A Theory on Optimal SKU Count.

by Jon Hedges

"How many products should I sell?"

That's a question that many managers have confronted at one time or another. Especially managers that have completed an 80/20 analysis so that they know where their profits are coming from, and where they're not. Remember: you are probably not making any money (and probably losing money) on your lowest-volume products.

The question of how many products to sell can only be answered using knowledge of your products, your markets, your customers, and of course, the data. There are plenty of formulas that determine the optimum number of parts to stock, but here is an easy-to-use tool to help you make decisions on which products to sell in the first place based on sales volume (important parts of a decision process should also be gross profit and inventory turns).

Standard deviation, for the statistically challenged (i.e., most of us), can be scary stuff. We'll simplify it here, and if you really want more information on standard deviation you can look at one of the million statistics websites out there. Standard deviation is simply a measurement of the variation, or "spread," in a series of numbers, in this case the total annual sales dollars by SKU.

Step One: List Your Products

List your products and their annual sales volume in a spreadsheet, using Microsoft Excel or similar spreadsheet program. Your spreadsheet will look something like this:

	A	B
1
2	Part Number	Total Sales$
3	10-100	$545,034
4	10-200	$845,453
5	10-300	$1,298,568
6	20-100	$343,406
7	20-105	$235,245
8	30-100	$184,456
9	30-200	$28,033
10	40-100	$15,545
11	40-200	$18,505

Next, calculate the standard deviation on the sales dollars. The formula in Excel will be:

=stdev(the cells you're analyzing)

and you will enter this formula in a separate cell (in this case we'll use cell B1). "Data" is the simply column that contains annual sales dollars. If the spreadsheet shown contained all the data you were analyzing, your formula would be =stdev(B3:B11) and we'll enter it in cell B1.

Your spreadsheet will look something like this:

	A	B
1		$417,899
2	Part Number	Total Sales$
3	10-100	$545,034
4	10-200	$845,453
5	10-300	$1,298,568
6	20-100	$343,406
7	20-105	$235,245
8	30-100	$184,456
9	30-200	$328,033
10	40-100	$15,545
11	40-200	$18,505

Step Two: Apply a Factor

Now that you've calculated the standard deviation for your product sales, now what? This is where you select a factor that best represents your business and multiply it by the standard deviation.

If you are a small specialized manufacturer with a simple product line, multiply the standard deviation by a factor of 0.2. If you are more of a broad-line manufacturer and/or multi-market manufacturer, multiply the standard deviation by 0.1. If you are a WD or a very broad line manufacturer, you may need to multiply by 0.05 or less.

In this case, the standard deviation for all parts sales is $417,899. Let's take that and multiply it by a factor of 0.1 and we get $41,789. That becomes our cut-off point. We'll look at products that have annual sales of $41,789 or less for further analysis (and if you're doing this calculation on your company's product line, I'll wager that if you take the number you came up with, and include all products with annual sales over that number, you have identified at least 90% of your total annual sales dollars). In this example, we would take a hard look at part numbers 40-100 and 40-200.

The point is, the number that you come up is a theoretical cut-off point. Determine which factor makes sense for your company, and revisit this calculation on an annual or quarterly basis. Annual sales volume below this cut-off point requires further analysis and some decision making. Ask yourself should you consider continuing to sell a product that has annual sales below that number? (Or maybe the product is new and does not have big sales volume yet).

Accountants may argue the point that products should be evaluated on gross profit and/or turns, and that's a valid point. This formula is calculating a cut-off point based on consumer demand and contribution to sales only.

Conclusion

Remember, there is no magic rule that determines how many products you should sell. That's what your brain is for. But this calculation establishes a quantitative cut-off point, a benchmark, so you can look at products below the cut-off and start to make correct business decisions for your company and your product lines. In most companies, this calculation will probably identify at least 90% of your total annual sales.