Skip to content

Hot Topic – Goodbye Average Position

Google have announced the end of a legacy metric for Google Adwords: Avg. Position. The deadline is set for the end of 2019. The announcement comes as what used to be a primary metric for advertisers has depreciated in usefulness over time and here’s why.

For those that don’t know, Avg. Position in Adwords is the metric that supposedly shows the physical position that a text ad is shown on the actual search engine results page (SERP) aka Google.  Ultimately, Avg. Position in Adwords is a statistic that describes how your ad typically ranks against other ads.

Bids are higher for higher positions. The higher the position, the higher the bid. This “average position” is for the time window that is set and can therefore change month to month depending on a multitude of factors. Such as your Ad Rank, which is a combination other metrics: quality score (landing page experience, ad relevance and expected CTR) and your bid. Your Ad Rank can change, causing its position on the page to fluctuate as well, so your average position can give you an idea of how often your ad beats other ads for position. However, the most important thing is to find what’s profitable for you, which might not be to show in the top position. By seeing how your ad typically ranks when it’s triggered by one of your keywords, you can try to influence your position by changing the keyword’s bid.

This all sounds very good, right. A super handy metric. However Average Position should have been named auction rank really as this would better reflect its meaning. The word position is actually better understood as a relative position compared to other advertisers on the SERP and has nothing to do with a physical position on the page where the ad is shown.

Back when Google used to have Ads to the right of the page and you could tell the location on the page because ads showed consistently, this was useful and a more appropriate name. Advertisers and marketers knew where the ad would be on the page, and could adjust bids accordingly depending on their strategy and budget. Advertisers that always wanted to be at position one would be more aggressive than those that would be happier slightly lower and able to drive more traffic. Now ads can appear on page one but underneath the organic listings, diluting the page 1 status further.

A major change to advertisers came from an intended improvement to the user experience. Google added a relevancy score into the equation, so now all advertisers could bid to be above the organic listings and the ads on the right side were scrapped (see below).

Being above the organic listings was historically reserved for big companies with big budgets billed on a cost per thousand impressions model (CPM) vs a cost per click (CPC) model. Now it is different. To get to the top of the page, an advertiser must be relevant to meet the criteria to be ranked.  An average position of 1-8 is generally on the first page of the search results (SERP), 9-16 is generally on the second page, and so on. Average positions can be between two whole numbers. This is all more confusing than we originally thought and Google Shopping made it even less straightforward muscling in at the top for e-commerce searches.

Relative ad position compared to competitors is not completely helpful and nor is auction rank. Advertisers are more concerned with where their ads physically are than where they place above or below other advertisers. What once was a transparent metric, and one that influenced bid strategies has naturally lost its gusto over time and that’s okay. Google is an ever evolving platform. So, Google will replace the redundant avg. position with four sparkly, new and better metrics:

  1. Impression (Absolute Top),
  2. Impression (Top),
  3. Search (Absolute Top) Impression Share
  4. Search (Top) Impression Share

These will give advertisers the control and clarity to bid according to absolutes that correlate with page position.

Watch out for my next blog to tell you all about these new and exciting metrics.