Since its beginnings, paid search has focused on keywords to show relevant ads that match searchers’ intent. Companies back then tried to figure out which words people use to look for products or services and then used those keywords to target potential customers with advertising. This simple and straight forward approach worked great on paper, but in practice, advertisers have found that it is hard to determine which keywords to target and which ones are a waste of time and resources. Research shows that only 6% of keywords produce conversion and sales, and the rest of them are advertisers trying to guess what words their customers will be typing to find them. Almost two years ago, Google announced that they are making changes to Google Ads match types, and that phrase match and broad match modifier keywords will include close variants of keywords you are bidding on. They intended to achieve their goal via function words and reordering.
What are function words?
Function words are prepositions, articles, and conjunctions. Google Ads can now ignore these words in a query, in an exact match keyword, or match what it considers similar intent function words. They call this “rewording” and say that it should happen only if the change does not affect the meaning of your keyword.
What is reordering?
Another way Google can apply close variants in Google Ads match types is through reordering. Your search will now result in ads that contain keyword phrases made from those words but in a different order if that doesn’t change the intent.
“Exact” doesn’t mean “precise” anymore (Thanks Google)
When Google Ads first introduced close variants, they were applied to broad match keywords, allowing Google to match singular and plural variations and misspellings. If we disregard some of the issues that have occasionally occurred when plurals boast a different meaning from their singular counterparts, this change simplified keyword building. Advanced advertisers still had an exact match run to lean on when precision was crucial. And now, sadly, that precision is fading away.
This change could increase revenue for Google Ads since around 15% of search queries on Google are unique and haven’t been used in the previous six months. With individual queries, it is highly unlikely that an advertiser would have added those keywords to its account since the queries would not have shown up in search query reports and keyword tools. Now, Google has more options when matching keywords to these types of searches.
Machine learning – A profit optimization engine for Google
Google is investing a lot of resources in machine learning and artificial intelligence capabilities. With them having to handle trillions of queries annually, they are putting a lot of faith and opportunity for revenue in artificial intelligence to determine how much a query can change before losing its original meaning and intent. Google stated that: “Early tests show advertisers may see up to 3% more exact match clicks on average while maintaining comparable click-through and conversion rates”.
However, as with any Machine learning platform, data is required to drive meaningful results. Do not be caught at the end of Google investing your advertising dollars into figuring out what works for your company. Here at Alpha Efficiency, we are very wary of utilizing Google’s machine learning with our clients or our advertising dollars.
What does this mean for advertisers?
This change affects advertisers who mostly use exact match keywords in their PPC campaigns. It could potentially expand their spending and search query reports significantly. Marketers with well-structured accounts and thorough keyword lists will now have to work even harder to adjust to this new fashion. Although this change does not affect the overall click and impression volume since modified broad match and broad match keywords were already recognizing these close variants, it does present an issue when it comes to bidding.
Companies mostly use exact match with queries that have a good return on investment so that a higher bid can be applied. The phrase modified broad match, and the broad match has lower bids as they are generally less precise Google Ads match types, and with the addition of close variants to exact match, some of these less valuable queries could be recognized as exact match level bids, making them more expensive and hurting your ROI. Marketers are now putting their faith in phrase match keywords to determine the word order and whether the function words should be included, excluded, or matched since this match type was not included in the update.
You determine relevance
Before this change, Google had to find an appropriate and relevant keyword in the user’s account before showing his ad. Now, this responsibility is on you. All they have to do is find a keyword that artificial intelligence recognizes as having a similar meaning and have their algorithm label it as an exact match. If a mistake occurs during this matchmaking, you will be paying for low-quality clicks and struggling to add the appropriate negative keywords to prevent it from happening again. This means reviewing your search queries more often and running filters looking for “Exact (close variant).” You also need to pay more attention to which queries are being mapped to which keywords and add negative keywords at the ad group level to push specific queries back from high-bid exact matches to lower-bid modified broad or phrase match runs.
When it comes to online advertising, change is the only constant. We depend on big platforms like Google to display our ads, so when a change like this happens, the only thing we can do is adapt. Luckily, Google is just as much dependent on us, so every time they make a significant change like this one is, it is either for the better, or the fix is in order.
This change in type matching policy certainly looks like the beginning of the end for keyword match types. The lines between Google Ads match types are getting blurry, and it probably won’t be long before we are only requested to identify our target keyword, and Google will do the rest. We are not quite there yet, but the writing’s on the wall.