So your Target ROAS (tROAS) Smart Bidding Search campaigns are doing well; that’s good to hear. But if you’re like most people, you want that campaign-y goodness to expand and grow so you can get even more robust performance out of those campaigns. You can try toggling the lever on your tROAS goal by decreasing the target percentage goal to see if that sparks some additional volume.
But what if you like where the tROAS goal is set, and it’s doing well – then what? Google is now recommending that you start to layer in some good old-fashioned broad match variations of your keywords into the campaign. The very nature of broad match tells the Google algorithm that it’s okay to cast a wider net in the search for converting customers.
As we all know, what customers want and how they are searching for what they want is a moving target. A diverse population will use different keyword combinations, have different intent signals, so the range of relevant queries that fit your product offerings is constantly in flux. It’s practically impossible to anticipate every single keyword variation that a potential customer might use to find your products.
From Google: “Using broad match and Smart Bidding together can help you reach more relevant queries that meet your performance objectives. For example, a broad match keyword like women’s hats could match to relevant queries you may not have thought of, like winter headwear for women or women’s accessories. By pairing this keyword with Smart Bidding, you can use auction-time signals to set the right bid for each of these queries. This means that you no longer need to anticipate and manage every potential search. It’s more important than ever for you to show up for your customers. By adopting broad match with Smart Bidding, it will be easier to reach more relevant queries and identify new growth opportunities.”
We were hesitant to use broad match due to less effective performance in the past. Still, Google’s Smart Bidding algorithm is continually improving and getting better and better at uncovering those predictive signals that users give the system via their browsing tendencies, so we decided to give it a try. We did a test where we added 258 new broad match variations of converting keywords. We labeled all of these keywords to keep tabs on them.
Of those keywords, 64 of them converted, so 24.81% did track a conversion. Of those 64, 40 tracked a conversion that met or beat our tROAS goal, so 15.50% of these newly added broad match keywords did perform quite well. Of course, that leaves 84.5% of the newly added broad match that didn’t deliver, but a few of them had minimal traffic, if any.
For a second broad match test in a different account, 508 broad match keywords received at least some traffic. Of those 508, 63 tracked a conversion, so that was 12.4% of the newly added broad match keywords. Forty of those conversions tracked revenue that was above our tROAS goal.
For both of the above tests, the broad match keywords were incorporated directly into existing Ad Groups. Another way to test broad match keywords would be to create a new campaign with just a few of your highest converting broad match terms, add negatives of your existing keywords, including brand terms, and then setting a high tROAS (above the non-brand search average) or low Target cost per action (tCPA) target.
Bottom line, it’s worth considering adding broad match keywords because of that potential for additional conversions. If you test, make sure to keep a tight eye on them as not all of them are going to deliver conversions at your tROAS or tCPA goal.