How to find price overrides in your invoice data

What counts as an override

Most ERPs store both the price the system calculated and the price that went on the invoice. When they differ and no contract or promotion explains it, a person made a pricing decision on the fly. Some of those decisions are good: a competitive bid, a damaged pallet, a customer worth keeping. The problem is that most businesses cannot tell the good ones from the habitual ones, because nobody has ever looked at them together.

Pull the right export

Ask whoever runs your ERP for twelve to twenty-four months of invoice lines with these fields. Most systems have them, even if no standard report shows them.

• Invoice date, customer, and ship-to location.
• Product and quantity.
• System price: the matrix, contract, or list price the ERP calculated.
• Price charged, and cost at the time of sale.
• The user who entered or changed the price, and any reason code.

If your system overwrites the calculated price instead of storing it, you can rebuild it from the price matrix in effect on each date. It takes longer, but it works.

Two numbers tell you whether you have a problem

Multiply the two and you have a rough annual figure. NAW published one distributor’s case in which cutting overrides from half of revenue to a tenth lifted gross margin by almost three points and EBITDA from 4 to about 7 percent of sales¹³. McKinsey described a building materials distributor whose reps overrode system prices on more than half of transactions. After it fixed its product data, overrides fell 80 percent and margin rose more than 200 basis points¹⁴.

Sort the overrides before you judge them

A single override rate invites a fight with the sales team. Sorting it turns the fight into a conversation. Group overridden revenue by who made the change, which customers received it, and whether a reason code was entered.

• Overrides concentrated in a few users usually point to habit, or to a rep protecting an account.
• Overrides spread across everyone on the same products usually point to a bad system price. Fix the matrix, not the people.
• Overrides on small repeat orders are the hardest to defend, because there is no competitive bid to explain them.

Write the hunch down first

Before anyone opens the export, ask your sales leader to write down what share of revenue they think runs on overrides and what it costs. Most leaders guess low. The gap between the guess and the record is often the most persuasive number in the room, and writing it down first keeps anyone from deciding afterward that the result was obvious⁹.

What to change, in order

Start with the system price. If reps override the same items again and again, the matrix is probably wrong for those items, and correcting it removes overrides nobody needs to argue about. Next, set a margin floor so that overrides below it need a reason code or a manager’s approval. Finally, show each rep a recommended price drawn from your own win history beside the quote. Test each change on a group of customers against a control, with a stop condition agreed in advance, so the result settles the argument.