Operations Research - Does the independent variable affect sales dollars?

#1
Does the independent variable affect sales dollars?

I have five branches in my company. Each branch has a showroom. Each showroom does not have the same mix of items. Some items are in all showrooms, some are in a few, and some are just in one showroom.

I want to see if an item's sales dollars at any certain branch (Example, Branch A) is affected by it existing in the showroom at Branch A or not. I'm not sure which test to use.

Attached is a screenshot of what the spreadsheet of my data looks like. An "x" under Br1, Br2, etc. indicates that the item in that row has a spot in that branch's showroom.

Any help would be appreciated! Thanks!
 
Last edited:
#2
Hi,
Apologies, I am finding it a little difficult to get my head round this. Do the crosses in the table relate to items or to the branches?

I think you need to provide further detail on the items, for example whether or not they are in show room (for example, does no cross mean the item is not for sale at that branch, or that it is not in the showroom for that branch)? Also, are the sales related to those specific items, or all sales, and does the cost per item vary?

Sorry I couldn't be more helpful.
 
#3
Thanks for asking! Sorry I wasn't more clear at first. The crosses in the table mean that there is a location in that branch's showroom for the item in that row. Below I've posted an extended image for illustration. Basically, the product is in the first column. Everything following the product in the same row is in regards to that product (showroom locations, sales dollars).

If there is a cross, then the item in that row has a location in that showroom. For example, Product A has a location in the showrooms of branches 2 and 5. Product B and C are only in the showroom of branch 2. Product D is in every showroom for every branch. If there is not an "x", it means that the item does not have a location in the showroom, but it is still for sale from the warehouse.

The sales are related to each specific item. Therefore, the overall sales for Product A in Branch 5 (both in showroom and warehouse, as indicated by the "x") are $401,763.10. The sales for Product C in Branch 4 (only from warehouse, as indicated by lack of an "x") are $38,886.39.

The cost per item DOES vary. Both the cost of goods on my end, and the price to the customer. I am trying to figure out if a product having a location in a showroom positively or negatively affects its overall sales.

Thanks again for your consideration! Hopefully this further detail helps! Please let me know if anything else needs to be clarified.
 
#4
I did try one way of comparison. Specifically, took the total 12-month sales across all branches for each specific item (the sum of the five sales numbers in one row), and using that total, I found expected sales dollars using specific multipliers for each branch (multipliers as decimals equaling 1.00). The multipliers were based on previous sales dollars volume out of each branch compared to overall sales dollars across all branches.

Then, I used formulas to find the percentage difference (positive or negative) from the expected sales dollars to the actual sales dollars for each product in each branch. For example, the percentage difference for Product A in Br1 was calculated by using "(Actual Product A BR1 - Expected Product A BR1) / Expected Product A BR1." However, to subjectively compare all of these percentages to see if they are significant would be way too arduous with the volume of products I am working with, and an objective answer is obviously preferred if possible.

Thanks!
 
Last edited:
#5
Okay, I think I understand a bit better.
The first thing to say is that you may see differences in costs based dependent on whether the item is or is not in the showroom, but that does not mean that this difference is due to it being in the showroom (cause or correlation).

The first thing I would do is calculate the average sales per branch for each product with and without being in a show room. I.e:
Showroom sales (av per branch) / Non show room sales (av per branch)
Product A
Product B
etc

Looking at this will give you an impression of any different for things being in the showroom. You may have to adjust things based on the total size of each branch (i.e if one branch has, overall, double the sales of another, divide by 2)

To look at the effect of being in the show room at a particular branch I would look at the data this way:
Showroom sales all products (average per product) / Non show room sales all products (average per product)
Branch A
Branch B
etc

You can go on to do some different stats tests after this, but I would start here to get a better understanding of the data.

hope that helps
 
#6
That is a wonderful idea. Unfortunately, I am unable to pull data on sales for an item ONLY out of the showroom. The report that would need to be run would be near impossible to write, and would probably crash the server :rolleyes:

Therefore, I only have access to overall sales for a product between both its showroom location and warehouse location. Then, as shown, I have information on whether these products have a location or not in any specific branch.