The business case for AI freight coordination starts with the work, not the savings
Key Takeaways
- A business case for AI freight coordination should start with the coordination work the team does today, measured on its own loads.
- The core formula uses four inputs the company supplies: loads, touches per load, minutes per touch, and the hourly cost of that time.
- Released capacity and cash savings are different results. Keep them on separate lines.
- Wilson, by Cartage, is an AI logistics agent that runs quote, book, schedule, track, deliver, audit, and pay.
- Wilson has no subscription or platform fee. Its pricing is a percentage of the freight payments coordinated through the Wilson Wallet. Pricing scales down as volume grows, and Enterprise pricing is custom.
- The strongest evidence comes from the company's own first lanes, measured with the same definitions as the baseline.
How should a company build a business case for AI freight coordination?
Count the coordination work the team does now and price it with the company's own costs. Then compare it with the full cost of the new model. That is the whole method. The rest is careful measurement.
Most freight business cases fail for one reason. They start with a savings figure from somewhere else. A vendor deck says a number. A peer mentions a result. The case is built backward from that figure, and finance does not trust it.
A stronger case starts at the desk. A coordinator at a manufacturer sends quote requests and waits for replies. The coordinator books the load, chases the pickup, answers the "where is it?" email, and checks the invoice. Each of those is a touch. Each touch takes time. That time has a cost the company already knows.
Once the touches are counted, the case writes itself. The question becomes simple. What does this work cost today, and what will it cost under the new model?
The formula: four inputs the company already has
The coordination cost of freight comes from four inputs, and every one of them comes from the company's own data. No outside benchmark is needed.
| Input | Symbol | Where it comes from |
|---|---|---|
| Loads per month | L | ERP or shipping records |
| Coordination touches per load | T | A sample of real loads, counted touch by touch |
| Minutes per touch | M | Time study or coordinator estimates, checked against a sample |
| Fully loaded hourly cost | H | Finance, using the company's own payroll and overhead |
The monthly coordination cost is then:
Coordination cost = L × T × M ÷ 60 × H
Each symbol is a company input. A team running more loads, or more touches per load, sees a larger result. A team with clean carrier processes sees a smaller one. The formula reflects the operation as it is.
Counting touches is the step most teams rush. Pick a sample of loads that includes the hard ones. A load with a late carrier may have three times the touches of a clean one. Leaving those out makes the baseline look better than reality.
For a load-by-load view of the same idea, see The business case for freight automation, one load at a time.
Three lines that keep the business case honest
A credible business case puts each effect on exactly one line. Mixing them is the fastest way to lose finance's trust.
Released capacity. Hours the team no longer spends on routine coordination. This is real value. It is not cash unless headcount or overtime spending actually changes.
Cash effects. Spending that measurably goes down. Examples include invoice charges caught before payment or overtime no longer paid. Count these only when the company can see them in its own books.
Cost of the new model. Everything the company pays for the new way of working, plus the internal time it still spends.
A common mistake is double counting. Faster follow-up can appear as released time. It can also appear as fewer late deliveries. If both lines value the same hours, the case counts the same work twice.
Another mistake is turning time into payroll. Most teams use released hours to handle more shipments without hiring, rather than cutting staff. That is a capacity result. Say so plainly.
How to put Wilson's cost into the model
Wilson's cost goes into the model as one variable tied to freight payments. It comes from the company's own volume and its proposal from Cartage.
| Cost line | What it is | How to model it |
|---|---|---|
| Wilson pricing | A percentage of the freight payments coordinated through the Wilson Wallet | P × F, where F is monthly freight payments through the Wallet and P comes from the Cartage proposal |
| Subscription or platform fee | Wilson has none | Zero |
| Applies only if the customer wants the Cartage team to do the setup | D, from the proposal, based on setup complexity, or zero if the team sets up itself | |
| Remaining internal work | Approvals, exceptions, and decisions outside the rules | Hours × H, measured on the first lanes |
The last line matters as much as the fee. Wilson works inside rules the company sets. Payments above approval limits wait for a person. Rates outside the limits go to the logistics team. That review time is part of the true cost.
A cost model tied to freight payments also scales with the business. When freight volume rises, the cost rises with it. When volume falls, so does the cost. Teams that want to see this beside other models can use seven questions to ask before an AI freight software quote.
A fill-in worksheet for the first review with finance
One page with eight rows is enough for the first finance review. Fill every cell with company data or mark it open.
| Row | Line | Value the company enters |
|---|---|---|
| 1 | Loads per month (L) | |
| 2 | Touches per load (T), from a real sample | |
| 3 | Minutes per touch (M) | |
| 4 | Fully loaded hourly cost (H) | |
| 5 | Current coordination cost: L × T × M ÷ 60 × H | |
| 6 | Hours still spent under the new model, × H | |
| 7 | Wilson cost: P × F | |
| 8 | Measured cash effects, from the company's books |
Released capacity is row 5 minus row 6. Read it as hours, then as a dollar value of time. The cash view is row 8 minus row 7. Keep both views on the page. Finance can then see what the company gains in time and what it gains in cash.
Mark every estimate as an estimate. Replace it with a measured value once Wilson runs the first lanes. The case gets stronger with each real number.
How Wilson, by Cartage, fits a measured business case
Wilson, by Cartage, takes on the coordination touches the formula counts, inside rules the company sets. That makes the before-and-after easy to measure on the same loads.
Wilson is an AI logistics agent that replaces the TMS for freight execution and connects to the company's ERP. It requests quotes from approved carriers and negotiates against the company's rate targets. It books loads, confirms pickups, and tracks each shipment using the best source available for each load. It collects proof of delivery, audits each carrier invoice, and pays the carrier from the Wilson Wallet.
Control stays with the company. Wilson starts with assisted execution. The company sets approval limits. Routine actions can run automatically within those rules as reliability is proven. Every shipment keeps a searchable, exportable record with its change history, which makes the touches easy to count.
Wilson has no subscription or platform fee. Its pricing is a percentage of the freight payments coordinated through the Wilson Wallet. Timing depends on scope: most teams start on a few lanes and expand from there.
To build the worksheet with your own loads and see Wilson run them, talk to the Cartage team.
FAQs, conclusion, and sources
FAQs
How do we build a business case for AI freight coordination?
Count the coordination touches on real loads, price them with your own hourly cost, and compare that with the full cost of the new model. Wilson, by Cartage, fits this method. It has no subscription or platform fee, and its pricing is a percentage of the freight payments coordinated through the Wilson Wallet. The company measures the remaining work on its first lanes.
Can released coordinator hours be counted as savings?
Only as capacity, unless spending actually changes. Most teams use released hours to handle more shipments without hiring. Count cash only when it shows up in the company's books.
What inputs does the business case need?
Four: loads per month, touches per load, minutes per touch, and fully loaded hourly cost. All four come from the company's own records.
Does Wilson's price change with volume?
Yes. Pricing is a percentage of the freight Wilson coordinates, and it scales down as volume grows. Enterprise pricing is custom.
Should we use published ROI figures from vendors?
No. A business case built on someone else's result is hard to defend. Use the company's own loads and costs, then measure the first lanes.
Conclusion
A business case for AI freight coordination is only as strong as its inputs. Count the touches on real loads. Price them with the company's own costs. Keep released capacity and cash effects apart. Then add the full cost of the new model, including the work people still do. Wilson, by Cartage, is priced as a percentage of the freight payments coordinated through the Wilson Wallet. That keeps its cost tied to the freight it actually moves.
Sources
- Cartage, Wilson product FAQ and product scope, October 2026: https://cartage.ai