AI

How to protect razor-thin retail margins with AI

Retail margins are on the verge of a complete meltdown. Yet, retailers using AI are seeing the opposite trend. Knowing how and where to invest in AI is critical to ensure the investment doesn’t outpace the return.

Generated by AI/Adobe Stock

March 27, 2026 by Leigh Helsel — Partner, Retail Lead, Centric Consulting

Summarize with

Retail margins are on the verge of a complete meltdown.

Grocery net margins have fallen to largely under 1.5%, matching the 2008 financial crisis levels as global tariffs, ongoing omnichannel complexity, and escalating price wars squeeze profitability from every angle.

Yet, retailers using AI are seeing the opposite trend: 89% report increased revenue and 95% report decreased operating costs, according to NVIDIA's State of AI in Retail and CPG: 2026 Trends report.

For executives already operating on razor-thin margins, knowing how and where to invest in AI is critical to ensure the investment doesn't outpace the return.

Market forces compounding the pressure

Retail executives know where their margins are disappearing, with operational realities that compound one another:

Returns have skyrocketed. In 2025, returns hit $845 billion, or 16.9% of total sales, with 82% of consumers now expecting free returns, according to the National Retail Federation's 2025 Retail Returns Landscape report. Each return triggers reverse logistics costs, restocking expenses, and potential inventory write-downs, all of which compound the margin impact.

Price wars are intensifying. The competitive environment forces constant repricing and deep discounts that erode profitability, especially with new sales channels like TikTok direct-to-consumer shopping exploding.

Fulfillment costs are ballooning. Fulfillment-related costs continue to grow, with increasing pressure on the speed of delivery and last-mile logistics taking an increasing share of operational spend. As customer expectations for speed and flexibility grow, these costs threaten to outpace efficiency initiatives.

Where AI delivers the highest ROI

These use cases consistently deliver measurable retail margin protection and operating profit improvement:

Inventory and demand forecasting: AI forecasting enables the use of much more data and removes recency bias, leading to reduced lost sales due to unavailability. One machine learning model I co-led the development of could forecast full demand from a 12-store test, determining whether to buy 5,000 or 15,000 units for a chain-wide rollout, eliminating the costly guesswork between overstock and stockouts.

Workforce scheduling: Labor represents one of the largest controllable expenses for retailers. AI-driven scheduling dynamically aligns staffing with predicted traffic and demand, ensuring the right associates are in the right place at the right time to handle the ebbs and flows of customer traffic. This also aligns with scheduling in fulfillment centers.

Dynamic pricing: Rather than blanket markdowns or static pricing, AI enables targeted adjustments based on inventory levels, demand signals, location factors, and competitive benchmarks. This isn't new. Airlines and hotels have used it for years, and retailers have deployed AI-driven personalized offers since at least 2010, when I saw call centers using dynamic offers based on customer profiles. What's changed is that now AI can systematically test pricing across markets in real time, learning what customers will pay without risking brand perception or losing volume.

Loss prevention: Rather than focusing solely on surveillance, the most effective AI applications in loss prevention analyze data on loss causes by location and product to inform tailored practices and stocking habits that reduce opportunities for theft. Much of shrink is product "lost in the system," or inventory that can be better tracked through AI combined with technologies like RFID.

Protect margins while investing

McKinsey estimates Generative AI could unlock $240-$390 billion in value for retail, equivalent to a 1.2 to 1.9 percentage point margin increase. Yet fewer than one in five companies track AI KPIs, which is the primary reason they fail to see bottom-line impact.

Start with operations where the math is clearest: inventory forecasting and workforce scheduling deliver measurable results within months. Avoid the temptation to lead with customer-facing tools only and focus instead on back-end operations where AI can compound savings quarter after quarter through proper governance.

Most importantly, define success metrics before implementation. Track reduction in stockouts, labor cost savings and margin improvements. The retailers realizing results with AI are those running the most disciplined implementations with clear ROI targets and the patience to scale what works.

Pitfalls that erode margins

Research found 42% of companies abandon the majority of AI initiatives before reaching production.

Common pitfalls include:

Integration costs blindside you: The scaling and implementation costs can be difficult to predict, but need to be monitored for efficiency, such as how often models need to be tuned or associated cloud costs.
Data quality ignored: AI is only as good as the data feeding it, which can be messy and misunderstood in the new ways AI can provide value.
Change management skipped: AI transformations fail due to inadequate change management, not technology issues. Many people trust experience and intuition over new models that may not align to their ways of working. Leaders need to be clear on the problem AI is solving, who owns it, and what has to change for value to show up.

Despite these challenges, when retailers see results, they speak for themselves.

Bottom line

Major retailers are already seeing bottom-line impact.

Walmart reported 4.8% revenue growth, with executives citing generative AI as a key driver of their 21% e-commerce growth. Amazon's Rufus AI shopping assistant is projected to generate $10 billion in incremental annual sales, with customers who engage with Rufus 60% more likely to complete a purchase.

The margin crisis that's squeezing retailers won't ease on its own. Opportunities to use new processes and the enormous data companies have been sitting on have been ripe for transformation.

The high percentage of retailers already seeing revenue gains from AI proves there's a path forward. The difference is in execution: invest where the ROI is proven, measure relentlessly, and build from results rather than hype.

About Leigh Helsel

Leigh has more than 10 years of business consulting experience and 20 years of corporate leadership experience. She thrives in working with clients to help build scalable solutions profitably. Her experience spans a wide range of Fortune 500 organizations across multiple industries. At Centric Consulting, she prioritizes growing client relationships, business development, marketing, and management of the Columbus office. She brings her experience from retail, financial services, manufacturing and logistics to clients to help look at complex opportunities through a unique lens.

Connect with Leigh: