Working with POS data

As we speak with manufacturers about they are handling POS data from their retail customers we find they are having trouble with the “noise” inherent in the data.  One of our clients get a very typical EDI 852 file each week with units sold and units on hand for each UPC for each store.  This is great data, but it takes up about 40 pages in a standard row and column format.  The issue of course is that it takes a long time to sort through and arrive at actionable information.  There is simply too much noise in the data.  And the problem is magnified when dealing with several retail customers at once.

We find it is much better to approach the analysis in a top down rather than bottom up fashion.  Top down analysis involves calculating sell-thru and weeks supply for each item.  Based on viewing these metrics one can much more easily spot the trouble areas than attempting to review line after line of sales and inventory with no context or relative performance information.  

When working with your POS data start by identifying leading indicators for your data like weeks supply.  Leading metrics are actionable and add needed context to your decision making.  In this way you can be much more efficient when dealing with lots of POS data.

Next talk to your customer about how how they analyze performance.   Some retailers look at sales by item and style, especially in apparel.  In soft goods this is less important because they have longer lasting SKU’s and more rigid replenishment guidelines.  By understanding what your customer is measuring and how often, you can align your metrics so you are both scoring performance in the same way.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: