Albertsons doubles down on AI to sharpen operations and customer experience
Albertsons is moving past pilot mode and scaling AI across its business. During its Q3 2025 earnings call, leadership reported net sales and other revenues up just under 2%, with online sales up 21%. The company is partnering with Google, OpenAI and Databricks to accelerate execution across four fronts: digital customer experience, merchandising intelligence, workforce management and supply chain optimization.
"We're working to scale it across the enterprise to fundamentally change how we operate and how customers experience Albertsons." - CEO Susan Morris
The tech stack and where it's pointed
Albertsons highlighted an advanced cloud data infrastructure as the backbone for scaling AI solutions. That foundation is critical for near-real-time decisions, shared data products and measurable outcomes across teams. For operations leaders, this signals a shift from isolated tools to integrated decision systems tied to concrete KPIs.
- Partners: Google, OpenAI, Databricks
- Capex: $462 million in store upgrades, digital and supply chain capabilities
- Focus areas: Customer experience, merchandising, labor, supply chain
Customer experience: search that drives bigger baskets
Albertsons rolled out an Ask AI search capability that led to a 10% increase in basket utilization for customers who use it. Autonomous shopping assistants are also being deployed to create more personalized journeys. Operational implications: higher conversion, better substitution logic, and reduced friction across search, discovery and checkout.
- Prioritize product data quality and taxonomy; garbage in still equals garbage out.
- Track AOV, add-to-cart rate, substitution acceptance and search success rate.
Industry context: peers are moving too. Kroger and Sprouts are among the first to test Instacart's AI shopping agent, which suggests generative search is becoming table stakes for grocery e-commerce. See Instacart's announcement for context: Ask Instacart.
Merchandising intelligence: pricing and promo with clearer signal
Albertsons plans to give merchants AI-driven insights to optimize pricing, transform category management and support promo decisions. Expect faster scenario testing, tighter price zones and more precise elasticity reads. The goal: fewer blanket discounts, more targeted value and cleaner margins.
- Stand up a weekly price and promo review using shared demand signals and elasticity models.
- Build feedback loops: tie outcomes to unit lift, margin dollars and cannibalization.
Workforce: scheduling and support that reduces friction
The company will use generative AI to assist managers and associates with scheduling and conversational tools. Think faster shift swaps, better traffic-aligned staffing and answers to policy or task questions on demand. The productivity upside hinges on clean labor standards and simple escalation paths for exceptions.
- Start with high-variance stores and departments where demand misalignment is costly.
- Measure schedule adherence, task completion time and employee sentiment by location.
Supply chain: forecast accuracy and on-shelf availability
Albertsons plans to apply advanced analytics to improve demand forecasting accuracy, product tracking and on-shelf availability. Closer to real-time inventory visibility and better signal blending (POS, promos, local events) should reduce stockouts and waste. Expect tighter store ordering and more precise DC replenishment.
- Baseline MAPE and OTIF, then set quarterly improvement targets by category and node.
- Close the loop: compare forecast vs. actuals and annotate drivers (promo, weather, vendor).
The investment thesis
"From a digital and technology perspective, we further invested in AI and digital transformation to create structural cost advantages, deepen customer loyalty and unlock new profit pools," said CFO Sharon McCollam. The message is clear: AI spend is being justified by durable cost and revenue levers, not hype.
What operations leaders can do next
- Anchor on a few KPIs per domain: CX (AOV, conversion), Merch (margin dollars, promo ROI), Labor (hours per task, shrink), Supply chain (MAPE, OSA, OTIF).
- Start with decision loops you run weekly: price/promo, labor scheduling, store ordering, DC slotting.
- Operationalize data quality: product, store inventory, promo calendars and labor standards must be owned and audited.
- Set human-in-the-loop boundaries: define when AI recommends vs. when it auto-applies changes, with clear rollback plans.
- Integrate with core systems: WMS, TMS, ERP, and workforce tools should write back outcomes for continuous learning.
- Train front lines: short, scenario-based enablement beats long docs. Measure adoption, not just availability.
Why this matters now
Grocery margins are tight and demand patterns keep shifting. AI that boosts forecast accuracy, trims labor waste and lifts digital conversion compounds fast at scale. For most teams, the limiter isn't models-it's clean data, clear ownership and tight feedback cycles.
If you're comparing retail AI stacks, Google Cloud's retail offerings are a good reference point: Google Cloud for Retail. And if you need structured upskilling aligned to leading AI vendors, see this curated list: AI courses by leading companies.
Your membership also unlocks:
