Summary and Questions

Any questions about what I covered up to this point? This is a natural stopping point, because what I've done is I've illustrated what exactly is micro-marketing. I explained it in terms of what we want to do to come up with a better pricing strategy at the store level. I gave you some sense of what the data looks like and I walked through some of the models. What's going to be important after the break is to think about the impact of the terms of the posterior profit function, because that's what's really going to drive these pricing strategies. Yes?

Q (Gatsonis):
What if Dominic's competes with each other? Here you are assuming a hierarchical conditionally independent model. So you are assuming independence between the stores. What do you do if they compete with each other? You would need a spatial model.

A:
Right, what I would want to do is augment the model and include the prices not just from my store but also from all the other stores. The problem is that since Dominic's essentially follows a uniform pricing strategy right now, it'd be impossible to parse out the effects of my pricing effects and prices of the other stores. What may be interesting is to augment the model by including Jewell's prices. Problem is, I don't have Jewell's prices or the other competitor's prices. So it's something that's missing from the model, but I'll come to other experimental results that sort of coroborate what the model is doing, that say that this does make sense, that even though the competition and their effects are there, they're not as important as you might think a priori.

Q:
Why is the 64 oz cheaper than the 96 oz?

A:
If we go back to the table we can see that what I've done is to standardize this stuff all to 64 oz, so it's not that the 96 oz is cheaper than the 64 oz.

Q:
I would have thought 96 oz was cheaper per unit but that doesn't happen either.

A:
Well, what happens is that the 64 oz are promoted all the time, so if you think about the base level price, the 96 is cheaper than the 64. But the 96 is promoted maybe a tenth of the time, the 64 oz is promoted every week in at least one of the brands, so if we think about the average prices, the average 64 oz is less than the average 96.

Q:
My other question was, would it be possible in your model for some stores to take certain brands out? For example, if I could make a larger profit margin on Minute Maid, maybe I'd want to take three other brands out and try to force someone in a particular direction.

A:
It would be possible -- that would be more of an assortment difference, while what I'm really interested in is just pricing. Let's figure out what the pricing effects are, and then let's back up and think about what the demand effects are. Just the base movement in demand. The other thing is, you might want to use this to think about where you would want to place a store, based on what the price sensitivity is going to be? You know, there's a lot of things that tie into it.

Previous Section

Next Sectionxo

Go to Table of Contents

Go to written version of paper