In the last two posts 1, 2, I tried to discuss how measurement and false metrics drive optimization towards low hanging fruits and in the end degrade ad effectiveness. I would like to follow up with a short example of how the issue extends into the paid search (e.g. Google Adwords) channel.
Search traffic is split up into two parts; the organic, free traffic and paid traffic. The following chart illustrates the difference.
For the paid part advertisers are allowed buy individual keywords (search terms that users might type in) on a cost per click basis. Each keyword sells for a different cost per click depending on competition, relevancy and click-through-rate. Usually the advertiser picks the keywords with lowest cost per click.
Keywords can be clustered into at least two broad categories: Brand and Generic keywords. Brand keywords contain the brand name of the advertiser while generic keywords do not and just relate to an (unspecific) inquiry. From a consumer journey perspective, generic terms are searched in the beginning of a purchase decision process, while brand keywords are used towards the end. Generally, that means that brand keywords have a much lower cost per click than generic keywords. Hence advertisers buy brand keywords all the time.
##So where is the catch?
It is safe to assume that advertisers are organically listed in the top positions for brand keywords, while they might not even appear on the first results page for generic terms. As the paid advertising section is shown above the organic section, a majority of users click on the paid links. Hence, an advertiser buying her “own” brand keywords is cannibalizing her organic, free traffic with paid search traffic.
The following chart illustrates this:
The share of incremental traffic for buying generic keywords is much higher compared to buying brand keywords. Using proper experimentation one can figure out what that share is and calculate the cost per incremental visitor as a more reliable metric.
The question arises why this is usually not done. I see three main reasons: (i) incentives, (ii) additional work and (iii) Google’s quality score.
For the team managing paid search it is much easier to communicate low costs per click (CPC) readily delivered by the system compared to relying on a derived metric that needs constant experimentation. As most people are familiar with CPCs they are also commonly used to benchmark channels and/or team performance. Hence increasing CPCs is a difficult story to tell. A part of the cost per click on a keyword is determined by the quality score, Google assigns to an individual advertiser. The lower the score the higher the cost an advertiser has to pay per click. Even though Google does not openly discuss the algorithm behind the quality score, it is pretty clear that the Click-through-Rate (which is higher for brand keywords) improves the score. As a result of their policy brands are forced or at least given a decent incentive to buy their brand keywords.
To sum up, in the paid search channel, the system is designed in a way to ensure that brands buy keywords for users who are very likely to click/convert anyways.