Why Google Ads is Sunsetting the Average Position Metric and What it Means for Advertisers

April 4, 2019
Sunset 1

In February of this year, Google announced that it will be sunsetting the Average Position metric in September 2019. The Average Position metric has been considered critically important in understanding how an ad ranks against competitors in the ads auction. This means advertisers will need to take a close look at some soon-to-be outdated bidding strategies, leveraging 4 new “prominence metrics” that were introduced in November 2018.

First, let’s look at why Google is sunsetting Average Position

Google explains that Average Position tells us where our ads stand in the search auction compared to competitors’ ads. But this metric does not necessarily tell us where those ads appear on the SERP (Search Engine Results Page). With Average Position, a position of 1 does not necessarily mean ads are displayed above organic search results.

Google takes the position that in general, it’s more important to understand where an ad is served on the page, rather than how an ad is performing in terms of outbidding competitors. In this context, the Average Position metric is less meaningful when compared to the new metrics discussed below.

How we will measure ad exposure moving forward

Google is promoting new metrics that give different information about where ads are showing:

  • Impr. (Absolute Top) %: The percent of your ad impressions that are shown as the very first ad above the organic search results.
  • Search (Absolute Top) IS: The impressions you’ve received in the absolute top location (the very first ad above the organic search results) divided by the estimated number of impressions you were eligible to receive in the top location.

ROI·DNA Note: Based on the definition, these metrics become a proxy for position 1:

  • Impr. (Top) %: The percent of your ad impressions that are shown anywhere above the organic search results.
  • Search (Top) IS: The impressions you’ve received in the top location (anywhere above the organic search results) compared to the estimated number of impressions you were eligible to receive in the top location.

ROI·DNA Note: These metrics become a proxy for positions 1-4, as they appear to overlap with the Top Impressions definition:

What we will be losing

We will lose the ability to quickly see our position on the search results page in relation to our competitors. Average Position was often used as a proxy to identify bidding success in competitive industries, and as a signal to make bidding changes.

This begs the question whether the team at Google believes they can encourage higher bids (and more advertising spend) by shifting focus from Average Position to Impression Share (IS) metrics.

For example: with an Avg. Position of 1.0 and a Search Impression Share of 75%, we know that our ads are served above any competitors’.

Without the context of an Avg. Position metric, the 75% Impression Share will likely create pressure to bid more aggressively in an effort to increase Impression Share, even though the ad may be beating out competitors.

This is another step in Google’s continued shift towards promoting automated bidding strategies, removing some of the ability advertisers had to fine-tune and optimize.

Our Perspective

These metrics seem to add increased focus to the top 4 positions on the SERP, because the only information that an advertiser will have about performance below the organic results will be a rough estimate based on the above-fold metrics.

If, for example, 60% of our Impressions appear between positions 1-4, then there is nothing more that can be deduced about the remaining 40% other than the fact that an ad appeared below the organic results.

There is no way to know whether that was in position 5 or in position 8.

Our scope of visibility will be reduced. We are going from Average Position data available at the decimal level, where for position 1 we can see metrics from 1.0 – 1.9, to now only viewing performance in 3 bands:

  • Top position (above organic results): Position 1
  • Anywhere above the organic search results: Positions 1 – 4
  • Below the organic results (subtracting the top metrics from 100): Below Position 4

With Average Position, 8 ad positions on the SERP translated to 80 data points, from position 1.0 – 8.9. Losing that visibility means that whole markets will be making bidding decisions based on just 3 pieces of information. This will no doubt create increased bidding competition as whole markets compete for those 3 position bands.

For advertisers who are lucky enough to have access to the Average Position metric today, we recommend downloading data from November 2018, when these new metrics were released and comparing them side-by-side to Average Position metrics in order to create a more meaningful baseline and reference point of these metrics for your account.

If we can add more meaning to these metrics in our current advertising environment, this will allow us to make more meaningful business decisions when the clear guide post of Average Position is removed.

We recommend revisiting any bidding strategies that use Average Position as a bidding signal and any automated bidding/bid optimization tools that rely on Average Position as a bid adjustment signal.

Rest assured that we will be reviewing these new metrics alongside Average Position and closely monitoring any impacts as the change takes place in September. Please feel free to reach out with any questions!

Hey There!

Thanks for reaching out.

Job Title
What're you interested in?

We'll be in touch shortly.

Until then, why not browse some of our work?

See our work