This post could also be titled: PPC Referrers Taking Too Much Credit?

The problem with tracking online sales is trying to determine why someone made the purchase. Was it because they saw our banner ad and just remembered our name? Was it because they found our ad on some SERP when searching for a product? Or did they come to our site directly and make a purchase but only because they previously came to our site through a paid ad and bookmarked the site later or just remembered our name?

I know when I search the web, I do it fairly fast and thorough. For most products, I know where to go. For instance, Half.com is where you buy books, tigerdirect.com & newegg.com are where you buy computer products. Dell is where you buy LCD monitors, etc. But a lot of the time, when I’m searching for something which there are many suppliers of at various competetive prices (cell phone & iPod accessories, etc.,) I look through many of the paid advertisers on the SERPs. I personally will open up 5-10 tabs of different sellers, and the ones with the highest prices get closed. So eventually I’m left looking at the product pages of several sites with the best prices but I usually don’t remember the name of the business/website selling them until I have bought from them a couple times. Honestly, when I get my product in the mail, I usually don’t even look at the who seller was, I’m just happy that I got the best value.

My point is that for any company has many different online marketing efforts going simultaneously at any time (banners, ppc, shopping portals, etc.,) we most likely have customers who visit us through more than one of our campaigns. The problem lies in the fact that each advertiser is biased to themselves. They all want to report how well they’re doing for you even if it means, they aren’t the most likely reason that a customer made a purchase. If we were to try and track our campaigns through each individual advertisers system, we would most likely have an overlap and over-report revenue generated. By using one system to track them all, we can eliminate some advertiser bias and level the playing field.

Ironically we use Google’s Analytics tracking system, but I would rather trust that they aren’t biasing themselves over any other of our advertiser’s systems. Consider this scenario - someone clicks a Yahoo! paid ad, and finds that we have pretty good prices, so they go do something else for a while, come back, and either types in our url (if they remembered it) or clicks the bookmark they made earlier. When they make a purchase, who should be given the credit? I would credit Yahoo!. What I am finding is that Google is not crediting Yahoo!, they are most likely crediting the sale to a ‘direct referral’ indicating that the cutomer came to the site directly (by typing in the URL.) This means to me that Google Analytics is only giving credit to referrers that cause a conversion before the user closes their browser after having clicked an ad in the same session because when they come back, they’re either doing it through a bookmark or by typing it in. Now in this situation, Yahoo! will show that they caused a sale because their tracking cookie was in there, and maybe rightfully so, but now we have a discrepancy in conversion reporting.

Another potential situation which might cause discrepancy is if a user clicks on a ppc ad at one point, and then comes back to the site through another advertiser afterwards before making a purchase. It’s possible that due to bias, both advertisers would think they should take credit for the purchase, being that the user came to our site through them at one point or another. I believe that if this scenario were to occur, the most recent referring source should get the credit.

Transparency in Google Analytics’ specific functionality will help to at least clear up what exactly is happening, and maybe some standards for how the effectiveness of online advertising should be reported will help get us all on the same page.

If anyone has any idea as to the specifics of how Google Analytics handles these situations, please comment.