McKinsey's Sternfels: 20,000 of 60,000 Staff Are AI Agents
OfficeChai reports on January 14, 2026 a major statement by Bob Sternfels (Global Managing Partner, McKinsey & Company): "my latest answer to you would be 60,000, but it's 40,000 humans and 20,000 agents." McKinsey headcount is now counted as humans + agents: 60,000 = 40,000 humans + 20,000 AI agents. Massive acceleration trajectory: "Little over a year and a half ago, that was 3,000 agents and I originally thought it was going to take us to 2030 to get to one agent per human." The human/agent parity initially projected for 2030 is now achievable within 18 months — an acceleration of >5× relative to the original projection. Explicit business-model shift: "We're migrating pretty quickly away from, let's call it pure advisory work... moving to much more of an outcomes-based model." — McKinsey is abandoning the billable-hours / advisory model in favor of an outcomes-based model where compensation is aligned with measurable results. McKinsey context 2024-2026: Lilli (internal chatbot) used by 70% of employees, 200 technology roles cut in November 2024, a 9-month-salary severance offer (April 2024). Major relevance for the 2026 dossier: this is the first top-tier consulting player to (a) publicly state and (b) precisely quantify the integration of AI agents into its accounting headcount.
By Article rapporté par **OfficeChai Team**// Source officechai.com ↗/Reading 2 min/.md// Auto-verified translation
#Bob Sternfels#McKinsey & Company#McKinsey Managing Partner Global#OfficeChai#60000 people 40000 humans 20000 agents#AI agents headcount#human agent parity 2030 to 18 months#acceleration 3000 to 20000 agents in 18 months
Summary
OfficeChai reports on January 14, 2026 a major statement by Bob Sternfels (Global Managing Partner of McKinsey & Company): "60,000, but it's 40,000 humans and 20,000 agents." McKinsey headcount is now counted as humans + agents: 60,000 = 40,000 humans + 20,000 AI agents.
Massive acceleration trajectory: "Little over a year and a half ago, that was 3,000 agents and I originally thought it was going to take us to 2030 to get to one agent per human." The human/agent parity initially projected for 2030 is now achievable within 18 months — a ×6.7 acceleration relative to the original projection.
Explicit business-model shift: "We're migrating pretty quickly away from, let's call it pure advisory work... moving to much more of an outcomes-based model." McKinsey is abandoning the billable-hours / advisory model in favor of an outcomes-based model where compensation is aligned with measurable results.
McKinsey context 2024-2026: Lilli (internal chatbot) used by 70% of employees, 200 technology roles cut in November 2024, a 9-month-salary severance offer (April 2024).
Major relevance: this is the first top-tier consulting player to (a) publicly state and (b) precisely quantify the integration of AI agents into its accounting headcount. The phrase "60,000 people but 20,000 of them are agents" is set to redefine the very conception of headcount in knowledge-based services.
Connection to the watch dossier: strong convergence with MediaPost MandeseBillable Hours Are Dead (March 2026), VoxCommRedesigning the Agency Value Model (Kessman/Williams, March 2026), Bain Rule of 40 (outcome-based pricing tailwind), Bain cross-system labor $100B. Convergence on "redefining headcount" with Curran/Intercom, Tatsyi/Raiffeisen (−75 people), Cherny (agents as team members). Productive tension with DORA ROI 2026's normative position "do not adopt headcount-reduction strategy" — McKinsey publicly owns the reduction in human headcount (200 roles) while reinjecting capacity into AI agents + outcome-based revenue.
Limitations: secondary source OfficeChai (not tier-1), the definition of "agent" is unspecified (autonomous vs. scripted workflows), no percentage figure for revenue already on the outcomes-based model, apparent contradiction between 200 roles cut (~0.5%) and the sweeping framing of the transformation.
To be leveraged for consulting/IT-services/agency executive committees (canonical wake-up call), HR strategy (redefining headcount), CFO pricing strategy (billable → outcome shift).
Key takeaways
Date / source.January 14, 2026, OfficeChai (Indian publication). Reporting statements by Bob Sternfels, McKinsey Global Managing Partner.
Pivotal quote."60,000, but it's 40,000 humans and 20,000 agents." ### Key figures | Metric | Value | |----------|--------| | Total McKinsey headcount (Sternfels statement) | 60,000 people | | Humans | 40,000 | | AI agents | 20,000 | | AI agents 18 months earlier | 3,000 | | Acceleration over 18 months | ×6.7 (3,000 → 20,000) | | Lilli (internal chatbot) usage | 70% of employees | | Tech roles cut (Nov. 2024) | 200 | | Severance offer (April 2024) | 9 months' salary | | Initial human/agent parity projection | 2030 | | Actual human/agent parity achieved | 2026 (18 months) | ### The business-model shift > "We're migrating pretty quickly away from, let's call it pure advisory work... moving to much more of an outcomes-based model."
