Why Retail CEOs Are Failing the AI Test

The retail CEO exodus isn't a leadership crisis. It's an overdue reckoning.

For years, too many retail CEOs got rewarded for engineering optics instead of building durable advantage. They leaned on promotions, store openings, buybacks, cost cuts, and polished investor narratives while avoiding the harder work of fixing merchandising, sharpening differentiation, modernizing operations, and rebuilding customer trust.

This worked when the environment was forgiving.

It doesn't work when growth is harder to find, consumers are more selective, and every weak proposition gets instantly exposed. CEO turnover at high-performing S&P 500 companies jumped from 7% in 2024 to 12% in 2025, nearly matching the 14% rate among bottom-quartile performers. Retail companies reported 51 CEO exits in 2025, with voluntary turnover in the sector reaching 26.7%, the highest among major industries.

The crisis isn't CEOs leaving. The crisis is too many should have been challenged earlier.

The Customer Does Not Shop the Narrative

Boards spent years backing compelling narratives while missing what the business needed.

Kroger is a clear example. The board rallied around a story where scale would solve the problem. The $25 billion Albertsons deal was positioned as the move to give Kroger leverage to compete with Walmart, Amazon, Costco, and discounters.

But the business needed something more basic. Sharper pricing, better store execution, stronger fresh, and a clearer answer for a pressured middle-income shopper.

After the merger collapsed, Kroger's focus moved exactly there. The company cut prices on thousands of items, leaned harder into promotions and private label, and reinvested savings into sharper everyday prices and better service. The messaging shifted toward fundamentals. Competitive prices, strong fresh, well-run stores.

The narrative the board rallied around was elegant and capital-markets friendly. We need transformative scale.

The business reality was less glamorous. We need to win the weekly shop.

Those aren't the same thing. When a grocery retailer loses sight of this, the customer notices price, quality, in-stock position, and store experience long before they care about the brilliance of your merger thesis.

What the Wrong CEO Does on Tuesday Morning

The wrong CEO spends Tuesday doing symbolic leadership instead of commercial leadership.

They're in the branding deck, not the business.

They spend time polishing the story for the board, refining transformation language, rehearsing the earnings call message, debating org-chart abstractions, approving another consultant's framework, or reviewing a glossy innovation initiative. Sounds strategic but doesn't change what the customer experiences this week.

Meanwhile, the right questions sit there unanswered.

Why are in-stocks down in key categories? Why is produce quality inconsistent by region? Why are promo items not converting? Why is labor not lining up with peak traffic? Why are baskets shrinking? Why are store managers drowning in complexity?

A narrative-driven CEO spends Tuesday managing perception upward. An operator spends Tuesday removing friction downward.

At Starbucks, Brian Niccol's turnaround has been tied to operational work. Simplifying the menu, adding staffing, and fixing mobile-order congestion because the in-store experience had become too chaotic.

At Nike, the work under Elliott Hill has centered less on grand positioning and more on repairing wholesale relationships, restoring innovation, cleaning up inventory, and improving execution in markets like China.

The best retail CEOs have an unromantic Tuesday. They force clarity around a few brutal questions. What's broken? Where's the friction? Who owns it? How fast do we fix it? Will the customer notice by next week?

Where AI Fluency Actually Shows Up

The AI-fluent retail CEO asks where the machine creates drag in the customer journey, and what AI fixes this week with measurable impact.

The narrative-driven CEO asks: "What's our AI story?"

The difference is clear.

On Tuesday morning, the AI-fluent CEO isn't starting with the model. They're starting with the choke point.

Which stores are losing margin because forecasting is wrong on perishables? Which promotions are creating traffic but not basket lift? Where are we overstaffed, understaffed, or scheduling against the wrong demand pattern? Which digital journeys have the highest abandonment, and what intervention would recover conversion fastest?

The narrative-driven CEO thinks AI is a capability to announce. The AI-fluent CEO thinks AI is a decision system to embed.

The question they ask when others miss it. Which repetitive, high-frequency decisions in our business are still being made inconsistently by humans, and how fast do we turn them into better, assisted decisions?

In retail, that might mean price recommendations, assortment tuning, labor scheduling, replenishment, promo targeting, service recovery, fraud detection, or identifying which stores are drifting operationally before the numbers make it obvious.

About half of sales growth between 2022 and 2024 is attributable to AI among retailers who adopted these technologies. Profit growth among retailers using AI was more than twice what it was among retailers who didn't use these technologies.

Yet having the tech doesn't mean knowing how to deploy it.

The Promotional Execution Gap

One of the clearest places human inconsistency shows up is promotional execution at shelf level.

The customer sees an item in the flyer, the app, or the email at one price, walks into the store, and hits confusion.

The shelf tag is wrong. The item isn't where it should be. The promotion didn't load properly to loyalty. The substitute online makes no sense. The feature item is already out of stock by mid-morning. The cashier has to override the price.

Now the customer is irritated, the employee is improvising, and the retailer calls it a small execution issue.

It's not small.

This is exactly the kind of inconsistency customers feel right away, and it happens because too much of the chain still relies on fragmented human execution across pricing, merchandising, replenishment, store ops, and digital systems.

