Earlier this year Google introduced a new AdWords match type functionality called broadmatch session-based. It’s not a separate match-type per se, but an extended feature of the existing broad match type that looks back at earlier searches that has happened in the same session.
This is another step of Google AdWords personalitsation of search results. For example if an advertiser have the broadmatch keyword ‘antivirus‘ and for a first search a user types in ‘antivirus‘ and gets exposed to the ad, for their second search for ‘windows xp‘ they may again get exposed to the ‘antivirus‘ advert.
So why is Google providing this new functionality? A first cynical view is that as Google makes money from advertising they will want to keep showing for most possible number of searches. However one need to consider that as advertisers only pay if a user clicks on an advert, it needs to still be relevant to provide value for both the advertiser, the user and Google.
One issue with Google’s experimentation of expanding the broadmatch functionality is that currently there is no way for advertisers to op-out of this functionality. So if your analysis of search terms prove that the broad match is not converting positively, the only way to avoid showing for these session-based broadmatches is to  reactively add them as negative keywords or switch of all broadmatch keywords.
By analysing a couple of different campaigns I notice that the share of session-based searches vary widely depending on the vertical. What does however stand out is that many times these session-based broadmatches have a much stronger CTR. A likely explanation for the higher CTR is that the advert that is trigger through session-based broadmatch connects better with the audience as it’s further down the search funnel which makes it stand out more in the search results. Whether or not it converts positively is however a very different question and from my analysis is not as conclusive yet.
Although its implementation is the topic of controversy amongst search marketers, the session-based broadmatch does at least not affect the quality score for advertisers keywords, thus can neither have a positive or negative impact on the keyword cost-per-click.
What’s your experience of the session-based broadmatch?
Related posts:
{ 1 trackback }
{ 0 comments… add one now }