Consulting Firms Selling AI Transformation Can't Deliver It

BCG partnered with Anthropic to give clients access to Claude. Bain allied with OpenAI and used Coca-Cola as a marquee case study. McKinsey built Lilli on top of large language model infrastructure. Deloitte announced a partnership with Anthropic to make Claude available to over 470,000 of its professionals. Every major consulting firm now has an AI strategy practice, an AI partnership, and an AI narrative. What none of them have demonstrated is the ability to fundamentally transform their own operating model around AI — which is exactly what they are selling their clients.

Key Takeaways

• McKinsey's Lilli was deployed to approximately 72% of its 45,000 employees by 2025, reportedly saving 30% of time on research and knowledge synthesis — yet the firm's core billing model has not changed
• Only approximately 25% of McKinsey's global fees are linked to outcomes; 75% remain tied to traditional effort-based billing
• Deloitte Australia refunded part of a A$440,000 government contract after AI-generated errors were found in the deliverable, including fabricated citations and a made-up court quote
• Global spending on generative AI consulting reached $3.75 billion in 2024, nearly tripling from 2023, per Gartner — but few firms have demonstrated repeatable enterprise-wide deployment success
• Zimmer Biomet sued Deloitte for $172 million over a failed software implementation in late 2025, alleging the team accumulated fees without delivering a working system

The Advisor's Dilemma

Consulting firms face a structural conflict when it comes to AI transformation. Their business model depends on selling human hours. AI's primary value proposition is reducing the need for human hours. A firm that genuinely transforms its delivery model around AI — smaller teams, faster delivery, outcome-based pricing — would generate less revenue per engagement under its current economic structure.

This is not a technology problem. It is an incentive problem. Every major firm has the technology. McKinsey's Lilli can search 100,000+ internal documents in seconds and generate slides and research summaries on demand. BCG's Deckster automates presentation formatting. These tools work. What has not changed is the business model that surrounds them.

The data confirms this. Even at McKinsey — arguably the most progressive firm on pricing innovation — only approximately 25% of global fees are linked to outcomes. EY's leaders have publicly acknowledged the pressure to move toward "service-as-software" pricing but described the shift as slow. Partner compensation structures built over decades around revenue from billable hours are not easily unwound, even when leadership recognizes the need.

What Clients See

Clients are not unaware of this dynamic. Companies are increasingly bypassing traditional advisors in favor of internal teams, citing consultants' limited hands-on experience with AI and a gap between marketing claims and practical implementation at scale. Gartner found that while global spending on generative AI consulting reached $3.75 billion in 2024, much of this investment remains in the exploratory or pilot stage.

The Zimmer Biomet lawsuit against Deloitte — filed in late 2025, alleging $172 million in damages for a failed software implementation where the consulting team accumulated fees without delivering a working system — reflects a broader sentiment. Clients are losing patience with engagements that produce advisory output rather than operational outcomes.

Boards have shifted their expectations accordingly. By 2025, boards had lost confidence in transformation programs that promised a cycle of change but whose implementation was perpetually one year away from delivery. Boards now want partners who share accountability for outcomes, not just partners who provide advice and absorb none of the consequences.

The Partnership Dependency Problem

The largest consulting firms in the world — organizations that sell transformation, strategy, and technology advisory to Fortune 500 clients — have responded to AI by partnering with the same handful of AI providers that their clients can access directly. BCG and Deloitte partnered with Anthropic. Bain partnered with OpenAI. KPMG aligned with Microsoft and OpenAI. The proprietary tools these firms have developed are, at their core, wrapper applications around the same foundational models available to any enterprise willing to build or buy an integration.

This raises a question that consulting firms have not answered convincingly: if the primary AI capability is sourced externally, and the client can access the same provider, what is the consulting firm's unique value? The traditional answer — domain expertise, structured thinking, implementation experience — remains valid, but only if the firm actually delivers those things. When the deliverable is a strategy deck generated partly by the same AI tools the client already has access to, the value proposition weakens.

The risk for consulting firms is not that clients replace them with AI. The risk is that clients realize the AI capability is available directly from Anthropic, OpenAI, or Google, and that smaller, more specialized firms can apply that capability with more focus, more speed, and less overhead than a global consultancy.

The Internal Transformation Scorecard

If a consulting firm were evaluating a client's AI transformation readiness, it would assess six dimensions: leadership commitment, operating model changes, incentive alignment, technology integration depth, measurable outcomes, and cultural adoption. Apply those same criteria to the firms themselves.

Dimension	Score	What the Evidence Shows
Leadership Commitment	High	Every firm has an AI strategy; McKinsey deployed Lilli to 70%+ of staff; Accenture restructured its workforce around AI
Operating Model Changes	Minimal	Pyramid staffing persists; billing remains effort-based; service lines still operate as semi-autonomous businesses
Incentive Alignment	Poor	Partner comp tied to billings disincentivizes faster, leaner AI delivery — a partner billing less earns less
Technology Integration Depth	Moderate	Internal productivity tools are widely deployed; client-facing delivery integration is uneven
Measurable Outcomes	Limited	Few firms publish data on how AI changed delivery quality, speed, or satisfaction; revenue growth is tracked, not transformation
Cultural Adoption	Mixed	Internal AI tools are used; the gap between marketing narrative and day-to-day delivery is recognized internally

A consulting firm evaluating a client with this scorecard would recommend significant remediation. The gap between stated ambition and operational reality would be flagged as a material risk to transformation success.

What Happens Next

Two trajectories are emerging. Firms that genuinely restructure around AI — changing pricing, flattening teams, embedding AI in delivery rather than just productivity, and sharing risk with clients — will maintain their position as the market's most trusted advisors. They will be smaller, more profitable per partner, and more aligned with client outcomes.

Firms that use AI to improve internal margins while maintaining traditional pricing and delivery models will face increasing pressure from two directions: clients who recognize the arbitrage, and specialized competitors who deliver faster at lower cost with more transparency.

The consulting industry has navigated disruption before — offshoring, cloud computing, digital transformation. AI may be different, because it strikes at the core of what consulting sells: knowledge work performed by people. When the knowledge work can be performed by systems, the value shifts to judgment, context, and accountability — qualities that are not unique to large firms and may actually be better delivered by smaller, more committed teams.

The firms that succeed will be the ones that can honestly answer a simple question from their clients: "If you're advising us on AI transformation, show us how you transformed yourselves." Right now, most cannot. The AI Transformation Readiness Scorecard covers all six assessment dimensions, the diagnostic questions for each, and the organizational signals that distinguish genuine transformation from marketing narrative.