What Role Do View-Through Conversions Play in the Display Network?

Just about everyone who has used the internet is familiar with seeing those pesky Remarketing ads. In fact, every time I tell someone that I work in the digital marketing field, it almost is directly followed by the same question, “Are you the ones that make those creepy ads that follow me around on every site?”

Over the years, I found that it is not just my friends and family who are annoyed about Remarketing ads, but CMOs and marketing executives are often asking the same question.  Does Remarketing really work? If someone visits a site, is seeing an ad in the corner of their computer screen really going to influence them to visit it again?

Remarketing campaigns can produce two different types of conversions.  The first conversion type is a direct conversion, where a user views a remarketing ad on a separate site, clicks directly on it, and then converts. The other, more controversial conversion, is a View-Through Conversion (VTC). This occurs when a user views a remarketing ad on a separate site, does not click on it, and at another time, navigates back to the original site through another channel (organically, directly etc.) and converts.

The question that then arises is, do VTCs really represent incremental sales and leads? How do we exactly know that the ad generated the conversion?  With VTCs, the individual can end up back on the original site, days or sometimes weeks after they first saw that specific ad.  What are the chances that that specific remarketing ad really prompted that conversion? How are we so sure that the user was not already planning on returning to the site? Additionally, a user may not have even seen the ad. Many times advertisements are located below fold on a site, or hidden away on side banners. It is likely that the user could have missed the ad completely.

Our agency set out to conduct an A/B test designed to prove whether or not VTCs are really influencing our performance.  To do so, our agency worked with Google to set up a PSA test.  Google carefully segmented our audiences into two groups: one group that will exclusively see an irrelevant public service announcement (PSA) advertisement, and another group that will see our client’s image ad. The audiences were set specifically so that there would be no overlap between who sees what ad.  Check out the results below:

Of all the users that were part of the remarketing list, and did not get remarketed to (saw the PSA ad), we generated 235 VTCs. Of all the users that were remarketed to normally (saw a client specific ad), we produced 306 VTC. Therefore, we can assume that our remarketing ads produced an incremental 71 VTCs (if we subtract 235 from 202).  The 71 VTCs makes up roughly 23% of overall VTCs (306). Of course this percentage will vary with each client and industry.

We can now apply this 23% to our total VTCs in order to more accurately determine the number of truly incremental VTCs.  For example, if in a given week, a remarketing campaign produced 133 VTCs, it is likely that 30 of those conversions are legitimate. If you are an ecommerce company, you can apply standard AOV numbers to calculate the expected incremental revenue that those VTCs added to your campaigns. Whether you are a strong proponent of VTC, or a skeptic, it is important to consider the findings of this test and learn that we can quantify the impact of VTCs.  These findings should be considered in reporting and they should influence optimizations within Display and Remarketing campaigns.  It is important that we do not neglect these forms of conversions that can sometimes be considered illegitimate or non-incremental.  Raising bids of display placements, keywords, or other targeting based on VTC volume has proven successful in impacting our overall client conversions.  Ultimately, we know that across Display and Remarketing campaigns some of the VTCs are incremental and some are not.  As advertisers, it is critical that we understand this rate of incrementality so that we’re properly attributing the correct value to each of our marketing programs.  Furthermore, each network and each campaign type (display vs. remarketing) may produce different rates of incrementality.  Therefore, it is important to test various networks and campaign types to understand how VTCs behave within each.

To learn more about Display and Remarketing strategy, please contact us by email at sales@synapsesem.com or by phone at 781-591-0752.