The consulting industry has always sold a particular kind of certainty: that if you pay for expertise, structured thinking, and disciplined execution, you will get a better decision, a smoother transformation, or a measurable improvement. But artificial intelligence is quietly attacking the assumptions underneath that bargain—especially the pricing logic.
For years, firms such as McKinsey, Bain, and BCG have relied on a familiar commercial structure: fees tied to time, seniority, and the volume of work required to produce recommendations. Even when contracts included performance components, the default remained anchored to effort—hours billed, workstreams delivered, decks produced, and milestones reached. Now clients are increasingly asking a different question: if AI can accelerate analysis, drafting, benchmarking, and even parts of implementation planning, why should they pay primarily for the production of outputs?
At the same time, buyers are becoming more comfortable with a fee model that looks less like “pay for time” and more like “pay for results.” That shift is not just about cost. It is about accountability. When advice is easier to generate, the value moves toward judgment, risk-taking, and the ability to drive outcomes in messy real-world conditions. And that is forcing major consultancies to rethink what they are actually selling—and how they should price it.
What’s changing is not that AI eliminates the need for consultants. It’s that AI changes the cost structure of many tasks that used to be labor-intensive. The implication for pricing is immediate: if the marginal cost of producing certain deliverables falls, clients expect the price to fall too—or they expect the firm to justify the premium by taking on more responsibility for what happens after the deliverable is handed over.
This is where the pressure is building. Clients are questioning the value of advice while simultaneously getting used to fees based on successful task completion. In other words, they want to pay for impact rather than for the act of advising. That preference is showing up in procurement conversations, in contract language, and in how deals are structured across strategy, operations, technology, and transformation work.
The uncomfortable truth for consultancies is that “outcomes” are harder to define than “effort.” A time-based model is straightforward: you can count hours, track staffing, and bill accordingly. Outcome-based pricing requires agreement on success metrics, attribution, and timelines—often across multiple stakeholders and external variables. Yet AI is making the old model feel increasingly misaligned with the economics of delivery. The result is a commercial tug-of-war: clients push for outcome-linked fees; firms push back on measurement complexity and risk transfer.
The debate is now moving from theory to contract mechanics.
Paying for expertise versus paying for delivery
One of the most persistent questions in these negotiations is what clients believe they are buying. Is it expertise? Is it delivery capability? Is it the final outcome? Or is it some combination?
In traditional engagements, the consultancy’s role often ends at the point of recommendation: a strategy is proposed, an operating model is designed, a roadmap is created, and internal teams take over. Under that structure, outcome-based pricing can feel like a mismatch because the client controls execution. If the client fails to implement, the consultant may still have done its job—yet outcome-linked fees would penalize the consultant for factors outside its control.
But AI is changing the boundary between “recommendation” and “execution.” Many firms are increasingly embedding themselves deeper into implementation, using AI-enabled tools to accelerate workstreams and support ongoing decision-making. That creates a stronger argument for outcome-linked pricing: if the firm is not merely advising but actively shaping decisions and execution, then it can plausibly share in the upside and downside.
Still, clients are not satisfied with vague promises. They want clarity on what portion of the engagement is advisory and what portion is operationally accountable. That is pushing consultancies to break work into components that can be priced differently. Instead of one blended rate card, deals are being redesigned around modules: discovery and diagnostic (where AI reduces effort), design and planning (where AI accelerates drafts and scenario modeling), and implementation support (where human judgment and change management remain critical).
This modular approach is one reason pricing conversations are becoming more granular. Firms are learning that clients will accept higher fees for the parts that require scarce human capabilities—leadership alignment, governance, stakeholder management, and risk navigation—while expecting lower fees for parts that AI can do faster and cheaper.
Measuring “success” when the system is complex
Even when both sides agree that outcomes should matter, the next challenge is measurement. Complex transformations rarely have a single cause. A cost-reduction program might depend on procurement reforms, technology adoption, workforce changes, and market conditions. A customer experience initiative might depend on product teams, marketing execution, and operational capacity. A turnaround might depend on macroeconomic shifts and regulatory constraints.
So what does “success” mean in a contract?
Clients are increasingly insisting on measurable indicators, but they also want those indicators to reflect the firm’s contribution. That leads to a growing use of hybrid metrics: a mix of leading indicators (process milestones, adoption rates, cycle-time improvements) and lagging indicators (financial outcomes, retention, revenue growth). The aim is to avoid the trap of tying everything to end results that may be influenced by factors unrelated to the consultancy’s work.
