Price Sensitivity Thresholds -- A new paradigm for supermarket profitability

I was in Walmart the other day, and I went to pick up a bottle of hydrogen peroxide. It is used in our household to kill the germs on a sponge or pot scraper. When they are put away wet, they are an excellent breeding ground for germs, and a shot of hydrogen peroxide sanitizes everything nicely.

The usual price is .99 for a bottle. I was shocked to see it priced at $1.27. This was interesting in many ways. First of all, it is a trend with Walmart. They quietly jack up the price on stuff that you don't notice, but when they raise the price, they drop it from say $1.29 to $1.27 to give the impression that they are .02 cents cheaper than the competition. The psychology is intriguing.

That got me to thinking. Walmart probably uses data-mining to determine these things. They probably crunch numbers and go through reams and reams of data to determine which products they can quietly raise the price on, while not losing sales. In other words, they determine which products have a low sensitivity to price thresholds.

If they jack up a loaf of bread by .29 cents, the public will notice. But the public doesn't buy hydrogen peroxide every day or every week, so if they do a huge price increase, pretend it didn't happen by lowering it a couple of cents from the new high, then over the hundreds and hundreds of stores, they get a real kick to their profitability. They must have many products that have a sensitivity threshold that consumers don't really notice. Or if they do notice, then they don't mind paying the few extra cents due to the convenience of already being in the store, and they aren't going to drive elsewhere for .27 cents.

This got me to thinking. I bet that the supermarket is full of products like this. With a bit of number crunching and data-mining, one could figure out what these price insensitive products are and jack up the prices by a few pennies.

Supermarkets are probably doing this already. I am told that the margins are low on supermarket products because of the intense competition. However, I am sure that this will be a burgeoning field in data mining as the box stores get bigger and bigger. In contrast, my gut instinct tells me that this would work for small boutiques as well.

I have a algorithm for this concept. Maybe it's time to incorporate it into a software program.

No comments: