates business value beyond cost savings. AI offers more ways to deliver value beyond productivity gains and expense reduction. Unlike traditional outsourcing and GBS technologies, it creates value by aggregating and analyzing data to improve decisions. This enables outsourcers and GBS organizations to expand into more complex, company-specific services, including supply planning, marketing and customer analytics, HR planning, and financial planning. The economic value of these AI-enabled services comes not only from transactional cost savings, but also from higher revenue, stronger cash flow, and broader enterprise cost reduction.
- AI changes the risk profile. While AI can improve business outcomes, it also introduces new operational and economic risks. More complex processes increase the risk of broader operational losses, not just GBS transaction errors–for example, lost revenue from inventory shortages caused by inaccurate supply planning. Switching providers may also become more expensive when a provider's AI agents are deeply embedded with the buyer's data and process knowledge.
Together, these economic shifts call for a fundamental change in how outsourced services are priced.
The Traditional Outsourcing Pricing Model Has Remained Largely Stable
For the past 25 years, the economic structure and standard pricing model for Business Process Outsourcing (BPO) have remained relatively consistent. BPO providers typically supply direct labor, facilities, bolt-on technologies, management, and support resources, while buyers provide the ERP or similar systems of record used to process work against agreed service standards. As noted above, direct labor has historically been the largest component of provider cost.
The standard BPO pricing model reflects this cost structure. It is designed to fairly compensate providers for building, improving, and delivering services. Buyers typically expect the model to support several objectives:
The standard pricing model typically includes the following elements:
- A base monthly charge for a defined work volume. This charge reflects a mix of the provider's fixed and variable costs and is typically developed by moving the buyer's current work and volumes into a new provider operating model and cost structure.
Operating model improvements and related reductions in base charges are reflected through lower forward pricing.
- A variable monthly charge for volume changes. Transaction volume changes are addressed through Additional Resource Charges (ARCs) or Reduced Resource Charges (RRCs), usually priced per resource unit. Where possible, resource units are direct transaction types,such as electronic invoices, manual journal entries, or benefits applications,that drive the level of effort. Companies also commonly use full-time equivalent employees (FTEs) as the resource unit. ARC/RRC rates may vary modestly by volume to reflect fixed-cost absorption and productivity gains from standardized processes.
- Transition and transformation charges. One-time provider charges cover the transition from the current operating environment and transformation of the work over time. These charges are fixed for a defined set of transition and transformation activities, with payments tied to milestones and agreed success criteria.
- Termination charges. Charges for ending the agreement are defined in advance and vary based on contract timing and the reason for termination.
- Service level credits. Predetermined monthly charge reductions apply when a provider fails to meet agreed service requirements, such as quality, accuracy, or timeliness. These are typically defined in a separate, but highly correlated, exhibit to pricing in an outsourcing contract.
- Market pricing provisions. Mechanisms allow the parties to benchmark pricing against the external market and adjust prices accordingly. Like service-level credits, these provisions are typically set out separately from the core pricing schedule in an outsourcing contract.
Although variations exist, these components cover most BPO services.
Outsourcing Pricing Models Need Four Key Changes
AI does not require abandoning the traditional pricing model, but it does require several fundamental adjustments.
1. Use fit-for-purpose pricing.
Outsourced service portfolios will become more varied. Some services will remain traditional and transaction-based, while others will become more complex, expertise-driven, and decision-based. The level of AI use,and ownership of the AI itself,will also differ by service. Pricing methods therefore need to become more flexible and tailored to the type of service and transformation involved.
2. Reflect a non-linear cost structure.
AI has a more fixed cost structure than traditional labor-centric models, at least within defined volume tiers. Both models include variable costs, but AI consumption is driven by factors such as tokens, compute units, API requests, and messages. As AI agents become embedded and always available across more use cases, a larger share of costs shifts from variable consumption to fixed capacity.
Pricing models should therefore reflect both the changing balance of fixed and variable costs and the new drivers of those costs. Most services will still include some labor tied to transaction volumes, but as AI replaces labor, transaction volume alone may no longer be the primary basis for ARCs and RRCs. Pricing should also account for AI consumption factors and include mechanisms, such as transformation price-downs, that encourage lower consumption and reduced overall cost.
AI technology ownership also matters. It affects who pays for the technology and how the outsourcing cost structure is reflected in pricing.
3. Adopt outcome-based pricing.
As outsourcing providers take on more complex, judgment-based processes enabled by AI, the value proposition changes. Traditional transactional processes create value by meeting defined service requirements at the lowest cost, which aligns well with the traditional pricing model described above.
Complex, judgment-based processes create value through broader enterprise outcomes. As noted above, the economic value of these AI-enabled services comes not only from transactional cost savings, but also from higher revenue, stronger cash flow, and broader cost reduction. More intangible outcomes,such as stronger customer or employee loyalty or faster merger integration,may also create measurable economic value.
Outcome-based pricing should reflect this shift in value. Payments are earned when specific goals are achieved. Where the provider directly controls or strongly influences business outcomes, pricing can be tied to those broader outcomes. Where provider control is more limited, pricing can be tied to derivative outcomes, such as an AI chatbot resolving a customer service ticket.
4. Reflect changes in risk profiles.
AI introduces distinct risks that pricing models should address:
- Operational and financial loss: AI can increase both the speed of transaction-processing errors and the business impact of errors in more complex processes. Adjustments to service level credit methodology can help address these risks.
- Transformation failures: At this early stage of adoption, individual AI projects may be less likely to succeed than more proven transformation initiatives. Pricing models can address this risk through portfolio-level management and by clearly allocating transformation costs and rewards between buyer and provider.
- Relationship breakage: When a provider owns AI technologies that are deeply embedded in the buyer's processes, the relationship may require a more vested partnership rather than a traditional arm's-length arrangement. Pricing should adjust termination charges and conditions to support that deeper partnership.
Pricing for an AI-Enabled Future
AI is reshaping the economics of outsourcing, giving buyers and service providers a rare opportunity to rethink a pricing model that has remained largely unchanged for 25 years. Rather than making incremental adjustments, both parties should take a more holistic view of pricing and align its elements with the objectives that support a successful outsourcing relationship.