Measuring delayed response to online advertising

(Click here to expand)

Third-party ad serving has tremendous benefits for brand owners by providing better visibility into the effects of their online advertising campaigns.

Ad servers allow marketers to identify end users—and the actions they have taken—by storing server-side profiles on each user that lands on their site after clicking on an ad (known as a click-through), or even those who did not click on the ad, but visited the site within a specified amount of time after being shown the ad (known as a view-through).

Through the use of pixel tags, marketers can track back-end actions like purchases from an ecommerce site, and tie that activity back to the advertising message (or series of messages) that initially drove the interest. Thus allowing you to optimize your campaign against conversion rates rather than click-through rates.

An even greater (but often ignored) benefit of 3rd-party ad serving is being able to understand the time it takes (on average) for a user to convert into a customer—after they’ve clicked on your banner ad. When a user clicks on your ad but does not purchase something in that same session, this does not mean that your advertising message was wasted on that individual. This person may return to your site later that day, or later that week, to complete his purchase. An ad server allows you to attribute the ultimate purchase to the advertising campaign, even though the purchase conversion itself was delayed.  [So, if you are not implementing pixel tags for tracking these repeat visits, you will lose this valuable information and ultimately underestimate the performance of your campaigns.]

But, a word of caution, when using pixel tags to track delayed response to your advertising, it is easy to distort the view of your campaign optimization efforts. As you can see in the chart below, if you measure total # of conversions from week-to-week, it looks like the performance of the campaign is steadily improving.

But remember, some of the conversions SHOULD ACTUALLY be attributed to advertising that occurred prior to the week you’re measuring (due to delayed response). As you can see in this example, 100 conversions (25 click-thru and 75 view-thru) of the 250 measured for week 2, are actually delayed response to advertising that occurred in week 1. This example actually represents NO improvement in the campaign’s performance from week-to-week.

As you become more sophisticated in tracking advertising effects, it becomes increasingly important that you attribute the right effect to the right effort.

Comments are closed.