For roughly a century, the consulting industry sold one thing above all others: asymmetry. The firm knew things the client did not. It had watched the same problem solved forty times across forty boardrooms, it held the benchmarks, it owned the frameworks, and it rented that accumulated pattern recognition back to clients at a premium. The insight was the product. The slide deck was merely its packaging.

That asymmetry is collapsing, and the most honest people in the industry already know it.

Consider what McKinsey has done to its own crown jewels. The firm built an internal AI tool, Lilli, trained on a century of proprietary research: more than 100,000 documents and interviews, the distilled intellectual property that justified its fees. Over 75% of its 43,000 staff now use it around seventeen times a week, and the firm reports it saves consultants roughly 30% of the time previously spent gathering and synthesising knowledge. Tasks once handed to junior analysts like research, first-draft decks, and proposal scaffolding, are increasingly handled by a prompt.

Read that carefully. The most valuable knowledge asset in management consulting has been compressed into something a consultant queries in seconds. And if a firm’s hundred-year archive can be reduced to a conversational interface, the client sitting across the table can build a credible version of the same capability against public models for a fraction of the cost. The “we know things you don’t” moat has not narrowed. For a large class of problems, it has gone.

This is the uncomfortable symmetry of AI in professional services: both sides now hold roughly the same instrument. The question for senior leaders, whether you sit in strategy, finance, technology or procurement, is whether that produces harmony or a harder fight. The answer is a harder fight. But not where most people are looking.

What actually got commoditised

The mistake is to assume that all of consulting has been commoditised. It hasn’t. One specific layer has: synthesis. The capacity to gather, structure and articulate known information, the work that filled the bottom two-thirds of every engagement pyramid, is now close to free.

The most rigorous evidence comes from the industry itself. In 2023, Boston Consulting Group ran a controlled experiment with Harvard, MIT, Wharton and Warwick across 758 of its own consultants. On tasks within the capability of the AI, consultants using it produced work rated around 40% higher in quality, completed roughly 12% more tasks, and worked about 25% faster. The gap between the firm’s strongest and weakest performers narrowed sharply. For structured, known work, AI didn’t merely help it flattened the difference between an average consultant and an excellent one.

If that were the whole story, clients would simply stop buying. But it isn’t the whole story, and the second finding is where the value goes.

Where the value relocated

The same study found that on tasks outside the AI’s reliable frontier, genuinely novel, ambiguous problems where the data conflicted with the human signals, consultants using AI performed about 19 percentage points worse than those working without it. The researchers called this the “jagged frontier”: the technology is brilliant on one side of an invisible line and confidently wrong on the other, and the line is not obvious from the task itself.

This is the hinge on which the whole industry now turns. The value of professional advice did not disappear. It relocated from having the insight to exercising judgement about it. Knowing which problem you are actually solving. Knowing which AI output to trust and which to interrogate. Knowing where the frontier’s edge sits on this particular question, in this particular business, under this particular set of political and commercial constraints. None of that lives in the model.

The cautionary tale already exists. Klarna, the fintech, went further than almost anyone, replacing large parts of its software estate and cutting customer-service roles in favour of AI, with its chief executive publicly championing an AI-run company. Within months, it conceded the approach had “gone too far,” that quality had suffered, and it began bringing humans back into the loop. The lesson is not that AI failed; it plainly delivered enormous efficiency. The lesson is that when you remove the human judgement layer entirely, the thing that breaks is the thing that was always hardest to buy.

What this means for the people who buy advice

For a FTSE leader, the strategic implication is liberating rather than threatening, provided you act on it.

You no longer need to buy the pyramid. The traditional model required clients to fund a large base of junior consultants whose principal output was precisely the synthesis work that AI now does in-house. Paying a premium for that layer today is paying for a horse when the car has already arrived.

What you should buy instead is scarcer, and you should buy less of it. The residual value is concentrated in senior, judgement-rich expertise: people who have genuinely sat inside the hard version of your problem and can tell you where the model is quietly lying. That expertise is expensive precisely because it is scarce, but you need it in far smaller, more precise quantities than the old model demanded. The intelligent purchase is no longer a team of thirty for six months. It is the right three experts, plugged into your own AI-augmented people, for exactly as long as the judgement is needed.

This inverts the commercial relationship. Power moves to the client. Deliveries get leaner. The consultancy’s leverage no longer comes from headcount, or from hoarding insight; it comes from how quickly it can place genuinely scarce expertise next to your problem, and how cleanly it gets out again.

The model that wins

The firms that struggle will be those still selling the insight premium, billing for synthesis that the client can now generate for itself. The firms that win will be those that have already accepted the commoditisation and rebuilt around the scarce part: flexible, on-demand access to deep expertise, drawn from a far wider bench than any single partnership can carry on its payroll, and deployed in precise increments rather than standing armies.

That is the bet B2E has been making for years, and AI has simply accelerated the logic of it. When the insight is free, the only defensible product is the right expert, in the right place, for exactly as long as you need them and not a daily rate longer.

The asymmetry of knowledge is over. The asymmetry of judgement is just beginning. The real question for any leader buying advice over the next eighteen months is a blunt one: are you still paying for what the machine now does for nothing, or for the scarce human capacity it cannot replace?

References

BCG / Jagged Frontier study

Dell’Acqua, F., McFowland, E., Mollick, E., Lifshitz-Assaf, H., Kellogg, K., Rajendran, S., Krayer, L., Candelon, F., & Lakhani, K. (2023). Navigating the Jagged Technological Frontier: Field Experimental Evidence of the Effects of AI on Knowledge Worker Productivity and Quality. Harvard Business School Working Paper, No. 24-013. Published 18 September 2023. Available at: https://www.hbs.edu/faculty/Pages/item.aspx?num=64700

McKinsey / Lilli

Smaje, K. (quoted in Entrepreneur, 2 June 2025). McKinsey Is Using AI to Create PowerPoints and Take Over Junior Employee Tasks. Entrepreneur. Available at: https://www.entrepreneur.com/business-news/ai-creates-powerpoints-at-mckinsey-replacing-junior-workers/492624

For Lilli’s document base and launch context, see also: AI Unlocks New Golden Age of the ‘Consulting Historian’. Innovators Magazine, 12 May 2025. Available at: https://www.innovatorsmag.com/ai-unlocks-golden-new-age-of-the-consulting-historian/

Klarna

Siemiatkowski, S. (quoted in Fortune, 9 May 2025). As Klarna Flips from AI-First to Hiring People Again, a New Landmark Survey Reveals Most AI Projects Fail to Deliver. Fortune. Available at: https://fortune.com/2025/05/09/klarna-ai-humans-return-on-investment

See also: Klarna Changes Its AI Tune and Again Recruits Humans for Customer Service. CX Dive, 9 May 2025. Available at: https://www.customerexperiencedive.com/news/klarna-reinvests-human-talent-customer-service-AI-chatbot/747586/

About the author, Dale McNeill.

Dale has spent the majority of his career consulting in IT Service Management, firstly with Northgate IS, then more recently with Accenture. An ITIL Expert with Cross Industry knowledge, Dale has led many consulting projects in large programmes across the IT Lifecycle. Dale joined B2E in January 2017, supporting some of our key client accounts and helping develop capabilities to support new clients, showcasing our expertise and growing our B2E community.

When not in the office, Dale enjoys going to a variety of gigs, visiting the cinema, cooking and talking to his budgie, Eric. A new addition to the family (Gus the Cocker Spaniel puppy) also keeps him busy and very fit!