Everyone in retail wants AI right now. That much is obvious.
What’s less obvious is whether most organizations actually have a plan for implementing it responsibly.
Because if you spend enough time talking to retail executives today, you start hearing the same thing over and over again: “We’re encouraging our teams to experiment with AI.” Which, on the surface, sounds smart. Progressive, even. Empower people. Let innovation happen from the bottom up. See what efficiencies emerge naturally.
And honestly? I get it.
If I were still leading a merchandising organization today, I’d probably feel the same pull. AI is exciting. The pace of change is staggering. Nobody wants to be the executive who moved too slowly while competitors figured it out first.
But after my latest Omni Talk Retail Technology Spotlight conversation with Judah Berger, AI Product Manager at Unframe.ai, I walked away thinking about AI implementation very differently.
Because Judah articulated something I don’t think enough people are talking about yet:
Retail may be on the verge of accidentally recreating one of its oldest operational problems at an even larger scale.
AI Could Become “Excel on Steroids”
One of the most interesting moments in the conversation came when we started talking about the history of retail technology itself.
For decades, retailers have survived operational complexity through what I’ll lovingly call “Excel duct tape.” Smart people inside organizations built clever spreadsheets, custom formulas, workarounds, macros, and reporting systems to solve problems quickly. And to be fair, a lot of those systems worked. Or at least worked well enough.
But they also created fragmentation.
Different teams solved the same problems differently. Logic became inconsistent. Institutional knowledge got trapped inside files only one person understood. Scaling became difficult. Governance became nearly impossible.
Now replace spreadsheets with AI prompts, workflows, agents, and custom logic.
That’s the real risk Judah sees emerging right now.
As companies rush to adopt ChatGPT, Claude, and countless other AI tools, employees are naturally building their own systems, prompts, and processes to solve problems independently. Which is great for experimentation. But potentially dangerous at scale.
Because unlike Excel, AI introduces variability directly into the reasoning layer itself.
Two employees can ask the same question in slightly different ways and get completely different answers. Multiply that across dozens of workflows, departments, and decision points, and suddenly you don’t just have operational inconsistency. You have operational chaos.
That observation hit me hard because it feels undeniably true.
The Retail AI Problem Nobody Wants to Admit
One of the smartest points Judah made during the conversation is that AI implementation is not really a technology problem first. It’s an operational discipline problem.
That distinction matters.
Too many organizations are approaching AI like a giant moonshot transformation initiative. They’re asking broad, almost impossible questions like:
“How do we become an AI company?”
But Judah’s argument is much more practical.
Instead of chasing futuristic transformation narratives, retailers should start by identifying repeatable operational pain points. What tasks consume hours every week? What workflows require employees to manually move data between systems? What reporting processes rely on tribal knowledge?
If you can explain a process to a reasonably capable new hire in a few days or weeks, there’s a strong chance AI can help automate or accelerate portions of it today.
That framing suddenly makes AI feel much less intimidating.
It stops being science fiction and starts becoming operational plumbing.
And frankly, I think a lot of retail leaders need to hear that right now.
Why Bottom Up AI Adoption Still Matters
Now to be clear, Judah is not arguing against experimentation.
In fact, he believes bottom up experimentation is critical.
Some of the best AI use cases inside organizations are emerging organically from employees themselves. People closest to the work often understand inefficiencies better than leadership ever will.
The danger comes when experimentation scales without governance.
Because eventually, organizations need consistency, auditability, security, and traceability. They need to understand why systems produce certain outputs. They need confidence that logic is repeatable and explainable across the enterprise.
That’s where many companies may underestimate the complexity ahead.
It’s relatively easy to build a compelling AI demo.
It’s much harder to deploy scalable AI systems that continue working reliably across thousands of operational decisions every single day.
What Unframe.ai Actually Does
This is where Judah’s role at Unframe.ai becomes interesting.
As an AI Product Manager, Judah works directly with enterprises to help operationalize AI inside their organizations. That means identifying business inefficiencies, understanding workflow bottlenecks, scoping the right use cases, and building AI systems employees can actually trust and use.
The company’s approach combines reusable AI infrastructure, governed enterprise data layers, auditability systems, and customizable AI agents into a managed deployment framework.
What stood out most to me, though, was the emphasis on traceability.
Every AI recommendation, workflow, or output needs to be understandable. If an AI system flags an inventory issue, recommends a reorder, or surfaces a business insight, employees need visibility into how that conclusion was reached.
Without that transparency, trust breaks down quickly.
And if employees don’t trust the system, adoption disappears.
The Footwear Retailer Case Study
The conversation became especially compelling when Judah walked through a real world example involving a major footwear retailer.
The retailer operated more than 140 stores with thousands of SKUs across shoes, handbags, and accessories. Like many retailers, they struggled with inventory visibility, manual analysis, and inefficient product transfers between stores.
Employees spent enormous amounts of time manually reviewing reports, analyzing sell through rates, and identifying inventory issues. By the time opportunities or problems surfaced, it was often too late to act effectively.
Unframe.ai built an AI powered inventory intelligence system that created daily operational briefs for managers.
The system flagged:
- products needing reorder attention
- inventory allocation opportunities
- local store performance anomalies
- trending products
- inventory risks
- forecasting insights
Managers could interact with the data conversationally, drill into SKU level performance, analyze store trends, and receive AI generated recommendations with traceable reasoning behind them.
According to Judah, the retailer ultimately saw a reported 40x ROI from the deployment.
Whether that exact number generalizes across every retailer is probably less important than the larger takeaway:
AI becomes significantly more valuable when it’s embedded directly into operational decision making.
The Bigger Strategic Question
The more I think about this conversation, the more I think the real threat facing retailers isn’t AI itself.
It’s fragmented AI adoption.
Because somewhere out there right now, AI native retailers are being built from scratch with centralized AI governance, operational intelligence layers, and scalable workflows embedded directly into the organization from day one.
Traditional retailers don’t have that luxury.
They’re layering AI onto decades of operational complexity, disconnected systems, inconsistent processes, and institutional workarounds.
Which means the companies that succeed likely won’t be the ones experimenting with AI the fastest.
They’ll be the ones implementing it the most thoughtfully.
To hear the full conversation with Judah Berger, listen to or watch this episode of the Omni Talk Retail Technology Spotlight wherever you get your podcasts.
Apple Podcasts | Spotify | SoundCloud | Amazon Music | YouTube
Be careful out there,
– Chris and the Omni Talk team
P.S. See our past 8 years of wonderful Spotlight Series podcast guests, featuring roughly 200 movers and shakers in retail, by clicking here
Music by hooksounds.com
*Sponsored Content*
Omni Talk® is the retail blog for retailers, written by retailers. Chris Walton founded Omni Talk® in 2017 and have quickly turned it into one of the fastest growing blogs in retail.