Are You Still Delivering to Elm Street Three Days in a Row?

Written on: October 1, 2026 by Phillip J. Baratz

Every heating season, I hear a version of the same story. A dealer is convinced their operation is “pretty efficient.” The trucks are rolling, there aren’t many run-outs and dispatch “knows their routes.” Then, we pull a few months of data and discover something uncomfortable: They’re hitting the same neighborhood, sometimes the same street—on three, four, even five consecutive days.

If you’re delivering to Elm Street three days in a row, you may not have a crisis, but you absolutely have a problem. Thankfully, it’s a problem that can be solved. Not with another whiteboard, not with another “good enough” plan, but with analytics, and increasingly, with AI.

“We Don’t Have Run-outs, So We’re Fine”… Right?
Many companies still equate “no run-outs” with “we’re doing a good job.” That’s a low bar. You can avoid run-outs and still:

Deliver too early and too often.
Leave capacity on the table every single day.
Burn driver hours and fuel on poorly sequenced routes.

Your degree-day or consumption estimates can be off by a lot without showing up as run-outs. The only way you see them is with reporting that tells you, clearly and unemotionally, where your estimates, drops and schedules don’t line up.

That’s what analytics are for. It’s the mirror you hold up to the operation—not to make anyone look bad, but to show where there’s money and capacity being left on the floor.

Analytics is Not Just a Dashboard
In our world, “analytics” touches almost everything that matters:

Hedging: Understanding your position, your exposure, your basis history and how your supply choices really performed versus the alternatives. The good old days of predictable rack spreads and “set it and forget it” basis differentials are gone. One supplier, one index, one basis is the new “maybe it’s good enough”—and it usually isn’t.
Business Intelligence: Knowing which customers, products, branches and drivers actually drive your margin and your risk. Not guessing, but knowing.
Consulting & Advisory: Having people who know how to ask hard questions of the data, not just how to make a pretty chart.
Dispatch Planning: Not just “what stop comes next,” but “how should we be using our assets across days, trucks, drivers and terminals?”

The common thread is that all of these are becoming AI-assisted. That doesn’t mean a robot runs your company; it means the heavy math and pattern finding are handled for you, so your staff can focus on decisions.

Routing: The Big Levers are Above the Stop Level
When people think “AI in dispatch,” they often picture a computer picking the next stop on a route. That’s nice, but it’s not where the biggest gains come from.

The big levers are above the stop level:

Which deliveries should go today vs. tomorrow vs. next week?
How do we use all our trucks, not just the first two we think of?
Where should we reload, and how many times?
How do driver hours, terminal constraints and service priorities fit together?

Most dispatchers are still doing this one day at a time, visually: drawing loops on a map, building routes from scratch every morning and adjusting on the fly when the real world moves the goalposts. They’re smart, experienced and often overworked. They are doing the best they can with the tools they have.

Unfortunately, “the best they can with manual tools” is no longer good enough when your competitor is using AI-driven planning that looks at multiple days, multiple trucks, multiple drivers, reloads, hours of service, time windows and business rules all at once—and hands their dispatchers a high-quality plan to refine instead of a blank map to fill.

That’s where modern routing engines, including what we’re building at Angus, are focused.

The Human Problem: “My Way Works”
Even in companies that have strong analytics, business intelligence and AI tools, there’s another problem that quietly kills value: adoption. If there isn’t clear, sustained, top-down expectation that the tools will be used, they won’t be. It’s that simple.

Most people don’t wake up in the morning saying, “I can’t wait to learn a new system today.” They say, “My way works. I’ve been doing this for 20 years.” They’re not wrong: their way does work—just not as well as it could.

Leadership’s job is to:

Set the expectation that decisions will be data-driven, not habit-driven.
Protect time for people to learn and use new tools.
Ask, “What did the system recommend?” before asking, “What did you do?”

If analytics and AI sit on the shelf while everyone continues to “do what we’ve always done,” you’ve just bought an expensive screensaver.

Supply, “Good Enough” & the Cost of Comfort
We’re also seeing the same pattern on the supply side. For years, many dealers survived—sometimes thrived—with very simple supply strategies:

One primary supplier.
One or two terminals.
A sense that basis and differentials were “predictable enough.”

Those days are over. Volatility in flat prices and basis, localized rack games and infrastructure constraints mean that “we’ve always bought from X” is no longer a strategy; it’s an exposure.

Analytics and AI can now:

Compare your actual costs vs. what you could have paid with a different mix.
Show how often “loyalty” to one supplier helped you—and how often it hurt you.
Model different hedging and supply scenarios before you lock in.

Again, the point isn’t to throw out long-term relationships, it’s to be sure those relationships are delivering the value you think they are.

Surround Yourself with Best-in-Class Teammates
The overarching theme here is simple: The world is becoming more efficient. If you aren’t, you’re probably falling behind—whether you can see it yet or not. To keep up, and ideally get ahead, you need three things:

1. Best-in-class tools. Analytics, routing engines, hedging and business intelligence platforms that are built for your business, not for some generic industry.
2. Best-in-class teammates. The right mix of in-house talent and outside consultants who know how to ask hard questions of your data and your processes.
3. Top-down commitment. Leadership that says, “We are going to run this company with the best information available, and we expect everyone to use the tools that give it to us.”

That combination is where we’re investing at Angus: taking the complexity of hedging, business intelligence, consulting and dispatch planning, and putting analytics and AI behind them so you don’t have to guess. You ask better questions. You get better answers. You make better decisions.

Therefore, the next time you see one of your trucks on Elm Street three days in a row, don’t shrug and say, “That’s just the way it is.” Ask yourself:

“Do I know for sure that this is the best we can do?”
“Or is this just good enough in a world where my competitors are quietly getting better?”

If that thought makes you even a little uncomfortable, that’s your signal. It’s time to surround yourself with better tools, better analytics and better teammates—and make “good enough” a thing of the past. ICM