This week, Google is set to deprecate the long-standing average position metric in favor of its new prominence metrics that offer more transparency and control for search marketers. Kenshoo’s Joe Bernal offers some comforting, final thoughts for advertisers on this shift.
The sunsetting of average position has a lot of search marketers on edge.
This is typically the case when Google makes a big change that at first seems like the end of the world (just Google removing right rail ads, Expanded Text Ad format, or exact match including close variants to see some of the “sky is falling” rhetoric leading up to previous big changes). But preparing and understanding the change should lead to a smooth transition as we move away from average position to the new prominence metrics.
The reason why many search marketers consider this to be a major change is because average position was one of the original metrics in Google AdWords so many years ago…
Average position, a core metric to the first era of SEM
Average position has historically been used as a proxy for how competitive a keyword is and how much room an advertiser has to work with when analyzing its performance metrics. Search advertisers would not only report and act on it, but bidding platforms would also use it as part of its algorithm to determine if it was worth pushing a keyword bid higher to avoid overpaying for the click if the maximum average position threshold was already achieved.
In my early days of paid search, average position was so important on a select group of keywords for the company I worked for that we outsourced a project where the keyword’s average position was captured in a spreadsheet every morning. Our team would then monitor the file daily and if there were any major swings in average position, we would investigate and try to rectify any nuances by modifying the keyword bids when we saw desired positions slip.
Average position would also be taken into account when setting up manual bid rules in an effort to see if a keyword would warrant a bid increase when minimum ROI goals were met or exceeded. As time has passed—and advertisers and algorithms have become more sophisticated—a lot of this manual guesswork has been taken out of our workflows, freeing up time to focus our efforts elsewhere while letting AI take this tedious task of bidding over.
Average position had its faults
Now that average position is being put to rest, the question is why and what should we do about it?
Google has made it clear that average position is not a true gauge of where your ad sits on the search engine results pages (SERPs) in relation to organic results. All that average position tells you is the order or rank of your ad in relation to the other ads—you could technically have an average position very close to 1.0 but be below the organic results without getting as much visibility as you think you would.
Or, you may have a broad match keyword averaging an aggressive position of 1.5 but sitting below the organic results and achieving your ROI goal, so you bid it up to try to achieve closer to a 1.0 position. Ironically, in reality, this tactic could backfire because increasing the bid for the broad match keyword may opt you into more competitive auctions matching to other keywords that may or may not be relevant to your business, thus driving ROI down while not even improving page position. With the new prominence metrics, you can see beforehand if there is even any room to bid higher to achieve position above organic, saving you time and money.
So because of this, Google has introduced the four new metrics:
Impression (Absolute Top) % – the percentage of your impressions that are shown as the #1 ad above the organic results
Impression (Top) % – the percentage of your impressions that are shown anywhere above the organic results
Search (Absolute Top) Impression Share – Absolute top impressions divided by the estimated number of eligible top location impressions
Search (Top) Impression Share % – Top location impressions divided by the estimated number of eligible top location impressions
Essentially, the first two metrics tell you how often your ads sit on top of the page above organic, while the second set of metrics tell you what your share is of the eligible top impressions. Remember, some paid results may not sit above the organic results, as this depends on a variety of factors.
Embrace the change
Now that we have a little background and definition of the new prominence metrics, how can we set ourselves up for a seamless transition from average position to the new metrics?
This would be a great time to re-evaluate your bid strategy if you have been optimizing to average position. It might make sense to move away from relying strictly on average position to a strategy targeting a ROAS or CPA goal.
But if average position is the KPI (especially true for brand marketers that simply need to dominate the top results), then it’s recommended to pull several months of historical data and trend average position next to the new prominence metrics. Chances are, there will be some correlation between these two sets of metrics. Hopefully, there will be a sweet spot between average position and absolute top or top impression share. Whatever that sweet spot may be, it could be used as your target for your target impression share strategy. As a reminder, Google Ads is only making this change and existing Microsoft Advertising bid strategies using average position can stay put as they are keeping average position as a metric and also have released their own version of prominence metrics.
If you don’t use average position as your KPI and instead use performance-based strategies, not much should change. Whether you optimize your program to ROI, CPA, or other goal, optimizing to those metrics should not change even though average position will be deprecated. If you want to determine how much room you have to bid up, that’s when you can use the new prominence metrics to see if it makes sense to increase the bids to try to capture more impression share. But, at the end of the day, your ads either met or didn’t meet your ROI/CPA goals and there should not be any major changes to your strategy.
As advertisers have increasingly become more sophisticated, Google has evolved its offering and realized that the average position metric has been slightly flawed by not informing advertisers if their ad is above organic or not, so it has introduced the new prominence metrics and putting average position to rest.
While this change has some advertisers on edge, with the proper preparation, new strategies should be implemented by swapping out average position and substituting with the new prominence metrics. It’s important to analyze average position alongside the new prominence metrics to ensure a smooth transition.
Kenshoo’s internal calculations and algorithmic decisions have substituted average position with a new model based on the new prominence metrics. It has been in the works for months to ensure a smooth transition once this long-standing metric is deprecated.