Our blog.

Google's Improved Ad Rank Calculation

Profile Image Posted

Steve Baker

Chief Analyst

I’ve always been an admirer of the Quality Score algorithm that Google use. In a nutshell, it rewards adverts that are relevant to the user and penalises the ones that aren’t, using the click through rate to determine how relevant each advert is (taking the position into account, of course). There are other ‘quality-related’ elements, but this is the main part of the calculation.

It’s always struck me as a good system. It pushes adverts that searchers are more likely to click on up the search results (which is good for searchers), ensures that Google make as much money as possible (which is good for Google) and ensure that advertisers with well-structured and managed accounts are rewarded (which is good for world class search marketing agencies!).

And in the past, Google have multiplied the Quality Score by the bid to determine your position (this value is called the Ad Rank) and the actual amount that your clicks cost – but now things have changed a bit.

Instead of determining your Ad Rank, and hence your position and cost per click solely on your click through rate and Quality Score, now they have decided to introduce another element – the Ad Extensions and Formats that you are using.

Google argue (quite reasonably) that an advert with a click-to-call extension is more likely to be relevant to somebody using a mobile phone, whereas an advert with sitelink extensions would be more relevant than a click-to-call link for somebody searching on a computer. But is this really the underlying reason for what they are doing? Are Google trying to improve the user experience?

To be honest, it’s quite difficult to tell. This may be a good thing for advertisers (and world class search agencies), it may be good for searchers, and it may be good for Google’s profit margins – or a combination of the three. OK, it’ll almost certainly be good for Google’s profit margins, but the other two are less clear-cut! Really, it all depends on how exactly these changes work.

Let’s start from the advertiser’s perspective. There aren’t many good reasons not to have some kind of advert extensions running – they allow advertisers to deep-link into various pages on their websites that may be particularly relevant to searchers, and (more cynically) take up more real estate on the homepage.

Google say that they will show the extensions that are the best-performing and most useful combination, so the safest bet is to have every type of extension that you can and let Google determine which to show. Of course, if Google gets it wrong, then it could end up hurting your click through rate, but I suspect that they’ll get this right (as it’ll cost them money otherwise).

If Google bases the decision on the past performance of your advert extensions, then what happens if you change your sitelinks, or add location extensions? And if they do look at the past performance of sitelinks, should you promote sitelinks that are going to get a higher click through rate in order to reduce your cost per click?

From the perspective of the advertiser (or their agency), without more detail on exactly how this will be implemented, the only conclusion is to use all relevant forms of advert extension, and hope that Google choose to show the right one. Hopefully, they won’t get too creative, and will simply look at which types of extension are the most effective for a given search, and promote that type of sitelink.

So if somebody is searching on a mobile phone, adverts with click to call buttons should be promoted. If they were searching for a garage, maybe location extensions would be more appropriate. And on computers, adverts with sitelinks will be pushed ahead of those without any.

If Google start to get too aggressive about the approach, and start comparing the performance of the sitelinks on different adverts, they could skew the results significantly.

So for most advertisers, it’s not really a good or a bad thing – you probably wanted to use sitelinks anyway, and for world-class search agencies (who always include sitelinks if possible), it’s probably a good thing, as their adverts will benefit to the detriment of lazy advertisers. You could argue that encouraging advertisers to use sitelinks will increase the click through rate of the top adverts, and reduce the click through rates of the adverts in lower positions, but it’s difficult to say to what extent this is true (after all, the side adverts aren’t being pushed down the page the way that the organic search results are).

So what about the Google users? Does this benefit them? Theoretically, where Google have historically promoted adverts with higher click through rates, this new adjustment will push adverts with lower click through rates but sitelinks into higher positions. I don’t really see how this is really going to improve the quality of the search results.

In addition, by encouraging advertisers to use sitelinks, Google are pushing their organic search results further down the page (something they’ve been working on for years) – so searchers who aren’t looking to buy something are definitely going to get a worse experience.

Overall, my feeling is that it’s a good thing for Google (so long as they don’t end up losing their users to Bing), good news for diligent advertisers, bad news for users, and bad news for SEO (they seem to be having a bad time of it recently!).

A more cynical analyst than I may suggest that by trying to make their paid search adverts more visible at the expense of organic search results, they are continuing their war on search engine optimisation, but I suspect this is simply Google trying to increase the click through rates on the adverts that pay them the most. I doubt that it’s going to make a massive difference to the overall auction, but only time will tell on that.