Only 45% of Retailers Scale AI Despite Threefold Earnings Advantage
In this article (9)
Only 45 percent of retailers are scaling artificial intelligence across their operations, even as businesses committing heavily to the technology generate three times higher revenue and earnings growth.
According to research highlighted this week by Boston Consulting Group managing director and partner Charlie Reid, operators treating artificial intelligence as a core discipline outpace rivals by triple on top-line sales and EBITDAR. Yet widespread experimentation across scattered pilots leaves the majority of commercial operators unable to convert software trials into balance sheet gains.
Fragmented initiatives and inadequate organisational changes continue to limit companies from translating investments into measurable financial returns, Boston Consulting Group noted in November 2026 findings.
Where Merchandising Delivers Fast Payback
Capital is flowing primarily into core store and supply chain operations rather than back-office corporate support. Retailers are largely bypassing enterprise applications across human resources, procurement, and finance because those departments offer lower immediate financial return.
Merchandising operations are providing the quickest path to measurable payback. Adjusting product assortment, pricing, promotion schedules, and vendor negotiations through artificial intelligence tools generates cash flow that operators use to fund broader technology rollouts. Getting shelf assortment right yields up to a 5 percent increase in like-for-like sales.
The Friction Behind Operating Models
Fragmented rollouts and weak internal adoption explain why most deployments stall at the testing phase. Boston Consulting Group measures technology implementation through what it terms a 10-20-70 operating breakdown: algorithms account for 10 percent of the transformation, data and systems represent 20 percent, while people, operating processes, and institutional structure dictate the remaining 70 percent.
“20% by the data and technology and then 70% by the organization, the processes um and and and and the people in the organization.””
Tools fail when buyers and floor staff lack clear authority or incentive to execute automated guidance. An algorithmic recommendation creates zero value if a category merchandiser overrides the system or if store staff fail to reset physical shelf space. Executive teams routinely hunt for novel software without identifying the exact financial metric they need to alter.
“So, at BCG, we call it the 10 2070 rule. 10% of the AI transformation is driven by the algorithm. 20% by the data and technology and then 70% by the organization, the processes um and and and and the people in the organization.”
Competing for Automated Buyers
Consumer interaction is shifting toward agentic commerce, where automated systems handle product discovery, price comparisons, and digital checkout tasks. Shoppers currently manage friction across physical aisles and mobile applications, frequently abandoning journeys when items show out of stock at checkout.
Software agents compress that multi-step search into direct prompts, assembling tailored baskets against specific household dietary rules and spending limits. Over extended periods, these systems track household consumption patterns and reorder domestic staples before pantries run empty.
That shift alters digital distribution outside merchant properties. Consumers searching through external platforms like Google and ChatGPT rely on software agents that scrape product attributes, ingredient registries, and live inventory across competing web domains simultaneously. Retailers must structure open, accurate product catalogues for algorithmic inspection rather than relying solely on visual merchandising to win human attention.
Pruning Unfocused Software Trials
Commercial operators face pressure to demonstrate cash returns on software capital expenditures. Treating machine learning as an exploratory technical exercise without strict hurdle rates has tied up working capital across fragmented vendor contracts.
Survival across competitive consumer markets requires pruning low-yield software trials to protect core operating margins. Retailers that narrow their engineering resources to inventory visibility, predictive replenishment, and machine-readable data feeds will command distribution as autonomous purchasing tools take over routine household spending.
Questions & Answers
Q.What financial advantages do retailers gain by scaling artificial intelligence?
What financial advantages do retailers gain by scaling artificial intelligence?
Retailers that treat artificial intelligence as a core discipline achieve three times higher revenue and EBITDAR growth compared to their rivals. Within merchandising, using automated tools to refine physical shelf assortment alone generates up to a 5 per cent increase in like-for-like sales.
Q.Why are most retail artificial intelligence deployments failing to scale?
Why are most retail artificial intelligence deployments failing to scale?
Deployments typically stall because of fragmented trials and inadequate organizational changes. People, operating processes, and institutional structure dictate 70 per cent of transformation success, meaning algorithms fail to create value if store staff and buyers lack the authority or incentive to follow automated guidance.
Q.Which business functions deliver the fastest payback on software investments?
Which business functions deliver the fastest payback on software investments?
Merchandising operations offer the quickest return through algorithmic adjustments to pricing, promotion schedules, vendor negotiations, and product assortments. Retailers are directing capital into these core store and supply chain functions whilst bypassing corporate departments such as human resources and finance that offer lower immediate gains.
Q.How is agentic commerce altering digital distribution for retailers?
How is agentic commerce altering digital distribution for retailers?
Autonomous software agents are increasingly conducting product discovery, price comparisons, and automated household replenishment on external platforms. To capture these sales, retailers must build open, machine-readable product catalogues for algorithmic inspection rather than depending exclusively on visual merchandising created for human shoppers.
Reader pulse
Where should retailers prioritise their artificial intelligence budgets?
19,291 votes so far