Outlining the problem
The issue with the above paradox is that the consumer does not always recognise the power that a marketing campaign has over his or her future actions. The consumer, especially on the internet, tends to think that he/she is all powerful and that every decision that is made is as a result of their own thoughts and judgements. Influencing a potential consumer without him or her even realising is one of the many benefits of display advertising and indicates the power that implementing a display campaign can have, even if the consumer does not agree.
Introducing the different types of conversions
There are two types of conversions that are recorded in DoubleClick: post click conversions and post view conversions. All conversions that can be attributed to a display campaign are classified as either post view or post click.
A post click conversion is a conversion that occurs after a user has clicked on an ad. A post view conversion (or view through conversion, if you are used to AdWords terminology) refers to a conversion where the user has previously been delivered an impression but not clicked on it.
A conversion is only attributed to a given impression or click within a period of time that your account manager specifies, called a look back window. The length of time used for a look back window can be altered dependent on a specific product's typical time to conversion.
Within this look back window the rule for attributing a conversion to a click or impression is "clicks trump impressions". Therefore, for example, if a user views 6 of your creatives and then clicks on one before converting at a later date then the impressions that were served before the click will not be given any credit for the conversion.
The significance of a conversion occurring after a click is immediately evident as the user has clearly been attracted to your ad on a website, clicked on it and then converted. In short, there are no issues when it comes to attributing credit for conversions to ads that are generating clicks.
Difficulty arises when attempting to attribute conversion credit to an ad after a user has been delivered one of your creatives but not clicked. As a display account manager this is troubling. Given that around 90% of conversions that we accrue in our campaigns are post view, this issue threatens to call into question the whole purpose of the campaigns that we run.
Now that we have got the definitions and theory sorted we can take a look at a real world example in an effort to highlight the importance of measuring both post click and post view conversions.
Below I have set out four very simplified examples of a user's path to conversion:
- A user is served one of your ads but does not click on the ad and then the next day goes straight to your site and converts.
- A user is served one of your ads, but does not click or convert. Then the next day the same user is served another ad, clicks on the ad and converts on your site.
- A user is served one of your ads, clicks on the ad and then the next day converts on your site.
- A user is served one of your ads, but does not click or convert. Then the next day the same user is served another ad and clicks on the ad. The following day the same user is served another ad but does not click on it. The user then goes straight to your site and converts.
Hopefully in light of your now superior DoubleClick conversion knowledge you were able to spot that examples 2, 3 and 4 would be recorded as post click conversions and only example 1 would be recorded as a post view conversion.
With greater awareness of how display campaigns work it becomes increasingly difficult to argue that examples 2 and 3 are worth more to us than example 1 when example 2 was actually initiated by a user viewing one of your ads. If this user had not witnessed the first ad then it is incredibly unlikely that he would then go on to click, never mind convert, after being served the second ad. The same logic goes for example 4 where the user was reminded of your brand with the final ad and yet this impression is not attributed any credit for the subsequent conversion.
If we were to only recognise the importance of post click conversions then we would be missing the fact that display advertising works even without a user directly interacting with one of our ads. A user who sees one of your ads may not be directly affected by the ad, hence why he believes that he has very rarely been influenced by a display advert, however once the user has witnessed the ad the brand is stored in the known brands part of his head to be called upon at a later date when he is in a more appropriate position to convert.
We must not lose sight of the fact that display advertising tends to be a lot higher up the conversion funnel than search and the majority of the time it is used as a push marketing mechanism, whereas search would be classified as a pull mechanism.
Building on this, the user that we are generally targeting in our campaigns has never interacted with the advertiser's brand before, and more often than not may never have even heard of the brand before.
Unfortunately in DoubleClick and Google Analytics, which both run off last click attribution models, we only ever know what a user's final interaction with our campaign was. If only there was a way to view a more detailed breakdown of a user's path to conversion which includes impression data!
As luck may have it, there are actually three different tools that can be used to begin to understand the value of marketing channels that have impact at different levels of the marketing funnel. To solve the nightmare scenario of credit not being attributed where it is deserved, the tools that can be used are:
Each of these tools work in slightly different ways and each has their own strengths and weaknesses; that's a debate for another day!
If you need help analysing the ins and outs of your conversion data or if you have any other questions regarding DoubleClick then don't hesitate to get in touch!