We use cookies. You have options. Cookies help us keep the site running smoothly and inform some of our advertising, but if you’d like to make adjustments, you can visit our Cookie Notice page for more information.
We’d like to use cookies on your device. Cookies help us keep the site running smoothly and inform some of our advertising, but how we use them is entirely up to you. Accept our recommended settings or customise them to your wishes.

CPCs & Average Position Subtleties

Q: Why is my average position lower on the page, even though my average CPC is higher?

We’ve been asked this question a few times recently, and, since the immediate answer isn’t intuitive, we thought we’d clarify.

Occasionally, an increase in a term’s average CPC can occur simultaneously with a decrease in average position. This seems counterintuitive. Why would higher CPCs mean lower page positions? If the bid is higher, shouldn’t the ad be higher up on the page?

Yes and no. For exact terms, we don’t see this happen; the relationship between CPCs and average position is generally positive. We do see this with broad and phrase match keywords, however, since in both cases a keyword is involved not in one auction, but many. Given the nature of the PPC competitive landscape, each potential broad-match query will have its own auction, so a higher bid for ‘widget’ could move ‘widget’ up five positions, but only move ‘red widget’ (to which ‘widget’ is broad-matched with) up one space.

The overall decrease in average position occurs because a higher bid could lead to an increase in the number of auctions (different search queries) the keyword is now eligible for.  For example, let’s say that at 20 cents the broad-match ‘widget’ ad has too low a bid to appear for the search query ‘blue widget’. If we increase the ad’s bid by 10 cents, though, its new bid of 30 cents is high enough for ‘widget’ to cross the first-page threshold for ‘blue widget’. Since the increase in bid is small, the ad will most likely appear low on the page. For this example, let’s presume that Google displays the ad at position 10.

Since ‘average position’ is determined by adding the numerical position of each impression’s ad and then dividing by the total number of impressions, this outlying, position 10 spot will skew the numbers and lower the reported overall average position for ‘widget’. If, after increasing its bid by 10 cents, ‘widget’ on the query ‘widget’ moved from position 3 to 2 for 10 impressions, while on the query ‘blue widget’ it moved from not-appearing to position 10 for 2 impressions, then the average position for the term ‘widget’ would be 3.3 (40, the sum of the positions, divided by 12 impressions). In this situation, we increased the bid by 10 cents, but, to our dismay, the average position subsequently dropped from 3 to 3.3.

So, the ostensibly ‘worse’ average position is misleading. For no individual search query did the average position drop. However, the overall average position dropped because you now have your ad appearing in low positions for new search queries that were previously too pricey. This idea can be corroborated by analyzing your traffic. Did average position drop but impressions go up? If so, then you’re probably dealing with an increase in the number of auctions.

This is a good example of why the average position metric should be viewed with a grain of salt. While it is helpful in providing a general idea where your term is (such as position 1.3 versus 24.4), the number itself isn’t very useful in determining where exactly your term is for any individual auction. Nor does a lower average position necessarily indicate that the term is being pushed down the page. Indeed, such as in our example, a lower average position actually means you have more exposure and more traffic.

Join the Discussion