reference

Gen AI in Corporate Functions: Looking Beyond Efficiency Gains

McKinsey CXO survey (2024) of 276 senior leaders across finance, HR, IT, customer care, and legal. Identifies the "diffuse capacity" problem where AI automates parts of roles rather than whole jobs, making it difficult to redeploy freed capacity. Argues effectiveness offers more value than efficiency.

Source Information

  • Publisher: McKinsey & Company
  • Date: October 23, 2024
  • Methodology: Annual CXO survey, n=276 senior leaders
  • Functions: Finance, HR, IT, Customer Care, Legal
  • Industries: 18 industries across North America and Europe
  • Company sizes: 20% <$1B revenue, 46% $1-10B, remainder >$10B

Key Findings

The Diffuse Capacity Problem

"It can be hard to turn small time savings into meaningful cost savings. Today's gen AI systems are capable of automating parts of roles, rather than whole jobs, making it difficult for companies to redeploy people freed up by automation."

This is the central insight: AI saves 15 minutes here, an hour there—fragmented across roles—but these savings cannot be easily aggregated into deployable capacity.

Effectiveness Over Efficiency

McKinsey argues that "effectiveness offers much more potential value" than efficiency:

  • Improving forecast accuracy
  • Understanding markets more deeply
  • Optimizing capital allocation

These unlock "truly significant cost savings and growth opportunities" beyond incremental time savings.

Top Use Cases in Finance (by CFO deployment)

Use Case Deployment Rate
Cost analytics 47%
Optimizing accounts payable approvals 44%
Fraud prevention checks 44%

Implications for Engagement Design

The diffuse capacity problem directly informs Engagement Model: Concurrent Discovery, Coordinated Execution:

  1. Capacity Crystallization as explicit phase gate—freed capacity must be aggregated before it can fuel expansion
  2. Role consolidation methods—combining partial-role savings across positions
  3. Governance against reabsorption—preventing diffuse savings from being absorbed back into adjacent tasks
  • in engagement model
  • in glossary (terms.yml)