The AI-fluent CEO asks why we're still treating promo breakdowns as isolated store mistakes instead of a pattern we should be predicting and preventing.

AI identifies which promotions are most likely to fail before they do by combining historical lift, store inventory, labor availability, shelf-change volume, loyalty load errors, substitution rates, past execution issues, and complaint data. It flags Store 184 is highly likely to miss a Wednesday feature on strawberries, not enough inventory is allocated, the digital ad is live, and customer disappointment will spike by 11 a.m.

This gives the operator time to intervene before the customer feels the failure.

Most CEOs miss this because they think AI at too high an altitude. They talk about personalization, copilots, and innovation labs. Meanwhile, the customer is standing in aisle 6 wondering why the promoted product is gone, mislabeled, or never available at the promised price.

The Real Organizational Tell

The biggest organizational tell isn't whether a company hired a Chief AI Officer. It's whether core business leaders are expected to use AI to improve specific operating decisions, with accountability, metrics, and follow-through.

The box-checking CEO creates AI as a staff function. The real one makes AI a line-operating discipline.

In the weak version, there's an AI leader, a lab, pilots, demos, a quarterly update for the board, and language about transformation. But the merchant, store ops leader, supply chain head, digital lead, pricing lead, and customer service leader are still running the business the same way they did before.

In companies with AI fluency, AI shows up in the business review. In companies without it, AI shows up in the innovation update.

The difference is massive.

Once AI is in the business review, it has owners, metrics, deadlines, and consequences. It stops being inspirational and starts being operational.

The CEO who gets it doesn't measure success by number of pilots. They measure success by reduction in decision latency, increase in forecast accuracy, fewer stockouts, better labor productivity, higher promo conversion, lower service friction, or improved basket outcomes.

Survey respondents identified three clear goals for agentic AI in retail: increased process speed and efficiency (57%), enhanced customer experience and personalization (40%), and improved decision-making with real-time data (40%).

Merchant Judgment With Telemetry

Merchandising instinct, product judgment, and brand feel are still the source of differentiation. AI doesn't replace this. It sharpens it, stress-tests it, and helps scale it more consistently.

A brand like President's Choice wasn't built by an algorithm. It was built through merchant judgment, product conviction, and a real understanding of what would resonate with customers in the aisle.

What changes now is this. The old model allowed instinct to go unchallenged for too long. Today, the environment is faster, channels are more fragmented, feedback loops are immediate, and execution is more complex. Instinct still matters, but it has to survive contact with data, operational reality, and speed.

The human decides what the brand should stand for. AI helps reveal where the brand promise is breaking down in the real world.

The human senses what product has emotional resonance with the customer. AI helps identify where demand is emerging, where conversion is stalling, and where assortment needs to be tuned by region, format, or channel.

The merchant says, "This product, this pack size, this flavor, this price architecture feels right for our customer." AI says, "Here's how that instinct is performing by cohort, basket attachment, repeat rate, substitution risk, and margin."

This isn't the death of instinct. This is instinct being upgraded from art alone to art with instrumentation.

The risk isn't AI killing taste. The risk is weak leaders using AI as an excuse to abandon taste.

Operational Precision in Service of Brand Soul

The next bad hire won't look like the old bad hire. The old mistake was hiring the polished storyteller who lacked operating depth. The new mistake will be hiring the AI-literate optimizer who can make the machine run better but has no feel for why the customer cared about the brand in the first place.

This person improves conversion and still weakens the business.

Boards need to stop asking only if this person is fluent in AI. They need to ask if they know what should never be optimized away.

In a retail business, the soul usually lives in a few places. The product standards the company refuses to compromise on, the emotional logic of the brand, the curation instinct behind assortment, the way the store or digital experience feels when it's working, and the understanding of which details customers may never articulate but absolutely notice.

A CEO who gets this will use AI to strengthen these things, not flatten them.

Use AI to improve forecasting, but don't let it commoditize the assortment. Use AI to improve personalization, but don't let it cheapen the brand voice. Use AI to improve promo execution, but don't train the customer to think you only win on price.

The right CEO doesn't put the merchants on one side and the data people on the other. They create a culture where brand instinct and operating intelligence sharpen each other.

Per Bank at Loblaw is one of the better current examples. At Loblaw, AI isn't replacing the merchant idea. It's being layered onto an ecosystem with soul. President's Choice, banner differentiation, and loyalty. The company tied performance to differentiated value, quality, service, convenience, strong multicultural and Canadian assortments, innovative PC products, and personalized PC Optimum offers.

The technology appears to be in service of the retail proposition. Help Canadians find the right products, shop more efficiently, and receive more relevant offers, while preserving what already makes Loblaw's ecosystem distinctive.

The Interview Question That Reveals Everything

If you're a retail board staring at CEO succession, trying to avoid both the old mistake and the new mistake, ask this.

"Tell us about a time you improved the economics and operating discipline of a retail business without stripping out the very thing customers loved about it. What did you change, what did you refuse to change, and how did you know the difference?"

This question forces candidates to prove they hold both truths at once.

The future retail CEO isn't an AI CEO or a merchant CEO. It's a CEO who knows which parts of the business should become more machine-like and which parts must remain unmistakably human.

The goal isn't operational precision instead of brand soul. It's operational precision in service of brand soul.