Adam Goldberg

Getting a new customer is great.  Getting a new customer that becomes a repeat customer is even better.  When a customer makes a repeat purchase, a marketer needs to consider how to attribute the credit for the future sales back to the marketing that was used to acquire the customer in the first place.   Should the original ad(s) that acquired the customer get any credit for future sales?  Should the original ad(s) only get credit for the first sale?  What if the second product purchased was related to the first product purchased, would that change the attribution model a marketer would use?

During the Attribution Management Forum 2.0 (AMF) on Jan 29th, 2009, we posed that question to an audience of hundreds of senior online marketers and asked them to vote on the attribution model they would use in the following scenario:

In this scenario a consumer clicked on a PPC ad on Monday and the next day purchased peanut butter from that site.  A little over a week later they clicked on another PPC ad and the next day purchased jelly from that same site.

We provided the audience with 3 attribution models they could choose from.  They were as follows:

If you voted for:

A)     You believe the first search should get credit for the sale of the peanut butter and jelly because the 2 products bought are so closely related, the first search should get credit for the sale of the jelly as well.

B)     You believe the first search gets credit for the first sale and a percentage of the credit for the next sale as well. The second search also deserves a portion of the credit for the second sale.

C)     You do not believe in lifetime ad value (LAV) and think that each search gets credit for only the sale that directly follows it.

In almost all of the groups polled, over 60% chose attribution rule B, which favors giving credit for the sale of jelly to the first and second ads.  The exact percentage breakdowns were not chosen in this exercise.

When looking at this scenario there are a few other factors that one should consider. One factor is the type of search ad that was clicked on prior to the sale of jelly.  For example, if it was a branded term, I believe more people would have opted to give full credit for the sale of jelly back to the first ad.  The logic for doing this is that this customer chose search as their preferred method to navigate back to the site that they originally bought peanut butter from vs. using the address bar, an organic result, or a bookmark.

Another factor that might change the way people voted is if they knew the second search was for “jelly”, which is a general search term.  The use of a general search might be an indicator that the consumer was not thinking of the company they purchased peanut butter from at all.  Therefore, if they did not have that company in mind when they did the search for jelly, perhaps all of the credit for the second sale should go to the second ad.

In AMF 1.0, we presented a similar scenario but did not show the type of products that were sold.  In that forum, the majority of the audience still chose to attribute some credit for the second sale back to the original ad.  I would have thought knowing the sale of the second product was complimentary to the sale of the first product would have made more people opt to give all of the credit to the first ad.

These types of scenarios and rules are why we continue to strive to generate some consensus around Attribution Management.  If you would like to learn more about these scenarios or the scenarios from the previous Forum, please visit www.AttributionManagement.com.  Additionally, we love to hear from our audience, so please fill us in on what your thoughts about this scenario may be.

This entry was posted on Thursday, March 12th, 2009 at 9:49 am and is filed under All Blogs, Attribution Management, Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.