Retail Supply Chains Are Broken - Here’s How Unified Platforms Fix Them

old way of supply chain versus agentic AI

Subscribe to our newsletter


If you've spent any time in Enterprise retail, you’ve seen it: departments that don’t talk to each other, systems that weren’t designed to share data, and a patchwork of bolt-on software that leaves you with more questions than answers.

I had a conversation recently with Mandar Rahatekar, Senior Director of Product at Manhattan Associates, that clarified what’s broken and where the opportunity lies. Mandar leads the team designing their supply chain planning solutions, and what he shared confirmed what I’ve been hearing from retailers globally:

We're past the point of tolerating disconnected systems. It’s time for a unified retail supply chain platform.

Here are five key takeaways from that conversation:

The Problem Was Never Just Tech—It Was the Siloed Thinking Behind It

Early supply chain software was designed by need, not by vision. Procurement teams got their tools. Warehouses had their systems. Labor, shipping, and merchandising were built in isolation. “We didn’t deliberately build silos,” Mandar said, “but there was a natural limitation. Each department solved its own problem with the tech available at the time.”

That patchwork mindset still lingers. What started as separate programs evolved into full-blown infrastructure built on incompatible foundations.

It’s no wonder enterprise retailers now find themselves in what I call a “Tower of Babel.” Everyone’s using tools, but no one’s speaking the same language.

Even as APIs allowed limited data sharing, Mandar pointed out, “There wasn’t a shared language underneath. Just plug-ins trying to force cooperation.”

We Now Have the Computing Power to Do What We Couldn’t Before

The big shift? Processing capacity. It’s what makes real-time insights and AI orchestration possible across a supply chain.

Mandar explained it like this: "In the old days, if all I could do was place a PO by hand, I wasn’t thinking about optimizing inventory. I was just trying to keep up.” Today, that same procurement role can click a few buttons and see optimized replenishment proposals, yet many companies still operate as if each role is on an island.

And the move to cloud isn’t just storage—it’s enablement. At Manhattan Associates, every client gets a secure, private cloud environment. “We use Google as our cloud provider, but every customer has their own secure box,” Mandar said. “If a customer has multiple applications, they can finally talk to each other in that box.”

The iPhone Analogy: Unified Systems Replace Single-Purpose Tools

Remember how revolutionary the first iPhone was? Before that, we all carried around separate devices: one for calls, one for music, one for internet. When Apple unified them, we didn’t just gain convenience—we unlocked new possibilities.

Mandar sees Manhattan Active in the same light. “We used to have a warehouse management system, a planning system, and an order management system. Now, we can unify them.”

He’s not just talking about efficiency. He’s talking about visibility—every team seeing the same data, making decisions from a single source of truth. And if you’re thinking that sounds like a headache to implement, Manhattan’s Agent Foundry is built to reduce that friction. It lets teams build AI agents that can automatically connect data, highlight exceptions, and surface insights—without needing a full IT team to wire it together.

From Firefighting to Forecasting: AI Enables the “Why” Behind Inventory Problems

The most valuable insight Mandar shared may have been this: Retailers know when they have too much or too little inventory, but rarely can they explain why.

A planner might overorder after a heated call from a salesperson. Or maybe a buyer expected a promotion to drive demand, but it didn’t. “There are so many human factors influencing planning decisions,” Mandar said, “and those are hard to model. But what if a system could tell you the story of a SKU?”

Imagine that. Not just knowing your numbers, but having AI tell you:

  • When the forecast was adjusted
  • Who overrode it
  • What social or marketing signals were firing at the time
  • How that change impacted the next six weeks of sales

Mandar calls that a "life history of a SKU." And if agents can deliver that story, you go from reacting to fires to preventing them.

Change Management Is the Hardest Part—But the New Generation Is Ready

So why aren’t more retailers jumping on this? Because change is uncomfortable.

“Even if we show the same inventory number to a planner and a warehouse manager,” Mandar noted, “it still takes a shift in mindset. People are used to waiting for reports, verifying against their own numbers. Now they don’t have to.”

He likened it to smartphones: “My son doesn’t even think of his device as a phone. That’s just one app. He uses Discord, Snapchat, FaceTime. His whole mindset is different from mine.”

It’s not about getting older generations to catch up; it’s about preparing leaders for radical transparency and autonomy. The system will flag what needs to be done. The human has to act on it. As Mandar put it: “You still need the human element. Tools complement us, they don’t replace us.”

If you're an enterprise retailer still stuck in disconnected systems—where store teams, inventory managers, and ecomm planners are running their own playbooks, it’s time to step back. You don’t need more dashboards. You need clarity.

Check out Manhattan Associates

Their Manhattan Active suite is a cloud-native, unified supply chain commerce platform that handles:

  • Warehouse Management
  • Order Management
  • Transportation
  • Labor Planning
  • Inventory Optimization

It's built to connect all your data, eliminate the Tower of Babel, and empower your people with real-time insights that finally make sense.