enhanced sitelinks: measuring the relevancy
Hot damn! Google has just announced the launch of Enhanced Sitelinks and I am pretty darn impressed. Traditional sitelinks (below) have served a role by encouraging a click on the ad rather than as a selection of alternative landing pages. They provide additional information and real estate on the SERP (which many users may interpret as a signal of authority), but don’t demand enough attention to draw users in to click on them. This means conversions (and ultimately advertiser’s re-investment in Google) are potentially jeopardised as users go through to the ad’s main landing page, rather than the alternative landing pages in the sitelinks which may be more relevant for the individual user.
So with the new sitelinks being a lot bulkier and imitating the natural listings even more, they should manage to live up to their intended purpose (assumed purpose) and drive users through to more relevant landing pages.
measuring the relevancy
All good so far – happier customers less likely to bounce after reaching the site, but Google has also said that the content of the enhanced sitelinks are actually just ‘related’ ads from your campaigns, though they’re not clear how they select these ads. Google claims it picks ads that are ‘closely related’ to your sitelinks, but how does it measure this relevancy? Google measures the relevancy of keywords to search queries by using crowd-sourced click through rates, from both your account and other advertisers. If it applied the same technique, this would mean testing several of the campaign’s ads until it could establish a pattern of reliable click through rates, except as other advertisers don’t use the same ads as you, it would be based entirely on your data – and collecting data/ testing costs money. Is it unfair for you to foot the bill of this testing when we’d already stipulated strategically chosen sitelinks?
one possible method
Perhaps Google is venturing into the semantic realm and identifying keywords that exist in both the ad copy and sitelinks? Very much doubt that though. More likely, they’re identifying ads which received a greater CTR improvement from the addition of sitelinks compared to others ads in the rest of the campaign. At the moment, an ad receives a good CTR when matched to a particular keyword and deemed relevant and eligble to display sitelinks. So equally if the particular ad-keyword-sitelink combination receives a good CTR, then the ad can be deemed relevant to the sitelink and eligible for enhanced sitelinks. This is of course just a theory and I think there’s going to be a lot of testing, theories and discussions before the truth starts to become clear.