Your keywords are about to start matching to queries they’ve never matched to before.
That’s right: Following up last September’s decision to make exact match keywords less exact, Google announced earlier today that phrase match and broad match modifier keywords will soon begin matching to search queries that include same-meaning close variants.
After providing some context and going into the details of what’s happening to your keywords, we’ll share the perspectives of WordStream experts as well as folks from the PPC community.
Some background: The evolution of close variants
Let’s get something out of the way: In the context of search advertising, a close variant is a query that’s (in theory) synonymous with your keyword. We can trace the origins of this back to 2014, when Google stopped allowing advertisers to keep their exact and phrase match keywords from matching to things like accidental plurals and misspellings.
Then, in 2017, Google further expanded the definition of an exact match close variant by allowing your exact match keywords to match to queries that ordered the terms differently or included function words like articles and prepositions.
The common idea behind these decisions is that an ad for dog day care is still relevant to someone who searches “dog day xare” or “day care for dogs.” Though the point of keyword match types is to give advertisers much-needed control over the queries that trigger their ads, Google has seen fit to balance that control with some flexibility.
There’s a difference, however, between accidentally misspelling a word and deliberately typing a different word altogether. That’s why search marketers were so thrown when Google announced the further extension of exact match close variants to include same-meaning words. All of a sudden, a business targeting the exact match keyword [yosemite camping] could have their ads shown for queries like “yosemite campground” and “camping in yosemite.”
Google’s defense of this decision was three-fold:
- Even if the semantics of a query doesn’t exactly match your keyword, the user intent may be all the same.
- Loosening the definition of exact match enables businesses to connect with a larger pool of relevant prospects.
- Nobody has time to find every single keyword that’s relevant to their business.
The results? A mixed bag. As our own Mark Irvine reported last November, some advertisers benefited considerably from their heightened exposure on the SERPs—others didn’t. The commonalities across these winners and losers were (1) an increase in the proportion of their impressions and clicks being attributed to close variants and (2) an increase in overall costs.
OK—now that we’ve contextualized today’s announcement, let’s dive deeper into it.
The extension of same-meaning close variants
At some point in the next few weeks, your phrase match and broad match modifier keywords will begin matching to search queries that include same-meaning words: synonyms, paraphrases, and the like. Given the importance of this change, a quick refresher is in order:
- Traditionally, a phrase match keyword matches to queries that include the entirety of the keyword in its specified order. For example, the phrase match keyword “dog day care” is eligible to match to queries like “boston dog day care” and “dog day care costs.”
- Traditionally, a broad match modifier keyword matches to queries that include each term specified in the keyword in any order. For example, the broad match modifier keyword +dog +day +care is eligible to match to queries like “day care for my dog” and “take care of my dog during the day.”
Up until now, phrase match and broad match modifier keywords have been eligible to show for close variant queries like accidental plurals and misspellings. Now, like exact match keywords, they’ll also be eligible to show for same-meaning close variants like synonyms and paraphrases.
Google provided an example for each match type in their blog post. Take a look at this one, which illustrates the changes coming to broad match modifier:
And here’s the phrase match example:
You may be wondering: “Wait. What if I were already targeting both ‘lawn mowing services’ and ‘grass cutting services’ as phrase match keywords? Which keyword would get triggered?”
Don’t worry—Google is making the appropriate change to keyword selection preferences. Even if you’re targeting two separate keywords that are synonymous with one another, the semantically relevant one will continue to be entered into the ad auction.
Reactions from WordStream HQ
Navah Hopkins, WordStream’s services innovation strategist and one of PPC Hero’s top 25 PPC experts, noted that given the limit of 10,000 negatives per campaign, this change makes it difficult if not impossible to protect a true SKAG structure. (SKAGs, or single keyword ad groups, restrict your ad group to a single keyword and a single ad.)
You can’t optimize your ad copy for a single keyword when that keyword is liable to match to a wide range of queries. Instead, Navah advises, “focus on the themes of your keywords and create copy that speaks to the true pain point of the user.” She added: “This change is actually quite liberating from a management standpoint as you’ll have fewer ad groups and campaigns to build. Also, it’s yet another signal from the ad networks that health indicator metrics are being depreciated in favor of ‘We have data—trust us’ metrics (e.g., top of page instead of average position).”
I also spoke with Kristina Simonson, who manages our own paid search and social accounts. Here’s what she had to say:
Considering the shift towards automation, I’m not surprised to hear about this update—and honestly, I’m not mad about it either. As Google points out, 15% of queries are new; unfortunately, humans don’t have the ability to predict brand new queries that a prospect may search on a given day. As the manager of our own account, I’m happy to know that Google will look for opportunities to expand our reach to further relevant queries.
Reactions from around the PPC community
What do other digital marketers have to say about the news? Generally, they share a concern that the weakening of keyword match types limits control and hinders their ability to optimize their ads according to different types of queries.
There’s that word again: intent. Indeed, whether the broadening of phrase match and broad match modifier proves beneficial to advertisers will depend on Google’s ability to successfully decipher the intent behind users’ queries. Some, evidently, are skeptical.
Here, “nuance” is synonymous with “intent”—Julie is essentially echoing the sentiment expressed by Melissa in the previous tweet. I’m intrigued by the idea of a “bad match reporting mechanism.” If advertisers had a direct line to Google’s algorithms that they could use to call out misinterpretations of user intent, that would certainly be helpful.
Though the notion that match types are dead is obviously an exaggeration, Jon has certainly gotten to the heart of the issue: As the divisions between keyword match types become less and less clear, advertisers have less and less control over which queries trigger their ads. True—for some, gaining exposure will lead to more profit overall. But for others, it’ll lead to greater costs and diminished click-through rates.
Aaron Levy, for his part, has a much more positive outlook on the situation than some of his peers—one that echoes Mark’s conclusion that last year’s changes to exact match yielded a mixed bag of results. Whereas some will reap the benefits that come with more impressions and higher click volume, others will bear the costs. Which camp you fall into depends on a variety of factors (e.g., goals, industry, colloquialisms, etc.)
Looking ahead
As these changes to phrase match and broad match modifier roll out, it’s imperative that you keep an even closer eye on your search terms than usual. Though results will vary across verticals, it’s likely that you’ll see an uptick in impressions and clicks from users whose queries are irrelevant to your business—that is, an uptick in queries that you should add as negative keywords. As long as you’re vigilant about keeping irrelevant users away from your ads, you should be able to mitigate the potential cost increases that may come with these changes, and focus on the potential upside.
More broadly, I recommend taking Navah’s advice to heart: Do your best to write ads that—above all else—speak to the unique pain points your prospects are experiencing. Does it become more difficult to do that as the intent behind your keywords and the intent behind the queries triggering your keywords grow further apart? Of course. But despite its imperfections, Google is pretty darn good at discerning what users are trying to accomplish. If you do what you can to be helpful and relevant, you’ll be fine.