However, hybrid metrics introduce their own disputes. Leading indicators can be gamed or improved without delivering the intended business impact. Lagging indicators can be delayed or distorted by external events. And attribution remains contentious: if the client’s internal team executes poorly, should the consultancy bear the financial consequences?
AI complicates this further because it can make certain improvements easier to achieve quickly. For example, AI-assisted analytics might identify cost-saving opportunities faster than before. But turning those opportunities into realized savings still requires organizational change. Clients may argue that the consultancy should be able to deliver more value per unit of effort because AI reduces the time needed to find and model options. Firms respond that the hard part is not analysis—it’s execution.
This is where the negotiation becomes less about pricing and more about role definition. Outcome-linked fees force both parties to articulate the consultancy’s responsibilities in a way that time-based billing never required. If the firm wants to be paid for outcomes, it must be willing to accept a clearer scope of accountability.
The rise of “task completion” models
The phrase “fees based on successful task completion” captures a middle ground between pure time-based billing and pure outcome-based pricing. Instead of tying payment to broad business results, contracts can tie payment to specific deliverables that represent meaningful progress toward outcomes.
Task completion models can include milestones such as:
– achieving a defined process redesign and operational readiness score
– delivering a working prototype or validated model that meets performance thresholds
– implementing a data pipeline or decision engine that passes agreed reliability tests
– training internal teams and reaching adoption targets for a new workflow
– completing governance and control frameworks that enable safe scaling
These milestones are often easier to measure than long-term financial outcomes. They also align with how AI changes delivery: AI can accelerate the creation of artifacts and the testing of scenarios, but it does not automatically ensure adoption, governance, or operational integration. That means the consultancy’s value can be framed as ensuring that the “last mile” is completed successfully—not just that a report is produced.
For clients, task completion pricing offers a practical advantage: it reduces the risk of paying for work that looks impressive but does not translate into operational capability. For consultancies, it offers a way to share risk without taking full responsibility for business results that depend on many variables.
Yet even task completion models require careful design. If milestones are too easy, clients will pay less than the firm believes it deserves. If milestones are too strict, consultancies will resist because they effectively become insurers against organizational failure. The best contracts tend to specify objective criteria, define what counts as “done,” and clarify how disputes are resolved.
AI’s hidden effect: it changes what clients think is “normal”
Beyond the direct economics of AI, there is a psychological shift. Clients are getting used to faster turnaround times and more iterative work cycles. They see AI tools generating drafts, summarizing research, and producing first-pass analyses quickly. Even when those outputs require human review, the speed changes expectations.
That expectation bleeds into pricing. Buyers begin to compare the consultancy’s cost to what they could achieve internally with AI-enabled workflows. If a consultancy can produce a benchmark in days rather than weeks, clients ask why the fee should remain anchored to the old timeline. If AI reduces the time needed to draft a business case, clients ask why they should pay for the drafting time.
This is not simply a cost-cutting impulse. It is a redefinition of value. Clients are trying to separate “work that can be automated” from “work that requires human judgment and accountability.” They are also trying to avoid paying for the parts of consulting that feel increasingly commoditized.
Consultancies, for their part, are responding by emphasizing the human elements that AI cannot replicate: leadership alignment, negotiation, governance, and the ability to manage trade-offs under uncertainty. But those arguments only land if the pricing structure reflects them. If a firm claims it is providing high-value judgment while charging for time spent producing AI-assisted outputs, clients will see the mismatch.
That mismatch is driving the move toward outcome-linked structures.
Why McKinsey and peers can’t solve it with a single pricing tweak
It would be tempting to assume that the industry will converge on one universal model: outcome-based fees with clear metrics. But the reality is messier. Different types of consulting engagements have different degrees of controllability and different timelines.
Strategy work often ends with recommendations and depends heavily on client execution. Technology transformations may involve implementation support and thus offer more controllable milestones. Operating model redesigns may require deep change management and governance. Procurement and sourcing optimization might have clearer financial levers and measurable savings.
Because of these differences, firms cannot simply replace hourly rates with outcome fees across the board. Instead, they are experimenting with combinations: fixed fees for defined scopes, variable fees for milestones, and