Artificial intelligence. Machine learning.
These are buzzwords being used ad-nauseum (even more than the word “buzzword” itself) by Google & Facebook to explain how their systems know better than us humans when it comes to ad targeting. They say that rather than set super-specific targeting for our ad campaigns, we should just leave the targeting to them, and they’ll figure out who to show our ads to.
And, in theory, I’m cool with this. If their AI can give us the best possible ROI, then that’s great. Go for it.
But there are some times when data clearly shows that Google (I’m pointing my finger directly at Google in this case, and not Facebook) is being too smart for its own good, and is taking intelligently targeted ad campaigns, targeted by real human brains, and bringing them to inefficient places.
Here are two examples which show what I mean.
But First, A Disclaimer
I love Google Ads. They’re awesome, and bring great results. The goal of this blog is not to say that Google Ads don’t work. They usually work great. But sometimes there are issues that you need to be aware of to avoid wasted ad spend, and this is a classic case.
So, that being said, here are two examples of Google going rogue when it comes to keyword targeting.
We’re running ads for a real estate company which rents out commercial office space. “Office” is the key term here. We do not want to target any users unless they are specifically looking for office space.
So, naturally, we wanted to target users who searched for the term “commercial office space”, and we did so using exact match, phrase match, and broad match modifier. You would expect that our ad would only appear when someone searched for a term which included the word “office”. This could be something like “rent commercial office space”, “commercial office space near me”, or “how to find commercial office space” – really, any term which fits this theme.
We were surprised, then, when we started receiving leads from people who were looking to rent out gas stations and store-fronts. Why would someone looking for these types of properties search for an “office” keyword?
Well, as it turns out, they weren’t. Google overrode our keyword targeting. It was ignoring our requirement to have “office” in the search term, and was showing our keyword whenever someone searched for terms like “commercial property” and “”commercial real estate”…and these terms do not mean that someone is looking for office space!
Users could be searching for “commercial real estate” because they were looking to rent out space for a pizza shop, yoga studio, clothing store, gas station, etc, etc.
In other words, Google was taking our most targeted keyword, disregarding its most important word, and showing ads to a completely untargeted audience. It made our #1, go-to keyword completely irrelevant!
We’re running ads for a manufacturer of skin care products, and we want to direct ads to a product recommendation quiz which is specifically for products for your body (and not your face).
So, naturally, we want to target users who are searching for terms like “body skin care routine”, “even body skin tone”, etc. The common denominator is “body”! We only want people who are searching for terms with “body”. Got it, Google? “Body”. We need “body” to be in that search term. If that word’s not there, we’re not interested. Ok?
Well, nope. That’s not how it turned out. Once again, Google showed our ads on irrelevant search terms.
Check out the screenshot below. The highlighted terms do not include “body”, even though this word is clearly required by our broad-match-modifier match type.
Google Has Decided To Help This Situation By Making It Worse
So, you might be saying to yourself, “At least I’ll have the search term report, which gives me transparency into when my ads are showing”.
Well, no. You do not have transparency. Or at least, not enough.
In September, Google announced that they will be cutting back on their search term reporting.
Case in point – check out this screenshot from the commercial office space client account. Google only gave us search term data on 377 out of our 955 clicks. This means that 60% of our clicks and 57% of our ad spend went towards search terms which we can’t see. That’s crazy!
(In Google’s parlance, “Other search terms” means “Total of search terms not shown in this report”)
So, What Can You Do? Follow These 4 Steps
There are a few steps which I recommend to Google Ads agencies to cope with this new reality.
#1 – Realize that Google Ads is still great, and offers you targeting unmatched in the history of advertising. I’ve run Google Ads to successfully efficiently reach people searching for family lawyers, cosmetic surgeons, tire inflators, accountants, weight-loss solutions – you name it.
So Google Ads is still great. You just need to make some adjustments to your thinking.
#2 – Remove the idea from your head that you have control over your search terms. And don’t take it for granted that you’re going to get your intended search terms by simply implementing non-broad match types. No matter how well you structure your keyword targeting, Google can still go rogue on you.
#3 – Check your search terms report, even more often than before. Yes, it is true that Google is giving less data than ever before in this report. But still, check the data that it does give you, and make sure that the terms that you can see to make sure they are in-line with the audience you’re trying to reach.
#4 – Make sure that your conversion tracking is rock-solid, and that you are measuring (and optimizing for) conversions. Because after all, in many cases Google may really know best. And If Google’s sometimes-frustrating-machine-learning-based targeting gets you the ROI that you’re looking for, then just roll with it!
To do that, you’ll need to make sure that:
Follow these four steps, you’ll still be able to generate great ROI from your Google Ad campaigns!
Barry Buckman is Director Of Client Strategies at SunHouse Marketing. He helps clients develop laser-focused strategies which are, quite simply, aimed at increasing their bottom line to the highest degree possible. His areas of expertise include Conversion Rate Optimization, Google Ads, ads across any social network, and SEO