In the US, 74% supply procurement executives say tariffs are driving a significant shift in strategies, because they disrupt every stage of inventory management from procurement to distribution

Retailers are ramping up investment in AI to navigate an uneven economy and tariff volatility, but external pressures are exposing limits to how data is collected, governed, and used. While momentum is building, many retail organizations are unprepared to scale AI across their operations.

Chief data officers (CDOs) hold the key to unlocking enterprise-wide value with AI, but they must treat data modernization as a growth strategy, while improving data governance and uniting insights into customers and supply chains.

Consumer confidence has rebounded somewhat after hitting a 12-year low earlier this year, but tariff uncertainty has clouded financial forecasts in retail. In the US, 74% supply procurement executives say tariffs are driving a significant shift in strategies, because they disrupt every stage of inventory management from procurement to distribution.

Like other industries, retail is going all in on AI, particularly generative AI, to navigate market uncertainty, improve supply agility, and capitalize on customer preferences. Fewer than half (46%) of retailers say they’ve fully deployed genAI in some part of their business, but 90% of retailers consider scaling genAI to be “very” or “extremely important” in the coming year, according to EXL’s Enterprise AI study.

Despite the ambition, data threatens the industry’s ability to unlock enterprise-wide value with genAI. Data silos across disparate systems prevent retailers from personalizing offerings at scale, forecasting real-time inventory needs, and reacting to market shocks that impact consumers or supply chains. Nearly 70% of retailers mistrust the quality and accessibility of their data. Only 32% trust the transparency of their models, and just 42% believe their data is consistent.

The industry’s CDOs are at the center of this challenge. Today’s savvy CDO needs a new playbook for data management and governance to enable AI throughout end-to-end workflows. The value is clear for companies that can drive AI at scale. Retailers deploying AI report 21% higher revenue and a 20% reduction in costs, according to our study.

Current state: GenAI adoption and early wins

Retailers are actively rolling out genAI, but successful pilot programs and the ability to scale will determine the industry’s winners and losers. Slightly more than half of retailers (52%) have deployed genAI solutions to some degree in customer care and e-commerce, according to EXL’s recent study. Nearly 60% are piloting genAI in customer experience, and another 52% are either actively piloting or exploring genAI to optimize supply chains and manage inventories.

Early wins with genAI are generating positive outcomes for about half the industry. 54% improved their ability to launch new products and services. 46% reported improvements to areas such as customer experience, operational efficiency, and existing products. For an industry built on finding the next “must have” products for consumers, current adoption trends foreshadow a doubling down on genAI across the enterprise.

Why data challenges are holding retail back

A core objective to recent technological transformation in retail has been to make products more seamless to buy in a digital environment, like an app or online store. Equally important is the effort to provide personalized experiences that improve customer loyalty and increase sales.

Success so far is mixed. For example, retailers want to recommend new products based on past purchases, but many suggestions are still primarily based on matching colors or brands as opposed to individual style, context, and intent. Virtual try-ons are also gaining prominence and are increasingly important given reductions to in-store inventory and online buying models, but this capability requires robust data governance to address privacy concerns, as well as solutions to generate a more accurate picture.

Finally, tariff volatility makes it harder to align customer demand and inventory against external factors. Consider how a genAI-powered workflow could transform a ubiquitous customer touchpoint: the cart-abandon notification. Virtually every consumer gets an email reminding them of products left in a virtual cart. Sometimes, reminders include a coupon for future savings. However, few retailers are equipped to provide consumers a call to action such as a data-driven, personal coupon if purchased immediately. Maximizing revenue and profit on individual cart reminders would require a deep understanding of customer buying habits, how quickly a product is selling and real-time inventory.

Despite the industry’s appetite for genAI, most retailers have yet to knock down the barriers and integrate silos that stand in the way of embedding genAI into end-to-end workflows. Data is collected and managed across a patchwork of point-of-sale platforms, e-commerce systems, CRMs, supply chain tools, social media, and other external feeds. Each has its own formats, standards, and ownership. This fragmentation gets in the way of building a unified view of customers, inventory, or performance, and it undermines efforts to personalize experiences or respond quickly to market changes.

