Deloitte put it plainly this spring. AI is now a standing topic in the boardroom for the retail and consumer products industry. The capital markets noticed before the P&L did. While company financials are yet to reflect the transformative potential of AI, capital markets are rewarding retail and consumer products companies for making strategic AI investments and announcements.
Read that twice. You are being paid for the press release, not the proof.
That is the most dangerous incentive structure in retail right now. And it is exactly why the memo you write to your board this quarter matters more than the roadmap you present next quarter.
The optimism memo is killing you
Most retail CIOs are about to write the wrong memo. It will open with agentic AI, cite a McKinsey cost-reduction range, list pilots in merchandising and contact centers, and ask for more budget. The board will nod. The press release will publish. The stock might even tick.
Then the variance report lands.
Here is the reality your CFO already smells. Leading retail and consumer products brands are training tens of thousands of employees on AI, embedding AI agents in contact centers, logistics, merchandising, and finance, and announcing partnerships across the technology landscape. Yet for all the transformative promise, the P&L story is not yet taking shape. The reality is, while there are small wins and bright spots, the impact is muted in the noise of broader enterprise financials.
The spending is not the problem. The spending is happening. A significant proportion of retail and banking executives expect to spend up to 20% more of their 2026 tech budgets on AI and machine learning, according to a Bain survey. The problem is conversion. The average enterprise runs 14 AI projects simultaneously, up from 8 in 2023, though most organizations report that fewer than half are delivering measurable business value.
So write a different memo. Write the Cost of Doing Nothing memo.
The CODN memo, in four parts
The board does not need another slide on potential. It needs a number on standing still. CODN is not a rhetorical device. It is a line item. Build it.
Part one: the CODN number. Quantify the margin you forfeit by waiting. The research already gives you a defensible anchor. Most retail organizations are at Stage 2 of AI maturity, with proven pilot ROI but lacking the business case for enterprise investment. IMRG data shows a 2.3 percentage point market share loss over 24 months for retailers without AI personalization. Translate that to your revenue base. Put a dollar figure on inaction and watch the conversation change. The point of CODN is to make “wait and see” the expensive option, because it is.
Part two: three production use cases with full cost accounting. Not pilots. Production. Model two to three use cases with full cost accounting: include data infrastructure (shared cost across use cases), implementation, ongoing MLOps, change management, and governance. If your memo hides the MLOps and governance line, it is fiction.
Part three: the self-funding ladder. Retail P&Ls cannot absorb a single moonshot. They can absorb a staircase. The solution is staged investment: start with quick-win use cases that deliver payback within one quarter, then reinvest returns into larger initiatives. This self-funding model avoids the need for a single large capital commitment that retail P&Ls cannot absorb. The sequencing is knowable. Quick-win use cases like recommendations and chatbots break even in 8 to 12 weeks. Demand forecasting and inventory optimization reach payback in 4 to 6 months. Dynamic pricing requires 6 to 9 months due to governance setup and model tuning.
Part four: named owners and hard limits. This is where most memos go quiet, and where the money leaks. Uncontrolled AI usage, nicknamed “tokenmaxxing,” has produced real cost overruns, including one company that reportedly spent $500 million in a single month after failing to set usage limits. Assign a name to every dollar and a ceiling to every workflow.
Why this memo wins the cycle
The failure rate is not a technology story. Global enterprise AI spending is projected to reach $665 billion in 2026, yet three out of four AI deployments fail to achieve their projected return on investment. The gap between AI spending and AI value has become the defining strategic challenge facing enterprise technology leaders. No model upgrade closes that gap. Governance does.
The winners already figured this out. The dynamic is shifting as organizations prioritize targeted use cases and lay the groundwork for scalability and ROI. Cross-functional steering committees and specialized task forces are emerging as critical building blocks to identify, prioritize, and align use cases to enterprise goals. Eighty-three percent of IT leaders confirmed their organizations either have such structures in place or are planning them within the year.
The board memo you write this quarter is not a funding request. It is a positioning statement. It says: we measure inaction, we ship to production, we self-fund, and we own the spend.
In 2026, ambition is free and announcements are cheap. Governed value is the moat. Write the CODN memo before the budget locks, not after the analysts ask why the P&L still has not moved.