In order to free your bidding strategy from the paralysis that comes from accounting for all possible variables, certain assumptions have to be made. Commonly, one of these assumptions is that increasing a keyword bid increases the ad position, and increasing the ad position increases the CTR. But it’s not always true.
To test, I looked at five high traffic, highly competitive keywords in an account. Four of them followed this model very well.
But the fifth keyword didn’t seem to get it. KW5 responded exactly the opposite – increasing the bid actually decreased the CTR.
For this keyword, increasing the bid did increase the position. The impressions for each data point are all more or less the same, and conversion rate was more or less the same.
So what’s the deal? Breaking down the universality of this assumption brings up a lot of questions, and it defies common sense. One factor is that in higher positions, this keyword shows on more of Google’s partner network, where the CTR is lower. But two important facts are that this pattern is seen only for this keyword, and changes in impressions were only weakly correlated with changes in bid. Is there a difference in the serve ratio between Google and its partners for this keyword? Is there something about the keyword itself that draws eyes down the page? How much do competitor ads on this term affect CTR differently? If we kept raising (or lowering) the bid, what would happen to CTR?
The bottom line is that despite the reason for the unique keyword behavior, it has to be considered when making bid changes. Does your account have any of these Giffen Keywords that violate the unproven “laws” of PPC Bidding?
We need to make certain assumptions in order to make bid changes, but it’s important to know when and where these assumptions break down or else changing bids can harm the account. So pick a couple top keywords and watch the effects of your bidding. And don’t think that bid tools are exempt… they can face the same challenges as the manual bidder.