How AI is revolutionizing consulting firm structure

Artificial intelligence is shifting the traditional “pyramid” consulting model to a narrower “obelisk” model, according to a Harvard Business Review article by consulting experts David Duncan, Tyler Anderson, and Jeffrey Saviano.

While some alarmists say that AI advances could render consulting obsolete, the authors contend that consulting isn’t disappearing but being reshaped.

The traditional “pyramid” structure of large consultancies relies on a wide base of junior consultants who work on research, modeling, and analysis – and support a narrow apex of senior leaders who oversee strategy and client relationships.

Gen AI is undermining the model by automating the tasks of junior consultants – and forcing consultancies to consider a revamp of their delivery model.

Boston Consulting Group is using Deckster, a tool that can spit out presentation decks in minutes, while McKinsey is using proprietary AI assistant Lilli, which is used by 72% of its workforce and has reduced research and synthesis time by approximately 30%.

The new model that effectively leverages the efficiencies of AI systems is a tall, narrow obelisk with fewer layers and smaller teams built around three roles:

• AI facilitators are early-career consultants AI tools and data pipelines who design and refine AI-driven workflows.
• Engagement architects are experienced consultants who lead projects, define problems, interpret AI outputs, and translate them to actionable strategies.
• Client leaders manage relationships with senior executives and overall consulting firm strategy.

Duncan et al say that nimbler, AI-native boutiques are already putting the model into action. Unity Advisory, launched by former Big Four partners and backed by $300 million in private capital, uses pods of senior consultants who work closely with proprietary AI tools. The firm doesn’t hire large cohorts of entry-level analysts or rely on hierarchical, middle-management structures.

The authors note that incumbents rarely disrupt themselves, so large firms may struggle to evolve the pyramid model that has long shaped their culture, economics, and delivery. And although consulting firms trumpet their sizable investments in AI, those capabilities often remain siloed.

Traditional firms also face a talent realignment challenge. The major consultancies are built on recruiting and training large amounts of generalist MBAs, but the new model demands smaller cohorts fluent in AI tools. Although the firms are launching upskilling programs, the culture and incentives still favor hours billed over insight delivered, Duncan et al say.

Firms that cling to old structures risk becoming slower, more expensive, and irrelevant.

The transition to an obelisk model isn’t an easy one, though, because it requires a wholesale restructuring of talent models, workflows, and compensation models.

The new model also requires guardrails to ensure AI-driven decisions are understandable, equitable, and accountable. The old model has the benefit of layers of review by analysts, managers, and partners that makes it easier to catch errors and assign responsibility.

“Despite the challenges for consultancies, this is not a moment for incrementalism. The winners will be those who move first and reimagine the industry before someone else does it for them,” the authors conclude.