A dealer once told me of a special talent his grandfather employed when he ran the lumberyard. “He could just look at you,” the dealer said, “and he could tell whether you were a credit risk.” That always sounded crazy, but recently I heard about a woman who noticed the way her husband smelled had changed as he developed Parkinson’s. Perhaps while that lumberyard grandpa stared, he could smell fear on the customer?

It would be great if every construction supply operation had someone like that on the payroll, but alas. Most operations struggle when it comes to knowing whether a particular customer ultimately makes or costs the dealer money. The best these dealers can do is apply averages to the specific: average cost of making a delivery, average cost of processing an order, average value of goods returned, and so forth.

Negotiating with customers based solely on averages isn’t good enough. As former Dealer of the Year Leonard Safrit once put it: “How in the world can you sell Customer A, who requires you to go to the jobsite 30 times and builds 200 miles from the store, and Contractor B, who lives across the street and requires you to go to the site three times, for the same amount of money? We’re doing it all the time, and that’s the stupidest thing in the world.”

Against this backdrop, I’m seeing more dealers investigate how they can produce better numbers on what it costs to serve each customer, and thus know what to charge that particular person. Safrit was among the pioneers, creating an enormously long formula that he says has helped his company not only bid more profitably but also know when the deal being pursued could be a money-loser once all the costs are tallied. Safrit says using the formula has helped him add several points of net profit.

Safrit could write his formula because he could pull the data from technology he was using, particularly his ERP and dispatching systems. They definitely can help, as could having a warehouse management system. But you can still begin this process without them.

In Utah, Sunpro operates without dispatch/delivery software or a warehouse management system, and its credit plan is nothing exotic; it even lacks a prompt pay discount. Nevertheless, Sunpro has created a grid showing 40 different price levels, with about a 25-point difference between the lowest and higher prices, to use when dealing with customers. This pricing strategy is adjusted every week based on four factors: the customer’s current or potential volume, the cost of credit for that customer, the cost of serving that customer (measured by factors like proximity to a yard and how many returns it makes), and the number of different products the customer buys.

A consultant I know has been working with a dealer recently to use basically the same numbers as Sunpro to analyze how fixed and variable costs vary among the dealer’s biggest customers. At this dealer, when you deducted the average total cost to serve from average gross profit margin, the net pre-tax profit was about 5%. But when you looked at the results for each of the biggest customers, the results ranged from a 14.8% profit to a 3.3% loss.

Now more than ever, it’s vital to recognize which customers generate a profit and which cost you money. Technology will help you get there faster, but it’s not mandatory. What must happen is that you take the first step.