Generated Title: McKinsey's AI Gamble: Trading Billable Hours for Boardroom Scorecards
McKinsey & Company, the consulting behemoth, is undergoing a seismic shift. Forget the image of consultants billing by the hour; the firm is increasingly tying its fees to client outcomes, particularly in AI-driven transformation projects. About a quarter of McKinsey’s global fees now come from this "performance-based" model, according to Michael Birshan, a managing partner. This isn't just a tweak; it's a fundamental change in how the consulting giant makes its money.
The rationale is straightforward, if a little self-serving. As Kate Smaje, McKinsey’s global leader of tech and AI, puts it, clients undertaking "big career bets" appreciate having their consultants share the risk. "Your scorecard with your board is our scorecard," she claims. In theory, it aligns incentives: McKinsey gets paid more if the client hits its targets (investor targets, revenue goals, customer satisfaction scores, etc.).
The Devil's in the Scorecard
But here's where the data analyst in me gets twitchy. What constitutes "success"? Who defines the metrics, and how easily can they be manipulated? A customer satisfaction score, for example, can be juiced with targeted marketing campaigns that offer short-term gains at the expense of long-term brand loyalty. Revenue goals can be met through unsustainable cost-cutting measures. (The acquisition cost was substantial (reported at $2.1 billion).)
Consider the incentives at play. McKinsey, naturally, wants to showcase its AI prowess and secure lucrative, multi-year transformation projects. Clients, under pressure from investors and boards, want quick wins and demonstrable ROI. This creates a potential for misaligned priorities, where the focus shifts from genuine, sustainable improvement to hitting easily gamed metrics.
And this is the part of the report that I find genuinely puzzling: If McKinsey is truly a "genuine partner" and not a "supplier" or "vendor," as Smaje claims, why the need for such a tightly defined scorecard? Shouldn't true partnership be based on shared values and a long-term vision, rather than a series of quarterly targets? This feels less like a collaborative endeavor and more like a high-stakes poker game, where both sides are incentivized to bluff.
The "Service-as-a-Software" Mirage
The shift towards outcome-based pricing isn't unique to McKinsey. Raj Sharma of EY notes that AI agents may force firms to adopt a "service-as-a-software" approach, where clients pay based on results. The idea is that AI can automate many of the tasks traditionally performed by consultants, leading to greater efficiency and measurable outcomes. AI is reshaping how McKinsey makes money
However, the promise of "service-as-a-software" is, in my opinion, largely hype. AI, at least in its current form, is a tool, not a magic bullet. It can analyze data, identify patterns, and automate repetitive tasks, but it can't replace human judgment, creativity, or strategic thinking. Furthermore, the effectiveness of AI depends entirely on the quality of the data it's fed. Garbage in, garbage out, as they say.
The risk is that consulting firms will become overly reliant on AI, sacrificing the nuanced understanding and personalized advice that clients truly value. Instead of acting as trusted advisors, they'll become glorified data processors, churning out reports based on flawed algorithms and easily manipulated metrics.
Trading One Problem for Another
McKinsey’s move towards outcome-based pricing is a clever marketing ploy, but it doesn't address the fundamental problem plaguing the consulting industry: a lack of accountability and a tendency to overpromise and underdeliver. Tying fees to client scorecards may create the illusion of alignment, but it also creates new opportunities for manipulation and misdirection. Ultimately, the value of consulting lies not in hitting arbitrary targets, but in providing sound, objective advice that helps clients make better decisions. And that's something that no AI, no matter how sophisticated, can ever replace.
