Without accurate stock level information in your computer, effective inventory management is impossible. No matter what sophisticated tools you have in your inventory management system, if the computer thinks you have 100 pieces of an item, and there are really only three on the shelf, the system won’t replenish your inventory when it should or order the right quantity.
Unfortunately, most distributors only verify the stock balances in their computer once a year, when they perform a physical inventory. During the physical count, every item is counted and, if necessary, the balance in the computer is adjusted to reflect the actual quantity on the shelf. Even if you assume that the physical inventory results in an accurate count of each stocked product (a big assumption for many distributors), how long do the counts remain accurate? One month? Two months? Six months? Eleven months after the physical count, what percentage of stocked products still have an accurate available quantity in the computer?
In order to receive all of the benefits from a good inventory management system, stock balances must be at least 97% accurate, every day of the year. This means that the actual available quantity of every item in the warehouse is no more than 3% greater or less than the available quantity displayed on your computer inquiry screens. If the computer says there are 100 pieces of an item on the shelf, there should be no less than 97, nor more than 103. Note that 97% is the minimum acceptable standard. One hundred per cent accuracy is the optimal goal that you should strive to attain.
The best way to ensure that a minimum of 97% accuracy is maintained is to continually count your products. That is, count part of your inventory every day, and count each item several times per year. This process is called “cycle counting.”
If the answer is so simple, why doesn’t every distributor abandon their annual physical inventory and cycle count? Do distributors enjoy the annual wall-to-wall count so much that they refuse to give it up? I don’t think so. In fact, many distributors have implemented cycle counting programs only to abandon them when they see that their inventory accuracy hasn’t improved, it may even have gotten worse!
For all of its benefits, there is one logistical problem which makes cycle counting more difficult than a complete inventory. That problem is material movement. Think of the environment that exists (if you follow the guidelines in the previous two articles) when you conduct an annual physical inventory:
- All stock receipts have been placed in their proper bin location.
- All printed sales orders and transfers for stock material have been filled.
- Computer records for these receipts, sales orders, and transfers have been updated.
- All customer returns have been processed and the material returned to stock.
- No customer orders are filled nor material moved until the counting of all products in the warehouse is completed
All material movement has stopped. You are counting a fixed target.
It would be nearly impossible to recreate these conditions every day when you cycle count. After all, you have to continually receive material and fill customers orders to remain in business. As a result, cycle counting is like trying to count a moving target. How difficult is this? Well, go down to your local pet store and try to count the goldfish in a tank. How do you know that some quantity of the products you are counting today isn’t sitting on your receiving dock having just been entered in the computer? Or, is it possible that an order for an item being counted has just been filled but the computer records won’t be updated until tomorrow? It’s no wonder that many well-intentioned distributors throw up their arms in frustration and abandon cycle counting programs after only a few days and weeks.
Like the procedures necessary for a successful annual physical count, there are several necessary guidelines for a cycle counting program to produce the desired results:
Decide which cycle counting method to use. A good plan for determining the frequency with which products are counted ensures that no items will be skipped, or counted more often than necessary. There are two methods for determining when to cycle count items that can form the basis for a good plan:
- The Geographic Method
- The Ranking Method
Using the Geographic Method, you start at one end of your warehouse and count a certain number of products each day until you reach the other end of your the building. This results in counting all of your items an equal number of times per year. Because you are systematically examining the contents of each shelf and bin, the Geographic Method facilitates the “discovery” of misplaced or lost material, especially the stuff that has been “stashed” between bins. If you implement a geographic count system, try to count each stocked item at least four times per year.
The other method is the Ranking Method. Research shows that the more often a product is received or shipped, the less accurate its computer stock balance. This makes sense. Every time someone goes to the bin is an opportunity for a mistake (or to coin the new term, an “unquality event”) to occur. For example, material can be put away in the wrong bin, or the wrong product can be taken to fill an order. The Ranking Method directs you to count the items with a large number of dollars flowing through inventory (i.e. with the highest annual cost of goods sold) more often than slower-moving products. The ranking is based on “Pareto’s Law” (named for the late Italian economist Vilfredo Pareto) which basically states that, in general, 80% of the results of any process is produced by 20% of the contributing factors. Applied to inventory, this means that approximately 20% of your inventory items are responsible for 80% of your stock sales. Though Pareto came up with this theory nearly 100 years ago, it still is generally true. And, for the reasons we talked about, we want to count these items (i.e. the top 20%) more often:
- “A” rank items (responsible for the top 80% of sales) count six times per year
- “B” rank items (responsible for the next 15% of sales) count three times per year
- “C” rank items (responsible for the next 4% of sales) count twice per year
- “D” rank items (responsible for the last 1% of sales) and products with no sales count once per year
Though not as effective in finding lost material, the Ranking Method usually works best for maintaining accurate inventory counts. Because the primary purpose of cycle counting is to verify the quantity on-hand of each item, most distributors prefer the Ranking Method.
When should cycle counting be performed? Most distributors experience some time during the day when material is not moving. Usually this is just before, or after, normal working hours. The chance of counting errors is reduced if cycle counting is performed during these “off” hours. Consider counting for one hour each morning before your warehouse opens for business. Or, if you prefer, counting can be done for an hour each afternoon, after the last order has been filled.
Most distributors are open for business about 250 days per year. This means that if you use the geographic cycle count method, and have 10,000 items in inventory, you will have to count about 160 products per day:
10,000 items counted four times per year = 40,000 counts
40,000 counts / 250 days = 160 products counted per year
Though how fast items can be counted is dependent on many factors, one conscientious, experienced person can usually count between 100 and 150 products in an hour.
The number of products counted each day using the Ranking Method will vary depending on the number of products that are assigned to each rank. To give you some idea of how this method works, we’ll look at another distributor with 10,000 stocked items:
- Rank “A” contains 2,000 products
- Rank “B” contains 3,000 products
- Rank “C” contains 4,000 products
- Rank “D” and the dead stock category contain 1,000 products
Notice that, in our example, the Ranking Method requires fewer products be counted each day:
| 2,000 “A” items counted six times per year | = | 12,000 counts |
| 3,000 “B” items counted three times per year | = | 9,000 counts |
| 4,000 “C” items counted twice per year | = | 8,000 counts |
| 1,000 “D” items counted once per year | = | 1,000 counts |
| Total | 30,000 counts | |
| 30,000 counts / 250 counting days = 120 products counted per day | ||
Greater inventory accuracy resulting from 25% fewer counts! No wonder rank based cycle counting is a popular alternative to geographic counts or an annual physical inventory.
Determine who should count. When you conduct a physical inventory, you have to count every piece of every item in your warehouse in a short period of time. To accomplish this task, most distributors draft, and put to work, anyone in their organization who can breath and count at the same time.
Cycle counting is different. Only a limited number of items are counted each day. To ensure the counts are accurate, only knowledgeable, experienced warehouse people should do the counting. Sure, you’ll have to put them on a slightly different schedule. That is, they’ll have to come in early or stay late. But the one or two hours of their time spent counting each day will be well worth the investment. Think about how good you’ll feel when you, and all of your employees, have confidence in the stock balances in your computer!
As I hope you can see, there are some definite advantages to implementing a cycle counting program. In our next article, we’ll look at cycle counting procedures, as well as the differences between counting with and without bar-coding equipment.