Share:

From Data Chaos To Decision Intelligence: How AI Continues To Reshape Logistics

Supply chain organizations have never lacked data. What they have often lacked is the ability to connect it, normalize it, trust it, and turn it into better decisions.

That was one of the central themes in a recent conversation between Scott Luton and Matt McKinney, Co-Founder and CEO of Loop. Drawing on his experience helping build Uber Freight and now leading a logistics technology company, McKinney offered a clear-eyed view of where AI is already creating value, as well as why the data foundation still matters more than the model itself.

The Real Problem Is Not Data Volume

McKinney traces Loop’s origins back to a recurring problem he and his co-founder (Shaosu Liu) saw at Uber Freight: rapid growth on the revenue side paired with a back office still dependent on people manually reconciling invoices, contracts, proofs of delivery, and surcharges. The bigger realization was that this was not unique to Uber.

“The thing people miss is that the value isn’t in the quantity of data, it’s in the utility of it,” McKinney explained. “Most companies have plenty of data. What they don’t have is data that connects.”

That disconnect is especially painful in logistics, where data is scattered across PDFs, emails, spreadsheets, phone calls, carrier systems, customer systems, and legacy platforms.

Even seemingly simple inconsistencies, such as different spellings of an address or different names for the same carrier fee, can prevent systems from recognizing that two records refer to the same thing.

Why Verticalized AI Matters

McKinney believes AI is the first technology capable of solving many of these normalization problems at scale. But he draws an important distinction between general-purpose AI and logistics-specific intelligence.

“A general-purpose model can read a document,” he said. “It doesn’t know what a bill of lading is, why a pickup date has to precede a delivery date, or that those two fee names are the same charge.”

That is why McKinney sees verticalized AI as critical. The technology must understand the terminology, business rules, relationships, and workflows of supply chain – – not simply process words on a page. Once that domain context is applied, raw information can become a unified data foundation capable of supporting automation, analytics, and increasingly autonomous workflows.

Freight Audit Was the Doorway

Loop began with freight audit and payment because it is one of the most data-intensive and error-prone areas in logistics. The opportunity was immediate: structure the data correctly, then automate the process to generate transportation savings and reduce manual work. But freight audit was never intended to be the final destination.

Once customers had trustworthy, structured information, McKinney says they began applying it to procurement, contract compliance, decision intelligence, and network planning.

The lesson is an important one for supply chain leaders: solving a foundational data problem can unlock opportunities far beyond the original use case.

Where AI Is Already Delivering Value

For McKinney, some of the most practical AI opportunities today are hiding in the back office.He estimates that a significant portion of supply chain operational information remains trapped in offline and unstructured formats. Once AI can extract and structure that information, organizations can automate processes that historically required extensive manual review.

The payoff can include carrier disputes resolved in minutes instead of weeks, automated audits, and finance teams closing the books using actual costs rather than estimates. That kind of practical value matters even more as shippers face tariffs, sourcing changes, geopolitical instability, margin pressure and more.

Trust Has to Come Before Autonomy

Even when the technology works, organizations may hesitate to allow AI agents to make consequential decisions. McKinney believes the answer is transparency and sequencing.

“Trust comes from sequencing, giving teams visibility into accurate, validated data before asking them to hand over autonomy,” he explained.

Unlike legacy black-box systems, well-designed AI should allow users to see which documents were read, which rules were applied, and why a decision was made. That visibility can turn AI adoption from a mandate into something employees actively want.

Modernization Starts with the Foundation

For companies modernizing their logistics technology stack, McKinney’s advice is straightforward: resist the urge to begin with flashy features.

“Get the data foundation right first – – validated, audited, normalized – – before you try to layer automation or AI on top of it.”

That sequencing may become one of the defining differences between organizations that scale AI successfully and those that remain stuck in pilot mode.

The Future Moves Continuously

Looking ahead, McKinney believes the biggest surprise may not be that AI makes better decisions, but rather how quickly those decisions begin happening. Network decisions that once occurred quarterly, supported by last month’s data, could increasingly happen continuously.

The result is a fundamentally different way of managing logistics: less rearview mirror, more real-time intelligence. For supply chain leaders, the opportunity is enormous. But the path starts somewhere decidedly less glamorous than AI hype.

It starts with getting the data right.

Where to Learn More

Learn more about Loop, including its recent $95M Series C, via its company website:  https://www.loop.com/ . We also invite you to tune in to a past Supply Chain Now episode, where Matt joins Scott and a few industry leaders to talk about digital transformation across global supply chain: click here.

More Blogs

leadership
Blogs
April 29, 2026

From Integration to Impact: Lessons in Modern Supply Chain Leadership

In a recent conversation, Supply Chain Now’s Scott Luton gained perspective from Sylvia Wilks, Chief Supply Chain Officer at Lamb Weston, who shared a powerful point of view on what it takes to lead in today’s increasingly complex, high-stakes supply chain environment. Her journey, from leading transformative initiatives at Starbucks to shaping global operations at Kimberly-Clark and REI, offers a consistent message: Success in supply chain isn’t just about systems or strategy. It’s about people, integration, and clarity of purpose. Wilks’ passion for supply chain was sparked during her time at Starbucks, where she led a bold initiative to insource instant coffee production. What began as a business case evolved into a transformative opportunity. “Seeing the entire chain, from strategy through operations, work seamlessly toward a common goal reinforced how much value organizations unlock when supply chain subfunctions operate collaboratively rather than in silos,” she explained. The idea of breaking down silos to create an integrated value chain has remained a central theme throughout Wilks’ leadership career.   The Power of People and Integration Across organizations of all sizes, Wilks sees a common thread: The challenges may be similar, but outcomes depend on how well teams work together. “My passion…
Gartner Supply Chain Symposium
Blogs
May 21, 2026

From AI Anxiety to Workflow Reinvention: Key Takeaways from Gartner Supply Chain Symposium 2026

At Gartner Supply Chain Symposium/Xpo 2026 in Orlando, some of the sharpest minds in supply chain gathered to tackle one central question: what does the next operating model for supply chain actually look like? In a special livestream conversation, Scott Luton sat down with Mike Griswold, VP Analyst at Gartner, alongside fellow Supply Chain Now hosts Karin Bursa and Jake Barr, to unpack the biggest themes emerging from this year’s symposium. The consensus? Supply chain leaders are moving beyond AI fascination and toward something much bigger: redesigning how decisions, workflows, and organizations operate.   AI Is No Longer the Story. Outcomes Are One of the strongest themes from the event was a more mature, pragmatic approach to AI adoption. According to Mike Griswold, many organizations are finally moving past the “shock and awe” phase that dominated conversations a year ago. “People need to figure out exactly what problem or problems AI is going to solve for them,” Griswold explains. That may sound simple, but it represents a significant shift. Instead of experimenting with AI for AI’s sake, companies are becoming more disciplined about identifying operational value and measurable business outcomes. Griswold also warns against a familiar trap: creating “highly efficient…