Ronald Clarke's fuel card ads ran the same line to small-business owners for years: use the card, save at the pump, pay no setup fees, no transaction fees, no membership fees. A federal court found every one of those fees on the bill. They just had new names. The advice to "shop around for the best discount" fails here, not because operators are careless, but because the mechanics of pricing work against them the moment they sign.

Most operators who manage vendor costs have heard that advice. Compare rates. Pick the lowest number. Call it due diligence. It sounds right. It is not.

The FTC built its case against FleetCor, now called Corpay, over several years. The agency found that FleetCor promised savings and zero fees to tens of thousands of small-business customers. The invoices told a different story.

Six fee categories showed up under fresh labels: Account Administration Fees, Program Fees, High Credit Risk Account Fees, Convenience Network and Out of Network Fees, Minimum Program Administration Fees, and Late Fees paired with Interest and Finance Charges. Each one matched a fee the ads said did not exist.

When a customer caught a charge and got it removed, FleetCor swapped in a different one. That detail matters. It means the fee structure was not an error. It was a system. Remove one line item and a new one fills the gap.

Fees did not appear on the main invoice. They were buried in separate account reports. Some did not show up until several billing cycles had passed. FleetCor charged late fees to customers who had paid on time, and to customers FleetCor itself had blocked from paying on time. The FTC documented hundreds of millions of dollars in total harm.

FleetCor's own records showed customers did not get the savings the ads promised. Internal emails showed the company knew. A federal appeals court upheld the judgment against FleetCor on all counts, but vacated one count against its CEO. The settlement: $100 million. FleetCor's annual revenue: $4.5 billion. The fine was a cost of doing business.

There is a name for this in the literature. Researchers Santana, Dallas, and Morwitz ran six studies on the pattern, published in Marketing Science. They found that when fees show up after a buyer has committed to a base price, the buyer tends to stay with the more costly option. Even after seeing the real total. Even when given the chance to switch. The cost of starting over feels higher than the fee. So the buyer absorbs the charge and moves on.

Earlier work by Morwitz and colleagues found that 23% of buyers simply did not register the added charge when asked to recall what they paid. Not because they were not paying attention. Because the base price had already set the anchor. The discount locks the choice. The fees come after the commitment. The brain does not recalculate.

That is not a discipline problem. It is a measurement problem. And no amount of "shopping around" fixes a measurement the buyer was never set up to take.

The platitude treats the quoted rate as the cost. It is not the cost. The cost is the net number after every line item the vendor can add. The flaw is not that operators fail to compare. The flaw is that comparing on quoted rates rewards the vendor who hides the most on the back end.

The better principle: calculate net cost per unit after all fees, on any vendor that bundles a savings pitch with a locked-in payment product. Once that is clear, three moves follow.

Pull three straight billing cycles from any vendor where you were promised a discount. Line up every charge against the original contract, line by line. Note every item that does not match the deal you signed. Look at the main invoice and any side reports tied to the account.

This is not quick work, and it is not the kind of thing that feels productive while you are doing it. Most of the charges that matter are the ones you were never told to look for.

Add up every dollar you paid the vendor across those three cycles. Divide by the total units: gallons, transactions, shipments, whatever the vendor counts. Compare that number to what you would pay at the open-market price with no vendor deal at all. If your net cost per unit sits close to or above that baseline, the discount is not real. It is a rate card with a better story attached to it.

Before you sign any new vendor deal, ask the vendor to list every fee category in writing. Not a summary. The full list, with every line item named. Then check those categories against your first three invoices once the deal starts. If a charge appears that was not on the list, you have your answer before the switching cost locks you in. If the vendor will not produce the list, that is also an answer.

Running this for one quarter does something the advice never did.

The gap between the quoted discount and the real net cost becomes a number, not a hunch. Vendors who survive the audit separate from those who do not. The place where hidden margin recovery lives shows up on paper, in ink, not as a feeling in your gut. And fee swaps, where one charge goes away and a new one fills the space, become visible in the pattern across months.

At the end of the quarter, ask three things:

→ What moved the net number?

→ What looked like savings but left no trace on the bottom line?

→ What fee showed up more than once under a different name?

That is the gap between advice that sounds right and a system that proves itself.

The discount is still printed on your invoice. It always was. The question was never whether the rate looked good. The question was what else sat on the bill that no one measured. Now you know what to count. Not the rate. The net number.