Aaron, The only way to get a good comparison is if the numerator (in your case Internet Ad Budget+BDC labor) and denominator (Internet Sales) are calculated and tracked identically.
Let's say twin brothers run the internet departments at two similar stores in the same market. Each dealership sells 50 cars a month out of a 100 car inventory. Both stores sell about a dozen cars that can be directly attributed to an email or phone call to the internet department, but the second store has an aggressive tracking program in place and can account for 25 more cars sold to "walk-ins" that visited the Internet before they came in. The second store classifies those walk-ins as internet sales and has a cost per sale that is a third of the first store. The two stores are having identical success, but because management calculates and tracks the denominator differently, one brother looks like a hero and the other a heal.
Complicate this further, by giving our brothers some cousins across the country. One cousin works at a store that is diligent about stocking Low Market Days Supply vehicles and prices them to market. The store has worked hard and is up to 18 inventory turns a year. The other cousin works at a store with high gross mentality that consistently prices $3,0000 to $4,000 above market average in order to give them room for the occasional "Home Run". They don't hit as many as they used to, but they are still pretty proud of their PVR. They are only hitting 4 turns a year. The first cousin gets to brag about his low, low cost per internet sale at the family reunions while the second cousin gets laughed at. The funny thing is the second cousin runs the best operation in the family, he's just hamstrung by the store's management philosophy (and happy the carrying costs and wholesale losses his store experiences don't affect him!).
There are so many moving parts, so many factors at play here, that it makes store-to-store comparisons difficult.