Common Advertising Mistakes Related to Placement Selection

Some advertising mistakes are made due to inaction. Others are made as a result of overthinking and unnecessary tinkering. Both apply to mistakes related to placement selection.

Today’s post continues a theme of common advertising mistakes. Read the prior posts below:

When you create an ad set, one of the customizable options is placements. Over the years, Meta has gone out of its way to make manually adjusting placements more difficult, encouraging advertisers to use Advantage+ Placements.

Advantage+ Placements

When on, all placements are eligible for distribution, though they won’t be used equally. Whether or not you make changes here is the focus of this post.

Let’s address the mistakes related to placements and provide a path for what you should do instead…

Removing Placements When Optimizing for Conversions

More often than not, the advertisers who are guilty of this infraction have been doing it for years. It’s simply too much effort to remove placements these days that there needs to be some deeply ingrained habit that motivates it.

The practice of removing placements goes back as far as the option has existed. I wrote a blog post in 2013 about how advertisers should stop removing the sidebar placement. It’s so old that it’s not active because too much of it is outdated.

But the concept isn’t. The problem was that advertisers were so focused on the low action rate of sidebar ads — and negative perceptions about how they thought it performed — that they completely ignored the low cost to reach people. The result was a solid Cost Per Action. But advertisers removed it and focused entirely on news feed placements.

It actually makes sense that these advertisers never learned. This overreaction was rarely punished with terrible results because it was so insanely cheap to reach people a decade ago. You want to force the algorithm to show ads to people in the news feeds? It was inefficient and unnecessary, but it still provided results.

Of course, it’s a different landscape now. The ad algorithm is far smarter, adjusting distribution of your ads in real-time to get you the best results. Costs are far higher, and removing placements to focus on the “highest performers” will often lead to poor performance. At the very least, the removal of placements doesn’t actually help you.

There are far more placements available today, so advertisers have more options for sources of dissatisfaction. Common targets are right hand column, Audience Network, and Marketplace.

Audience Network has a negative reputation that’s deserved. It’s notorious for driving low-quality traffic due to accidental clicks, click fraud, and bots (before those nefarious actors are detected). When your ad account gets refunded, it’s almost always due to activity on Audience Network.

So, wait. I just gave a really good reason to remove this placement. Why wouldn’t you?

Keep in mind that the algorithm is always trying to get you as many of your goal events as possible, as defined by your performance goal.

Performance Goal

The algorithm will constantly make adjustments to delivery based on its ability to get you that action. If a placement either historically doesn’t lead to that action or isn’t currently, less of your budget will be spent there — if at all.

The concerns with Audience Network are related to low-quality clicks. But, low-quality clicks don’t lead to purchases. They rarely lead to any conversions at all. That’s part of what defines a low-quality click.

If you use Advantage+ Placements and optimize for any type of lead or conversion, you’ll likely see this reflected in delivery. Run a breakdown by placements

Breakdown by Placement

In the example above, the algorithm barely even tried to show lead ads in the Audience Network placements because it wasn’t expected to drive results.

Could you remove it? Sure. But, it’s completely unnecessary. If you always remove it, you may miss an opportunity.

Using Advantage+ Placements for Top of Funnel

This section will sound like a contradiction of what I stated in the section above. But once you understand how this works, it will make sense.

Ad sets that are optimized for top-of-the-funnel actions are notoriously problematic. They consistently provide good surface-level results, but the quality is often very low when put under closer scrutiny.

Why does this happen? It has everything to do with the ad distribution algorithm, and placements are a component.

For the same reason that you shouldn’t remove placements when optimizing for conversions, you should remove then when optimizing for top-of-funnel actions. Let me explain…

Once again, your performance goal will determine how the algorithm distributes your ads. This is what defines success, and Meta will do everything it can to get you as many of that action as possible within your budget.

Instead of conversions, let’s assume that your performance goal is link clicks or landing page views.

Link Clicks or Landing Page Views Performance Goal

Remember when we talked about how Audience Network can lead to low-quality clicks? That wasn’t a concern when optimizing for conversions because low-quality clicks don’t lead to conversions. Well, that’s going to be an important concern now.

The algorithm for ad set optimization and delivery is literal. Its primary focus is getting you as many of that goal action as possible within your budget. You defined it as the action that you want, and the algorithm is going to help you get it.

Because the algorithm is literal, it doesn’t care about what happens next. You might care whether people who click your link do other things after — like read, spend time on your website, or convert. But when your performance goal is link clicks or landing page views, the algorithm doesn’t care it all. It only wants to get you clicks.

So, it will go after people who historically click on things. Maybe they’re serial clickers. Maybe they’re bots who haven’t been detected yet. Maybe they actually are potential customers mixed into the nonsense. But you can bet that the algorithm will go after the cheapest clicks first in order to get you as many as possible.

Once again, that may sound nefarious. To a point, it’s not great. But this same approach is why you benefit from it when optimizing for any type of conversion. You want as many purchases as possible within your budget.

If you’re running a traffic campaign optimized for link clicks or landing page views, removing Audience Network won’t completely solve your problem. The algorithm still doesn’t care about quality. But you can at least eliminate a very likely source of low-quality clicks that the algorithm will surely draw from.

This problem isn’t limited to clicks, unfortunately. Another potential issue is when optimizing for ThruPlays. Once again, an Audience Network placement is the source of quality concerns.

I first stumbled on this problem when optimizing for ThruPlays and I was getting results that were too good to be true. Suspicious results. I was getting more ThruPlays than actual people reached.

Wow, my ad must be amazing! People like it! But, no…

When I performed a breakdown by placement, I spotted the culprit…

Audience Network Rewarded Video

The vast majority of impressions were within Audience Network Rewarded Video. How is it possible that the ThruPlay Per Reach was 132.6%? It all makes sense when you understand how the placement works.

From Meta:

Rewarded video ads are a fullscreen experience where users opt-in to view a video ad in exchange for something of value, such as virtual currency, in-app items, exclusive content, and more.

People aren’t watching your video because they find it interesting. They’re watching it because they’re either forced or incentivized to watch it in exchange for something of value, like virtual currency. Audience Network Rewarded Video ads placements are one way that apps monetize themselves.

And since your performance goal indicates that you want ThruPlays, the algorithm doesn’t care how it gets them. It assumes you don’t care either. You might hope that those who watch the video are inspired to take another action, but that’s never a consideration for the algorithm.

The result is likely to be lots of views that go nowhere.

What Should You Do?

This is such a critical piece to understanding how ad set optimization for delivery works. When you understand it, you’ll be much better equipped to leverage it and avoid its pitfalls.

As related to placements, it’s really this simple…

1. Use Advantage+ Placements when optimizing for any type of conversion.

2. Consider removing placements when optimizing for any top-of-funnel action.

I’ve isolated a couple of problematic placements, but that doesn’t mean others aren’t a problem. If you ever get results that seem too good to be true, perform a breakdown by placement. Is a large percentage of impressions happening in certain placements?

That doesn’t mean you remove it. But ask… What is it about this placement that may lead to more of these actions? What is it about this placement that may lead to cheap and low-quality actions?

“More” isn’t necessarily “cheap and low-quality.” In the case of Audience Network, there are specific weaknesses that lead to cheap and low-quality actions. And if you optimize for those specific actions, you’re susceptible to throwing money away.

This is how you protect yourself. But long-term, Meta needs to do something to prevent this. A solution that involves providing an option to optimize for quality top-of-funnel actions would benefit everyone.

Watch Video

I recorded a video about this, too. Watch it below…

Your Turn

How do you manage placement controls when running Meta ads?

Let me know in the comments below!