ERP for Distributors: The Price List and the Unit of Measure Decide the Fit.
Every ERP vendor with a distribution page lists the same modules: inventory, purchasing, order management, warehouse, reporting. A distributor reading five of those pages learns very little, because the modules are not where the systems differ. They differ in how an ordinary product behaves once it is in the system: how many units it has, how many prices it can carry, and what it really cost by the time it reached the shelf.
We suggest a different way to choose. Pick three behaviors that a distributor's margin depends on, and test each with your own data before you shortlist a vendor.
A distributor makes money on the spread between what it paid and what it charged, across thousands of products, many customers and several suppliers, each with different terms. A small error in any of the three numbers on a line, the cost, the price or the quantity, is multiplied by every line that shares the mistake.
That is why the item master matters more than the module list. The item master is the record of each product and everything that varies about it. When it cannot hold what is true about a product, the business keeps the truth in spreadsheets beside the system, and then the system is a record of the spreadsheets. This is the pattern we described in why the stock count, the job cost and the invoice disagree, and distribution is where it starts earliest.
A distributor buys a product by the pallet, stocks it by the case, picks it by the case or by the each, and sells it by whichever the customer asks for. The conversions between units are simple arithmetic, and they are where errors breed: a case of twelve received as twelve cases, a price per each applied to a quantity in cases, a count in the warehouse in a unit the order system does not know.
Ask the vendor to set up one of your real products with its true purchase, stock and sale units, then run it through a receipt, a pick, a sale and a return. Check that the quantity on hand, the cost per unit and the invoice line all agree in every unit. Check what happens when a customer orders a fraction of a case, and what the system does with the leftover. If the vendor needs a workaround for your most ordinary product, you will need the workaround for all of them.
A distributor's price for a customer is rarely a single number. There is a list price, a customer contract price, a quantity break, a promotion and sometimes a rebate paid later. Several of those can apply at once, and somebody decided which wins.
Pick five real customers, each with a different arrangement: one on a contract price, one on a discount from list, one with a volume break, one with a promotion running and one with no special terms. Ask the vendor to price a mixed order for each, and then to explain, line by line, which rule produced each price. If the explanation is "the system worked it out", you cannot audit your own prices. If the explanation names the rule, the customer's record and the date it took effect, you can.
The cost on the shelf is rarely the cost on the supplier's invoice. Freight, duty, handling and receiving differences all belong to the product, and a distributor that leaves them in an overhead account sees a margin that is higher than the real one.
Ask the vendor how a freight charge that arrives on a separate invoice reaches the cost of the products it carried, and how it is spread: by weight, by value, by quantity. Ask what happens when a supplier ships ninety-eight cases against a purchase order for a hundred and bills for a hundred. The right answer ends with a cost that is correct per unit and a variance that somebody can see. A wrong answer ends with a margin report that looks fine and a stockroom that is losing money on a line you cannot identify.
For the surrounding evaluation, the six general criteria we use are in ERP evaluation criteria for an operationally complex company. The three tests here sit beside those criteria. They are the distributor's version of one idea: test the system against your own data, not against the vendor's.
Imagine a distributor of cleaning supplies. One product is bought on pallets of 48 cases, each case holding 12 bottles. It is stocked by the case, sold to most customers by the case and to small accounts by the bottle.
Run the test. The receipt of one pallet should add 48 cases and 576 bottles of stock at one cost per bottle. A sale of 5 bottles to a small account should take five bottles off, leave the case count with a visible remainder and price each bottle at that customer's rate. Suppose that customer has a contract price 8 percent under list, and the product also has a quantity break that applies only to cases. The system has to choose a price for bottles, which are not cases, and say why.
Add freight. A pallet's freight arrives on an invoice four days later, equal to 3 percent of the product cost. Spread by weight across the pallet's lines, it raises the cost of the heavy concentrate more than the light one. If the system takes the freight into overhead, both products look 3 percent more profitable than they are, and the contract customer who was priced 8 percent under list may be buying below cost without anyone knowing.
None of that needs a vendor's best demo to find. It needs one product and five customers.
Score each vendor on the three tests with a plain pass, workaround or fail, and write down the workarounds. A workaround is a cost: someone will maintain it, explain it to a new hire and rebuild it at each upgrade. A vendor with two passes and a clean workaround may beat one with three passes that needs your item master rebuilt first.
Keep the test file. It becomes the acceptance test at go-live, which is the point at which most distributors discover that nobody wrote down what the system was supposed to do.
We start with a no-risk discovery: we take a sample of your products, customers and supplier invoices, build the three tests, and show you how your current setup answers them, so you can compare vendors on your own data. You pay for the discovery only if you go ahead with the work that follows. A build comes with a guaranteed estimate, so if we estimate low, we absorb the difference, and Managed Technical Operations keeps the item master and the price rules correct after go-live.
A module list tells you what a system might do. Twenty of your own products tell you what it will.
In a no-risk discovery we look at your CRM, finance and operations systems and who owns each part, and show what a connected system would do differently. You pay only if you proceed. Or see how we approach it.
More on the same problem:
"Something isn't working." "We're outgrowing our tools." "We're afraid to make the wrong move." No-Risk Discovery is a short, practical conversation that gets you clarity before you commit to anything big. We'll tell you if we're a fit. If we're not, we'll tell you that too.
Say who you are and ask. A person replies within one business day.