Before.Pure advisory work — McKinsey sells human brainpower time (billable rates: senior associate / engagement manager / partner).
After.Outcomes-based model — McKinsey sells measurable results, aligning its incentives with client success. Implication: if 20,000 of the "60,000 people" are AI agents, the billable-hours model doesn't hold up — an agent doesn't bill an hour. Pricing must shift to outcome-based. ### Three structuring Sternfels quotes 1. "I do think we're going to be in this period where enterprises are fundamentally changing themselves." (macro framing). 2. "My latest answer to you would be 60,000, but it's 40,000 humans and 20,000 agents." (the accounting figure). 3. "Little over a year and a half ago, that was 3,000 agents and I originally thought it was going to take us to 2030 to get to one agent per human." (the acceleration). ### Connection to the watch dossier #### Convergence on "billable hours killed by AI"
Sternfels/McKinsey."migrating away from pure advisory work to outcomes-based model".
MediaPost / Joe Mandese.Billable Hours Are Dead, AI Killed Them (March 3, 2026): agency margins 30% → 10%, creatives 5× output, 4-shift outcome-based framework.
VoxComm.Redesigning the Agency Value Model (Brian Kessman + Tim Williams, March 2026): "decouple revenue and profit from staffing numbers".
Bain Rule of 40. (April 2026): outcome-based pricing as a tailwind.
→ Cross-cutting convergence in knowledge-based services: consulting + agencies + creative — all simultaneously shifting from time-and-materials to outcome-based. #### Convergence on "redefining headcount"
Tatsyi/Raiffeisen. (2026-05-05): −75 people (−8%) with deliberate reinvestment in AI agents + new products.
Cherny Sequoia. (2026-05): "few dozen PRs/day, 150 PRs in a single day record" — agents as team members.
→ Convergence: headcount accounting in the classic sense (humans only) is being abandoned by leading players. Major HR/finance question: how should we count? #### Productive tension with the DORA position
DORA ROI 2026."we strongly recommend organizations do not adopt a headcount-reduction strategy".
→ Explicit tension: DORA = prescriptive a priori norm (no headcount reduction); McKinsey = descriptive a posteriori practice (actual reduction + replacement by agents). As with Tatsyi/Raiffeisen — norm and practice diverge. To be used as a debate pivot for executive committees. #### Convergence on acceleration figures
Sternfels. 3,000 → 20,000 agents in 18 months = ×6.7.
Cherny Sequoia. (2026-05): 100% of code generated since Oct 2025, 150 PRs/day record.
Curran/Intercom. 3× R&D productivity in 16 months.
Tatsyi. AI adoption 62 → 83% in 12 months.
→ Convergence: observed adoption trajectories substantially outpace initial projections — a systematic 2024-2026 pattern. ### Limitations to flag
Secondary source. OfficeChai reports Sternfels's statements but without a full transcript or referenced official video — risk of truncated context.
Definition of "agent". unspecified: is Sternfels referring to autonomous agents like Claude/Devin or to scripted LLM tools (automated workflows, RPA, chatbots)? The figure of 20,000 is striking but its substance is unspecified.
Potential contradictions. 200 tech roles cut in Nov. 2024 = a small relative reduction (~0.5% of 40,000); is the actual movement as massive as the framing suggests?
No outcome-based figures given. Sternfels speaks of migration but gives no % of revenue already on this model versus billable.
OfficeChai = non-tier-1 source. should be cross-checked against Financial Times, Bloomberg, official McKinsey publications before reliable citation in an executive committee. ### To be leveraged for
Consulting / IT-services / agency executive committee presentations. the phrase "60,000 people, 20,000 of them are AI agents" is a canonical 2026 wake-up call.
HR / talent strategy.redefining headcount as a budgetary and organizational question.
CFO / pricing strategy. the argument "migrating away from pure advisory to outcomes-based" as a structural pricing-model shift.
FR / European framing. to be cross-referenced with French players (Wescale, Habert, Tatsyi/Raiffeisen) to position European consultancies / agencies within the trajectory.
Figures for reference. 60,000 / 40,000 / 20,000 / ×6.7 / 70% Lilli — memorable reference points to work into strategic presentations.
"60,000, but it's 40,000 humans and 20,000 agents"
— Bob Sternfels
The knowledge graph extracted from this fiche — 10 entities, 17 relations.
In this graph :Bob Sternfels · McKinsey & Company · Lilli · Migration pure advisory vers outcomes-based · Accélération projection 2030 vers 2026 · Redéfinition headcount services intellectuels · 200 postes tech supprimés nov 2024 · Offre départ 9 mois salaire avril 2024 · OfficeChai · Tension DORA-prescriptif vs McKinsey-descriptif