The tools work. The people don't change. Workflows stay the same, managers route around the technology, and the ROI never shows up on a P&L. AI adoption is two-thirds behavioral and one-third technical. Every engagement starts with the two-thirds. Ask most consultants what the ROI is and you get a productivity estimate. Here is the real answer: workflow redesign frees capacity, and redeploying that capacity is the ROI. Use Rate measures it. Redeploy or it isn't ROI.
Before you spend another dollar on AI, find out why the last dollar didn't work. The 2/3 Rule System has Eight Disciplines. The audit scores every one of them Green or Red. A Red you can defend is worth more than a Green you are hoping for. You start with the first one you're failing, because the ones after it depend on it. The audit names that one and sets your first Use Rate baseline. What it includes: a 90-minute scoping session with the CEO. Eight conversations across the functions that bought the tools. Observation of three live workflows where AI was supposed to land. A first Use Rate baseline. A written read-out with the Readiness Scorecard, the first discipline to fix, and the one workflow to redesign first. A 90-minute CEO read-out with a decision on whether to proceed, and where. Two to three weeks. Starts at $12,500. Larger scopes are priced after the fit call. Not tool selection, not a training program, not an IT review.
You bought Copilot. You bought ChatGPT Enterprise. You ran the training. Nothing changed. This is the part that makes it stick. A ground-level redesign of how your people actually work, built from what the audit found. Decision rights that say who uses the tool for what. Friction removed from the workflow so the tool is the easy path, not the extra step. Reinforcement that managers run, not HR. Milestones scored on Use Rate, not on logins. Haven't rolled out yet? Build this before the first license goes live. It is cheaper to design the work once than to retrain a habit twice. Typically eight to twelve weeks. Scoped after the audit.
The audit shows you the gap. The operating model closes it. Then comes the part nobody sells: staying with it. Two 90-minute working sessions a month with me. Direct access between them. A written Use Rate read-out every month: where AI touched the work, how many hours came back, and where those hours went. For CEOs, COOs, and CHROs who want a partner who has run the business, not one who briefs it. I ran a company through fast growth. I know what it costs when a change doesn't stick, and I know the CEO is the one who carries it. Monthly retainer, scoped after the audit.
I speak on The 2/3 Rule and what mid-market companies get wrong about AI adoption. Executive sessions, leadership offsites, and industry events. Topics: Why AI training doesn't work, and what does. Adopt does not equal adapt: the behavioral economics of AI adoption. What PE-backed companies miss about AI ROI. The 2/3 Rule in practice. Most sessions run 45 to 90 minutes and end with your team scoring their own eight disciplines.
1. Own Your Disruption. The Ownership Spectrum. "They're replacing us" becomes "I'm upgrading myself." 2. Define the Outcome Before the Tool. The CFO Test. "What can AI do?" becomes "What do we need done?" 3. Sequence for Momentum. The FSR Matrix. "Let's pilot everything" becomes "Let's win first, then expand."
4. Invest the Cognitive Surplus. The Surplus Audit. "How many people are using AI?" becomes "What are they doing with the time they got back?" Redeploy or it isn't ROI. 5. Build Human-AI Workflows. The Judgment Line. "AI vs. people" becomes "AI + people = capacity." 6. Speak the Language of the Business. The Translation Test. "We need buy-in" becomes "Here's the business case." 7. Create Multiplier Systems. The Multiplier Test, Five Gates. "Project" becomes "Platform."
8. Sharpen the Workforce. The Quarterly Capability Audit. "We trained on AI" becomes "We evolve with AI." Four questions, every ninety days, run by someone with authority to act. The organizations that win at AI don't outspend the competition. They out-behave them. From OutBehave: The 2/3 Rule for AI-Ready Organizations. The audit scores all eight. The operating model runs them. Advisory keeps them running.