PPC in 2010 was a different world. Match types meant something, algorithmic learning was barely out of infancy, and the entire industry was up in arms about Google’s new Quality Score metric. The idea was (and still is) that Google would give extra weight to unique ads and copy that matched up well with keyword and landing page content, effectively lowering the cost of a click compared to your competitor’s ads. The intent was to make ads more relevant and therefore generate more clicks for Google (never mind that competitors would do this too, creating a race to the bottom for every industry).
Everyone wanted to maximize their quality score over competitors, so everyone started asking how they get keyword content to match up with ad content when ad content was held at the ad group level.
Enter SKAGs, shorthand for Single Keyword Ad Groups. Advertisers would create an ad group for every keyword in their campaign (sometimes all match types together, and sometimes even each match type in a separate ad group) and saturate ad copy with the respective keyword for each ad group. In theory, this would make the quality score go up, and Google would “discount” the ad buys.
Did it work? To an extent, yes.
Google cleverly never went so far as to release how much the quality score adjusted the final click cost, so we’ll never know for sure, but there was a hint of change in the overall data that said it was worth doing, if not overdoing.
So, what happened to this major trend that took the industry by storm? In a word, automation. Google started releasing bidding algorithms that handled keyword bids at the ad group level or even campaign level, which meant segmenting keywords was ineffective. The more keywords and ads were grouped together, the more effectively and quickly machine learning generated value.
Next, Google released Responsive Search Ads (RSAs), which effectively did the leg work of matching ad copy to keyword content automatically. With RSAs, we still write ad copy (indeed, even more than before), but it’s more piecemeal now. Google takes the best ad copy match to each individual search term that triggers an ad to show.
And the nail in the coffin was the dissolution of keyword mapping to search terms. When exact match truly meant exact match, it made sense to split keywords into their own siloed ad groups so that keywords could be tailored directly to ad copy. Now, the match types and mapping of keywords have been blurred to the point of being almost irrelevant, and matching ad copy to every possible “variant” of a keyword isn’t possible. Because every keyword is effectively broad match, ad copy relevancy has become a function of RSAs alone.
Lastly, after all these shifts, the industry was simply overwhelmed. Making an ad group manually for each keyword was eating up man hours across the industry. Automation was welcomed as a benefit.
Now, it’s 2022, and SKAGs are a thing of the past. Automation is handled at the campaign level more and more, gradually making split ad groups more about organization than performance.
However, there has been a lot of industry talk of new strategies to take full advantage of the new Performance Max (PMax) campaign type. Instead of Single Keyword Ad Groups, could the next trend be Single Product Campaigns for PMax? Single Keyword Campaigns?
It will depend on what Google does with PMax campaigns and what data advertisers are given on this campaign type. Where automation was once a welcome change, many industry experts are starting to balk at the lack of valuable data within the highly-automated PMax campaign type and are starting to split these campaigns into these granular strategies for the sake of data clarity.