CDOs also contend with legacy systems, inconsistent data quality, and a lack of governance. Many retailers still rely on outdated technology for critical operations, and new data types, such as social media or real-time tariff updates, add to the complexity. Budget constraints, talent shortages, and unclear mandates further limit the CDO’s ability to drive change.

These challenges are not unique to retail, but they are especially acute in an industry where agility and customer insight are critical to success.

The CDO’s mandate: modernize, govern, and scale

Retail chief data officers are tasked with unlocking the value of genAI, but they face a tangle of technical and organizational hurdles. Tariffs and supply chain volatility are only the most recent example of the need for agility and precision at scale. Today’s CDOs need a new playbook to prepare their organizations to drive data-driven transformation.

Modernize architecture

Most retailers struggle to act quickly or strategically because their data is trapped in silos. Point of sale systems, CRMs, inventory systems, and external feeds like tariff updates all operate in isolation. Tariff volatility is a core area where decisions made too late can lead to cost increases or supply shortages if retailers lack the full picture of customer demand, inventory status, and how changes might affect sourcing or pricing.

A unified data environment is the only way to break down these barriers. Modernizing architecture means investing in cloud-agnostic platforms that connect all critical data sources, establishing clear standards and ownership to ensure consistency and reliability. AI platforms such as EXLerate.AI provide prebuilt connectors for retail systems, accelerating integration and enabling a single, actionable view of the business, which positions retailers to respond more quickly to tariff volatility and other market shocks.

Utilize AI-driven scenario planning

Once data is unified, the next challenge is using it to anticipate and adapt to change. Tariffs can reshape demand patterns and supply chain economics, but many retailers are unable to simulate their impacts. Traditional forecasting methods are slow or inaccurate, leading to overstocking, stockouts, or missed sales.

With AI, retailers can better analyze historical sales, market trends, and real-time tariff data, creating “what-if” scenarios and adjust strategies before new tariffs take effect. For example, a sudden tariff increase on imported goods might prompt a retailer to shift sourcing, adjust pricing, or run targeted promotions to clear inventory. Machine learning combined with genAI, such as the technology powering EXL’s AI solution for estimating elasticity and demand can help retailers stay ahead of impacts to inventory and pricing.

Optimize procurement and supplier management

Tariffs threaten to disrupt traditionally slow and reactive procurement processes and supplier relationships. When tariffs on imported materials increase procurement costs, they can limit supplier options, creating bottlenecks and margin pressure. AI solutions embedded into the workflow can provide real-time visibility that allows retailers to renegotiate contracts or shift sourcing strategies. For example, EXL’s AI solution providing end-to-end cost insights continuously update tariff data and supplier information, empowering retailers to explore switching suppliers from high-tariff to tariff-exempt regions.

Strengthen data governance and privacy

As retailers push into advanced AI use cases—such as personalized offers, virtual try-ons, and dynamic pricing—data privacy and compliance become critical. Customers expect their data to be handled responsibly, and regulators demand strict adherence to laws like GDPR and CCPA. Many retailers lack frameworks for data lineage, privacy, and compliance, which slows innovation and increases risk. Robust data governance builds trust in AI solutions and enables responsible adoption, including policies that require synthetic data for sensitive use cases and solutions that automatically mask personally identifiable information.

Measure impacts

All industries, including retail, are pushing to move beyond pilot programs. Scaling genAI throughout business workflow requires a disciplined approach to prioritization and ROI. Retailers should start with high-impact, low-complexity pilots, such as dynamic offers that respond to tariff-driven cost changes and then expand based on measurable outcomes.

Defining clear KPIs for each use case and tracking business impact is critical. By focusing on use cases that navigate a difficult trade environment, retailers can protect margins and improve agility. Closing data gaps will also build executive confidence, accelerate value, and prepare the organization to scale genAI across the enterprise.