The rise of AI is forcing a fundamental economic shift in consulting. One of those shifts is the move from a labour-intensive, time-billed service model towards productised, subscription-based advisory, writes Johan Devér, a former McKinsey consultant and co-founder of Grasp.
During my time at McKinsey & Company, one of the most common questions clients asked us to answer was “How do we double revenue in five years?” This question could take months to answer: analysing organic and inorganic growth opportunities by reviewing market reports, conducting expert interviews, building Excel models, and synthesising the findings into a 100-slide recommendation deck. A seven-figure bill would soon arrive through the post.
Traditional consulting exists entirely because strategic analysis is expensive and laborious to produce. You need smart, analytically-minded people, time, and access to information. Primarily, that data is very unstructured, so to be able to understand and use it correctly, you need to spend a lot of time finding, analysing, and synthesising it.
That work, manually reviewing hundreds of company websites, reading through lengthy industry reports, and building market maps from scratch, is the backroom work that doesn’t appear obvious when looking at the polished slide deck at the end.
Whether a client wants to increase revenue growth or reduce costs, the main two project types in consulting are valued by time spent, rather than outcome secured. A client is buying the hours, not just the outcome – whether revenue actually increases.
AI’s impact on consulting
Today, an AI system can produce that same analysis overnight. It can assess market opportunities, identify acquisition targets, map competitive dynamics, and present options in client-ready slide decks, all while the team sleeps. The system has full context of the client’s business, access to comprehensive datasets, and the capability to run multiple scenarios in parallel.
This raises an uncomfortable question for the consulting industry: when the deliverable can be generated automatically, what exactly is a consulting firm selling?
The current model of consulting has endured because the work has required human intelligence and judgment. Machines couldn’t grasp the nuance in a market trend or understand the strategic implications of a competitor’s product launch. The consulting firms that won were those that could deploy the most capable talent.
But, AI is fundamentally inverting these economics. The consulting model has always been low fixed cost and high marginal cost, because the primary expense is human time. Each new project requires assembling another team and billing another few thousand hours. AI flips this to high fixed cost and low marginal cost. The investment goes into building the system, but once it exists, answering another client query costs very little.
This is a shift we’ve seen before, though perhaps not as clearly in professional services. Salesforce replaced what sales teams did manually with complex databases, Rolodexes, and filing cabinets. Sales forecasting, pipeline management, and customer tracking all moved from human-intensive admin work to automated systems. When software can replicate what humans do, and package it neatly into a single product, industries restructure entirely.
The rise of productisation
What makes this technological shift particularly significant for consulting is that clients are also starting to expect product-style experiences from advisory services.
Consulting’s traditional edge has been access to proprietary benchmarks, frameworks, and insights from past projects. AI can now synthesise across all this existing intellectual property and apply this to the specific requests of the client at almost infinite scale, with better pattern recognition, uncovering otherwise overlooked insights. This shifts the value proposition, and the analysis itself becomes commoditised. What remains valuable is judgment, interpretation, trust, and the client relationship.
Johan Devér’s former employer McKinsey is one of the frontrunners in productised advisory
If AI’s impact is so clear, why aren’t more firms moving quickly to adopt it? There are three forces creating friction.
The first is a scale paradox. Larger firms have the most intellectual property to productise. Decades of past projects, proprietary frameworks, and industry-specific knowledge. But the bigger the organisation, the more organisational inertia. Changing how work gets done when you employ tens of thousands of consultants is monumentally difficult.
Secondly, the current consulting model is built on day rates, so how do consultancies transition to subscription-based or outcome-based pricing without destroying margins? The economics need to work, and at the moment, it’s not immediately obvious how. This means there is almost no personal incentive for decision-makers, who are typically compensated based on billable hours, to embrace AI. It’s rational for individuals to resist, even if it’s strategically necessary for the firm.
Finally, big consultancies also can’t keep up with the pace of technological advancements. By the time a large organisation builds and ships an internal AI tool, the frontier has moved. What seemed cutting-edge six months ago is antiquated by the time it reaches users, creating a perpetual sense of being behind. This creates a build versus buy dilemma: Should firms build proprietary AI products internally or partner with specialists?
The irony is that established firms have the best raw material. They have the intellectual know-how, the client relationships, and the industry knowledge. But they often have the slowest metabolism for change. Smaller, newer firms have less IP but can move faster. There’s a brief window where established firms can leverage their advantages before more nimble players close the gap.
From service-based towards productised advisory
The application of AI presents a shift from service-based advisory towards productised advisory. These are standard strategic questions, like competitive analysis or target identification, that can be automated and delivered through always-on AI systems that leverage existing intellectual property. This is where most of the volume of current consulting work sits.
Instead of assembling a team for each project, clients get subscription access to that capability, like having a research team on permanent retainer. The pricing reflects this: monthly or annual subscriptions, or fees per query. There’s also a version of this where clients connect their own real-time data to AI advisory systems.
Yet, there will always be genuinely novel problems, for which there’s no playbook. This work remains bespoke and expensive, but it will become rarer as a percentage of total advisory work. Where most of the future value for consultants will concentrate in the future is strategic work, where AI produces the analysis, but consultants provide interpretation and challenge. The client gets comprehensive data but pays for human interpretation of what it means. The pricing shifts to outcomes or milestones rather than time.
The decision to make
So, every advisory firm faces a decision. You can defend the current model, keep billing by the hour, and resist productisation. You can gradually adapt, acknowledge the shift, but move slowly, create pilot projects, and take three to five years to transition. Or you can move aggressively, productise core offerings now, and accept short-term revenue disruption for long-term positioning.
The firms that will lead in five years aren’t necessarily those with the most consultants today, but it is those who can deliver a truly differentiated, client-specific product. The consultancies integrating their intellectual property into AI systems fastest will be the first to achieve this, but they must recognise that early enough. Those who wait may well discover that the most expensive decision they made was to do nothing at all.
Source link
#future #consulting #billing #hour #productised #advisory
