AI in Distribution: The People Problem

The Real Reason AI Initiatives Stall in Distribution

Walk into any mid-market distribution operation that has invested in AI tools and struggled to see results, and you’ll find a consistent pattern. The technology is capable, use cases are valid,  but somewhere between the go-live and 90 days later, adoption has fragmented. Some people are using the tools and others have reverted to spreadsheets and tribal knowledge.

The change management research on enterprise technology is clear: when tools are introduced without addressing the beliefs, habits, and fears of the people who must use them, adoption stays shallow and the implementation never reaches its full potential.

AI sharpens this problem. Unlike a new ERP module or a process redesign, AI introduces something that feels extremely different to frontline employees: a system that can reason, recommend, and act. That raises a question your people will ask even if they never say it out loud… “what does this mean for me?”

As the CEO, your job is to answer that question before your people fill in the blanks themselves.

“The organizations that see the highest AI ROI are the ones that redesigned how work happens around the tools they deployed.”

Three Groups That Will Make or Break Your Adoption

Resistance to AI doesn’t come from everywhere at once. It concentrates in predictable places. Understanding where to expect it helps you get ahead of it.

The experienced operator.

Your 15-year warehouse manager or top sales rep has watched technology projects arrive with promises, create disruption, and underdeliver for years. More than that, their professional identity is built on their earned expertise, and AI appears to threaten the value of that expertise. The reframe that works here is truth: AI handles the routine so that experienced operators can apply judgment to situations no algorithm can navigate. Their knowledge is what makes the AI smarter.

The middle manager.

If AI can surface the answers that have historically required their expertise, what’s their role? Left unanswered, that question fills with resistance. Answered early, it becomes an opportunity: their role shifts from information conduit to decision architect. They can become the person who defines which questions the AI should answer and interprets its output in operational context. The organizations that see this coming and invest in redefining these roles will pull away quickly from the ones that don’t.

The new hire.

Counterintuitively, newer employees sometimes resist AI for a different reason: they came to learn from experienced colleagues. AI-mediated workflows can feel like they’re cutting off that path to their learning. The fix is to make AI part of the learning experience. It is a tool that explains the reasoning behind decisions and accelerates ramp time rather than bypassing it.

Where Is Your Organization on the AI Readiness Curve?

Adoption doesn’t happen all at once. It follows a progression and knowing where you are determines the right next move.

• Stage 1 — Awareness: Leadership understands AI’s relevance in general terms, but no one has translated that into specific operational use cases. Your people have formed opinions from LinkedIn and trade press. The priority here is narrative: what problem are we solving, and why now?
• Stage 2 — Experimentation: Specific teams or individuals are using AI tools, but without structure. Usage is driven by individual initiative, not aligned strategy. The priority is identifying the highest-value use cases and building governance that can take experimentation to scale.
• Stage 3 — Integration: AI is embedded in defined workflows. Usage is no longer optional. The priority is measuring impact rigorously and using those results to build internal credibility and secure further investment.
• Stage 4 — Transformation: AI has changed not just how tasks are performed but how the organization is structured. It determines which roles exist, how decisions are made, and what the competitive profile of the business looks like.

 Most mid-market distributors are between Stages 1 and 2 today. The operations that will own Stage 4 in three years are the ones beginning the Stage 2-to-3 transition now.

What the First 90 Days Should Actually Look Like

The most common mistake in AI rollouts is leading with the technology. What helps is leading with the problem it solves instead.

• Weeks 1–2: Name the operational problem you’re solving first and the specific friction your teams feel every day.
• Weeks 3–4: Identify your early adopters. This is the 10 to 15 percent of your workforce who are genuinely curious and willing to experiment. These are your internal credibility builders.
• Month 2: Run a structured pilot on one contained, high-value workflow. Measure rigorously and capture specific examples of time saved, errors prevented, and decisions improved.
• Month 3: Use those results. Now you have your own data to broaden adoption with a credibility base that the skeptics can actually engage with.
• Ongoing: Create a visible channel for AI feedback and concerns. Resistance that stays underground will continue to grow.

Trust Is the Foundation, and It Has Three Layers

Getting your people to use AI tools requires trust, and organizational trust in things like AI is layered.

Trust in the output.

Employees need to understand why the AI is recommending what it’s recommending. Black-box outputs breed suspicion. The platforms that work best in distribution environments are the ones that can explain their reasoning.

Trust in the governance.

People need to see that AI cannot take consequential actions without appropriate approval, that its activity is logged, and that mistakes can be identified and corrected. This is the fundamental architecture of organizational trust.

Trust in your intent.

Vague reassurances create more anxiety than honest acknowledgment of change. Your people need to hear directly from you: what AI will change, what it won’t, and what your commitment to them is through the transition.

The Bottom Line for Distribution CEOs

The AI tools available to mid-market distributors today are genuinely capable. Epicor Kinetic and Prophet 21 are embedding AI into the workflows your teams use every day, from intelligent procurement automation to natural-language ERP queries that surface answers in seconds. The platforms are ready.

The question is whether your organization is ready to meet them.

In our last post, we covered governance and risk management as the structural foundation of AI adoption. In Part 4, we’ll look at how to build an AI strategy for your distribution operation that compounds over time.

Not sure where your operation sits on the AI readiness curve?

We work with manufacturers and distributors on Epicor Kinetic and Prophet 21 to help you identify what’s actually ready to activate and what to sequence first. Let’s talk through where you are.