Inventory is the largest financial asset in a wholesale distribution company. For most organizations, it represents at least 75 percent of the cash in company. This is the big one. Given this fact, it is shocking how little attention is paid to getting the shelf count to match the computer count. When I discuss this subject in distribution management classes, several distributors admit to having somewhere around a 60 percent to 70 percent accuracy ratio. Of course, there are others who boast accuracy in the high 90th percentile, but I suspect that is more based on how they arrive at the percentage versus a real matching of shelf to system variance. If our warehouse was filled with cash (I mean stacks of currency), wouldn’t we want to know how much was in there every single day?
As I alluded above, I believe that some distributors are fairly delusional about their level of inventory accuracy. It is really difficult to achieve accuracy in the high 90s if you are working on a manual warehouse system. A warehouse management system (WMS) with barcodes and phaser, like scanners, makes it a great deal easier to achieve these lofty levels. But let’s face it, even those using advanced technology can screw up a warehouse count. For those boasting high levels of accuracy, I suspect that they are basing their variance on dollars versus units.
Basing the inventory accuracy on a dollar variance is a really poor way to determine the accuracy of the inventory on your shelf. This type of reconciliation is meant to satisfy bean counters and auditors. No offense to my friends who count the beans, but you know exactly what I am talking about. Let’s say that you have a supplier line with an average inventory value of $100,000. When a count is performed on the line, the inventory variance is only $2,500. In this scenario, the inventory accuracy would be reported at 97.5 percent. This makes everyone in the financial world grin from ear to ear. On the other hand, I suspect that a closer examination of the inventory count sheets would paint an entirely different story if we were to measure the unit variance of individual items.
Here is the tougher way to look at this: If the shelf count does not equal the system count, the item is inaccurate. I live by the pass/fail ruling here. Let’s say that the above $100,000 line contained 300 SKUs. When the items were counted, 43 of the items had some form of unit discrepancy. The item could be higher than the system count or lower that the system count. Either way, the system count did not match the shelf count. By the unit variance measure, this line could only boast an 85.6 percent accuracy ratio. I suspect that these figures might not be appealing to the financial partners in the business.
When people ask me why it is better to measure the inventory accuracy by units versus dollars, I ask them two simple questions:
1. How do our customers request material from us, in units or dollars?
2. How do we order products from our suppliers, in units or dollars?
This usually drives the point home.
Now that we have a better understanding how to measure inventory accuracy from a practical perspective, the real question is: Why is it important to be so accurate? Now, this may seem like a silly question, but if it is so silly, why do we continue to perform so poorly when it comes to matching shelf stock to system stock? Perhaps we should look at the costs, either financial or reputational, associated with sloppy inventory practices.
When teaching this subject, the first cost I discuss is labor. Yes, inaccurate counts add time to our material handlers, but it can also have an effect on our customer service labor cost. Let’s say that a customer calls our customer service representative and asks for 100 pairs of gloves. Our CSR pulls up the inventory level and sees that our system shows 100 pairs of the particular glove. Here is where a decision is made. Do we trust the count or do we cover our backside? In my experience, most CSRs would put the person on hold, and either walk out to verify the count or ask another person to verify it. Let’s go a step further; what if the customer was asking for 90 pairs and the system showed 100 on the shelf? Would your CSR roll the dice? It all depends on their threshold of trust and if they have been recently burned. I believe that this lack of trust in our inventory integrity will add significant labor dollars. We are really cycle counting by customer request, and I suspect that this is probably the most expensive way to do it. For a mid-sized distributor, this could be an additional body.
The next cost I typically point out is reputation. Let’s use the scenario from above. In this scenario, our CSR decides to roll the dice and tell the customer to come on down and get them. The customer jumps in their vehicle and drives a considerable distance to come down and pick up the 90 pairs we just promised we had on the shelf. When the customer arrives, we actually have 83 pairs available because the sales people have been showing them all over town and failed to tell anyone they were pulling samples from the shelf. Yes, this really happens. How do you think the customer will view your company when you can’t fulfill the request as promised? You have now become an unreliable supplier, and I doubt that you will be the first phone call in the future.
Finally, I believe that inaccuracy will drive up costs from the purchasing department. For those of you with multiple locations, stock inventory replenishment is usually accomplished from a central purchasing point. When replenishment decisions are being made for locations away from the central office, the inventory investor does not have the luxury of verifying the shelf stock. Sure, they could call someone at the branch to verify the count on an item, but let’s not kid ourselves. If the investor is buying for a branch that is notorious for low inventory accuracy, they are going to cover their backside and buy a little extra. Nobody gets chewed on for loading a little more on the shelf. If you let them run out, the phone starts ringing off the hook. Multiply this padding over thousands of SKUs in multiple locations, and I think you get the picture.
Inventory accuracy takes time and resources. Cycle counting every day is labor intensive, requires organization and attention to detail. Many distributors don’t want to invest the money to do it right. I would argue that you waste far more money doing it wrong. If you really want to start getting your arms around that inventory asset, make a commitment to find out how much you have on the shelf — every day. If you need help getting started, just ask. The first phone call is free.