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.

Goodnight Sweet Prince – A Farewell to Google Average Position

Announced yesterday in a Google Ads Help post, Google will be sunsetting the average position in September 2019. In lieu of average position, Google encourages advertisers to make use of newer metrics like absolute top impression share to assess the competitive positioning of ads on the page.

Such newer metrics should prove fine for allowing advertisers to adjust bidding strategy in light of how highly they appear on the page for relevant queries. However, the elimination of average position does eliminate one helpful diagnostic tool for assessing changes in the Google search results page.

Average Position Wasn’t Very Reliable, but Neither is Impression Share

Part of the rationale for eliminating average position is that impression share metrics give a ‘much clearer view’ than average position does of how prominent an ad is on the page. While these metrics do prove valuable, it’s hardly the case that they are very clear.

The specifics of how such metrics are calculated seem to be ever-shifting with updates such as changes to the definition of close variant matching, and it’s clear at times that advertisers are seeing shifts in impression share as a result of changes in the auction rather than advertiser-specific changes in competitive standing. For example, Google Shopping impression share skyrocketed across the board for Merkle advertisers in Q4 as Google Shopping impressions exploded, to the extent that it was clear that the increased impression share was not the result of our advertisers’ strategies.

In the case of both average position and impression share, neither should be heavily relied on in most cases since neither reflect actual costs or sales driven by ads, and rather reflect only the appearance of the ad on the page. While there are specific segments of accounts for which it might be reasonable to focus on how prevalent ads are, more often than not marketers will want to use less vanity-centric metrics and focus instead on something more indicative of the actual profitability of ad units. Google’s new click share metric is a helpful advancement but is also likely to be impacted by changes in the auction.

From Google’s standpoint, one must wonder if it believes it can get higher bids from advertisers by shifting focus over to impression share metrics. One Merkle advertiser sees an average position of 1.0 on its key brand term, but only around a 75% absolute top impression share. Will advertisers caught in this situation be more likely to boost bids when they don’t have a clean 1.0 average position metric telling them the ad is sitting pretty? It’s quite possible.

Even if average position isn’t really the first metric most advertisers turn to, it’s still been a nice arrow in the quiver for paid search analysis over the years.

Google Average Position Has Proven Useful in the Past in Analyzing Other Shifts

Way back in Q1 2015, Google click growth for Merkle advertisers slowed to just 0.2% Y/Y. At the same time, first page and top of page minimum bid estimates were more than double what they had been a year prior, and text ad impressions were in decline after years of reliable growth.

It was difficult to understand what was going on, and one might have guessed at the time that Google had run out of runway to continue increasing search ad traffic, though that’s a laughable thought given how the last four years have played out on this front. There was also the possibility that an influx of competition was hurting results of long-standing advertisers.

However, looking at average position we were able to discover that our ads were actually being displayed higher on the page.

Q1 2015 Google Non-Brand Text Ad Average Position

Given that it would be unlikely that average position would move up the page if it were in fact the case that competitors were sweeping past our advertisers, we were able to judge that the click and impression trends we were seeing were more likely caused by Google itself reducing the number of ads featured on the page.

Fast forward to Q4 2018, and the latest Merkle Digital Marketing Report explained why a decrease in the share of text ad clicks coming from positions lower on the page over the past few quarters can put upward pressure on cost-per-click (CPC), since higher position ads come with higher CPC.

Q4 2018 Google Non-Brand Position 3 & 4 Click Share

Google was unable to comment on if the adposition ValueTrack parameter used to create this chart will also be taken off the table, but it stands to reason that that might be the case if Google wants to push advertisers away from position metrics.


With the sunset of average position, it will be more difficult to get at some kinds of insights that have proven valuable in the past. While advertisers that are shooting for specific locations on the page should find alternative metrics like absolute top impression share adequate for assessing the competitive landscape and making optimizations, these metrics don’t shed light on some deeper trends that average position has helped identify over the years. Additionally, some advertisers might feel more compelled to raise bids based on impression share metrics when average position may have made them feel comfortable with bids in the past.

It’s certainly not the end of the world, but a sad day nonetheless, if only for nostalgia’s sake. Here’s to you, average position.