4 Mistakes to Avoid When Creating Google Ads Audiences

by | Jun 12, 2024

Subscribe to the Playbook Blog

Google advertising is a great tool to help advertisers connect with people looking for their product or service. That said, if you don’t use its full functionality well, you run the risk of missing the people who are actually qualified to purchase your product or service. 

It is possible to find the right people, of course; Google has spent a lot of time and money giving advertisers ways to refine their audiences. Still, advertisers tend to commit unforced errors that we find ourselves correcting over and over in new client accounts.

At Playbook, we’re all about raising the advertising bar. So, here’s a list of common errors we recommend you guard against in refining your Google Ads audiences.

1. Using too many audiences

I’ve seen brands upload up to 50 (!) audience lists for Google to use when finding new users. That approach can split the available data you have into too many pieces, which leads the algorithm to struggle to latch onto the right signals. Every click has to be assigned to either one audience or another, so too much of a good thing can become a bad thing.

Instead, I recommend curating audiences using one of a few techniques:

  • Paring and combining existing, related lists into a more granular subset
  • Using the (very under-utilized) Combined Segments feature to both slim down and create overlap in a few existing audiences
  • Setting very clear benchmarks by identifying how many conversions you need in your industry for the algorithm to thrive—then ruthlessly deleting or combining lists that don’t meet those criteria

One note here: sometimes it’s a very good idea to combine audiences to achieve the data density the algorithm needs to self-optimize. For instance, if you’re not closing your benchmark number of clients a month, you might need to create a lookalike list of closed-won and SQL contacts to get a big enough list while keeping the quality high.

2. Letting Google take control

Google, like many other platforms, is leaning hard into AI-powered bidding and targeting—most notably in Performance Max campaigns. But at this point, we’re still seeing better results when we opt out to create standalone campaigns for search, video, shopping, demand gen, etc. 

Performance Max may not be ready for mass adoption, even if it’s a Google focus right now. It has potential, but from our testing, it’s not quite there yet—at the moment, there’s too much reach and scope without enough visibility and control. Yes, it may find pockets of new users your more manual targeting won’t, but I haven’t found that volume to be worth the tradeoff of the low lead and conversion quality you get from Performance Max campaigns.

3. Ignoring GA4 data

GA4 is much-maligned for a few valid reasons, but it’s still a very under-tapped source of data that you can use in your Google campaigns.

Consider that GA4 pulls data from all of your traffic sources—Facebook, direct, organic search, LinkedIn, TikTok, etc.—not just the people who find your site through Google ads. If someone visits a pricing page, a services page, a product page, or some other high-intent property on your site, they’re giving you a signal. Setting up reports on GA4 (I really like recency/frequency reports and funnel-stage reports) and using those to build audiences for Google ads has produced some great results for our clients. If you’re not using all the data available to you in GA4, that’s a big missed opportunity.

4. Going too broad with YouTube campaigns

YouTube is a straight-up discovery channel, and Google wants you to go broad—you can’t use Combined Segments, for example. But “discovery” doesn’t mean “show your ads to everyone”—it means showing your ads to people qualified to take further action down the road.

Instead of letting YouTube run wild, layer first-party data and lookalike audiences, which is where we’ve seen the most success. And while this isn’t an audience-related note, keep in mind that YouTube tends to allow a much higher frequency than what we’ve seen perform well, so make sure you not only train the algo on who to find but set guardrails on how often to find them.

I’ve seen lots of growth marketers throw up their hands because they’re struggling to find the right audiences with their Google campaigns. Remember to cut down on the mistakes I just described, and you’ll see your audience quality and performance swing in a much better direction.

Subscribe to the Playbook Blog