Retail leaders are under pressure to deploy AI. But the data problem that has always slowed execution is the same one that will determine whether AI delivers or disappoints.
Retail leadership teams are moving fast on AI. Boards are asking about it. Technology vendors are pushing it. And the competitive narrative is clear: retailers who figure out AI faster will have a meaningful advantage.
But in conversations with leaders across Real Estate, Construction, Facilities, Store Planning, and Brand, the same honest tension keeps surfacing. The pressure to act on AI is real, and so is the recognition that most organizations aren’t actually ready for it. Not because the tools aren’t there. Because the data isn’t.
“Every senior leader we talk to wants to use AI to make smarter capital decisions, reduce execution risk, and move faster across their portfolio. The ones who are being honest with themselves know the foundation has to come first,” says Tia Kachman, COO of Immersion Data Solutions. “You can’t build predictive intelligence on top of data you don’t trust.”
The tension that defines this moment
Retail executives are navigating a genuine strategic paradox. The mandate to invest in and deploy AI is coming from the top. At the same time, the people responsible for the operation of hundreds or thousands of physical locations know that the physical portfolio has never been systematically captured, validated, or maintained as a structured data asset.
This is the AI investment vs. data integrity tension that shows up across every function in the enterprise.
- Construction leaders want AI to help sequence capital programs and reduce change-order exposure.
- Facilities leaders want AI to shift their teams from reactive firefighting to condition-based maintenance and risk prioritization.
- Store Planning leaders want AI to explain performance variance across formats and flag which prototypes are underdelivering.
- Real Estate leaders want AI to sharpen site underwriting and surface portfolio risk before it becomes a problem.
All of that is achievable, but it won’t happen without a reliable, validated, structured picture of the physical portfolio as a starting point.
Why physical data is the missing layer
Modern retail has invested heavily in systematizing two of its three core operating domains. Customer data flows through CRM platforms. Merchandise data flows through ERP and supply chain systems. Both are structured, validated, and continuously updated, which is exactly what makes AI applications in those domains credible.
Physical location data has never received the same treatment. It gets captured episodically, project by project, team by team. Construction does a site visit for a remodel. Facilities conducts a condition assessment. Real Estate commissions a pre-lease survey. The data from each of those efforts lives in different folders, different formats, different systems, and is often out of date by the time the next program needs it.
The result is a data environment that looks fragmented from the outside because it is fragmented on the inside. Standardized taxonomies don’t exist across the portfolio. Year-over-year comparison is limited or nonexistent. There is no validated baseline that Finance, Operations, and the C-suite can trust when AI surfaces a recommendation or flags a risk.
“The question we hear most often isn’t ‘Can AI help us?’ Everyone believes it can. The question is ‘Can we trust what AI is going to tell us?’ And that question always comes back to whether the physical data feeding it is structured, validated, and current,” says Nick Bonko, Immersion Data Solutions account executive.
What “AI-ready” actually requires
Retailers who are building durable AI capability in their physical portfolio programs are doing something specific before they deploy models: they are establishing a validated truth layer for their stores.
That means a few things. First, site data is comparable across locations, scored and evaluated against the same criteria, not assembled differently for every project. Data is time-aware, carrying trend lines and history rather than just a point-in-time snapshot. Data is also auditable, structured in a way that Finance, Risk, and Operations teams can stand behind. And data is accessible at portfolio scale: answerable to the kinds of questions AI needs to ask across hundreds or thousands of sites simultaneously.
This is not a technology limitation. The AI tools exist. The challenge is upstream, in building the physical data infrastructure those tools require to produce outputs anyone trusts.
“Retail has already done this work for customers and inventory. The companies getting ahead now are the ones applying that same discipline to their physical data portfolio,” says Kachman. “Capture the right data once, structure it correctly, keep it current, and AI insights become something you can actually act on.”
The cost of waiting
Retailers that delay the data foundation work while continuing to invest in AI tools are creating a specific and compounding problem. As they accumulate tools, their data often becomes more fragmented and difficult to trust. They might find that confidence in AI-generated outputs erodes precisely when leadership needs to rely on them most, such as during capital planning cycles, remodel sequencing, portfolio risk reviews, and growth program decisions.
There is also a competitive dimension. Retailers that get the physical data foundation right now will have a structural advantage that builds over time. As you accumulate validated site data, you’ll see comparability improve, forecasts stabilize, and decision velocity increase. That is not a one-quarter gain. It is a durable operating capability that becomes harder to replicate the longer you wait to build it.
The leaders who will look back at this period as a turning point are the ones who recognized that AI readiness and data readiness are not separate work streams. They are the same work.
Building the foundation: where Phygii fits
Phygii, from Immersion Data Solutions, is a retail property optimization (RPO) platform designed specifically to solve the physical data problem that sits at the center of this challenge.
The platform builds and continuously maintains Phygital twins of each site: validated, structured, time-aware representations of physical reality that serve as the data foundation across every function and every program. Construction, Facilities, Real Estate, Store Planning, and Brand all work from the same source of validated truth rather than maintaining separate, incompatible captures.
Phygii does not replace ERP, CMMS, or BI systems. It functions as the physical ingestion layer those systems have never had, structuring site-level reality in a standardized, auditable format and making it queryable at portfolio scale. When AI and analytics tools connect to that foundation, they connect to data they can actually use.
The model is straightforward: capture once, reuse everywhere. Rather than commissioning new site surveys for every program, validated site data accumulates and improves with each initiative. The result is reduced planning costs, elimination of redundant visits, and giving every team a shared picture of conditions, constraints, and readiness across the portfolio.
The path forward
The competitive pressure to deploy AI is not going away. But deploying AI without addressing the physical data foundation underneath it is an investment in outputs no one will fully trust.
The retailers who move from AI investment to AI advantage will be the ones who treat the physical portfolio as a managed, validated data asset, not as a collection of project files accumulated over years. That shift does not require replacing existing systems. It requires adding the one layer that makes every other system, decision, and AI initiative more reliable.
Truth first. Intelligence follows.
Ready to assess your organization's AI readiness comprehensively? Download our complete white paper, "Getting Your Property Data AI-Ready: A Strategic Guide for Multi-Property Retail Leaders," which includes the full AI Readiness Assessment Tool, detailed transformation roadmap, and practical frameworks for advancing through each maturity stage.
