Macy's AI Bet: Virtual Try-On Quintuples Spend Per Session

The 168-year-old department store is deploying machine learning across merchandising, demand forecasting, and customer-facing tools — and one metric stands out.

Written by Simon Reed · Bureau Finance · Jun 2, 2026

The Number That Matters

Macy's virtual try-on assistant quintuples spending per session. That is the figure the company wants you to notice — and, unusually, it is also the figure worth noticing. A 5x lift in session spend is not a vanity engagement metric; it maps directly to revenue per visitor, which is the variable most under pressure in physical and hybrid retail.

The caveat: Macy's has not disclosed what share of total sessions use the tool, nor the absolute revenue base it applies to. A 5x multiplier on a thin slice of traffic is a different story from a 5x multiplier at meaningful penetration. That denominator matters and has not been provided.

Demand Forecasting: The Quieter Bet

Less headline-friendly but arguably more consequential is the machine learning deployment in demand forecasting. Inventory management is where department stores bleed margin — overstock drives markdowns, understock drives lost sales. If Macy's ML models are improving forecast accuracy, the operating leverage shows up in gross margin, not in a press release.

Macy's has not published forecast-accuracy figures or quantified the inventory impact. Investors should watch gross margin trajectory in upcoming quarters as the more honest signal of whether this layer is working.

Manager Training and Labour Productivity

The extension of AI into manager training is the least-discussed element and potentially the most structurally significant. Labour is Macy's second-largest cost line after cost of goods. Tools that compress training time, reduce turnover costs, or improve floor-level decision-making compound across hundreds of stores. The economics are slow to surface in reported numbers but durable once they do.

Turnaround Context

Macy's has been contracting — closing underperforming stores, rationalising its nameplate portfolio, and defending against the structural shift away from mid-market department stores. The AI rollout is being framed as a component of that turnaround, not a standalone initiative.

The framing is reasonable. AI that improves conversion, tightens inventory, and reduces labour costs addresses three of the four main pressure points simultaneously. The fourth — traffic — is harder to solve with software.

What to Watch

The 5x session-spend figure needs a penetration rate to be fully interpretable. Gross margin in the next two to three earnings reports will indicate whether demand forecasting is delivering. And any disclosure on training-cost reduction would sharpen the labour-productivity thesis. Until those numbers surface, the AI story at Macy's is directionally credible but not yet quantifiably confirmed at the P&L level.

Key takeaways

Macy's virtual try-on tool produces 5x spending per session — the sharpest single metric in the company's AI rollout.
Machine learning is being applied to demand forecasting, which directly affects inventory costs and markdown rates.
AI-assisted manager training points to labour-productivity ambitions beyond the customer-facing layer.
The deployment spans the full business, not a pilot segment — scale is the operative word.
For a 168-year-old retailer under sustained pressure, AI adoption is being positioned as a structural turnaround lever, not a feature.

FAQ

What does the 5x spending figure actually mean?

Macy's reports that customers using its virtual try-on assistant spend five times more per session than those who do not. It is a conversion-quality metric — it measures spend intensity, not total revenue contribution, which depends on how widely the tool is used across all sessions.

How does AI in demand forecasting affect Macy's financials?

Better demand forecasting reduces excess inventory, which lowers the need for markdowns and protects gross margin. The impact is not immediate but accumulates across buying cycles. Gross margin is the line item to monitor.

Is this AI deployment a pilot or company-wide?

According to Fortune's reporting, Macy's is deploying AI across its entire business — not a limited pilot. That scope increases both the potential upside and the execution risk.

What are the main risks to the turnaround thesis?

Traffic decline is the variable AI cannot easily fix. If footfall and digital visits continue to fall, higher spend-per-session may not offset volume loss. Execution risk across a large, legacy retail infrastructure is also non-trivial.

Citations

An AI overhaul at Macy's is fueling the 168-year-old retailer's turnaroundMacy's virtual try-on assistant quintuples spending per session; AI is deployed across demand forecasting and manager training.
Fortune RSS FeedBureau research source: Fortune
An AI overhaul at Macy's is fueling the 168-year-old retailer's turnaroundMacy's is deploying AI across its entire business as part of a broader turnaround strategy.