Understand How Meta Advertising Actually Works

So one of my biggest frustrations is watching conspiracy theories about Meta advertising spread. The reason they spread is that not enough people have enough knowledge of how things work to put a stop to them. They could say this is not even possible because of how things work.

While those conspiracy theories will always thrive, you would be giving yourself a gift by knowing when they are based in nonsense. There is no reason to waste your time or spin your wheels on something that does not lead to results because you know how things work.

I realize it is not easy to get to that point of understanding. My challenge to you is to invest time into how things work. You are making an attempt right now by listening to this podcast, so you are on the right track, but pay special attention to what Meta says about how things work.

I am not talking about the PR spin that Meta puts out about how effective certain tools are. That stuff is not helpful. Focus on the Help Center articles that articulate the mechanics of what happens. This is incredibly useful information, and there is a reason I focus so much of my education there.

What Meta says about how things work defines an important foundation of fact. There is no reason for Meta to lie or mislead about mechanics. Everything else outside of that foundation of fact is theory based in experience. My experience will be different from yours, and how we interpret that experience, whether accurately or not, is also important.

Because of that, everything outside of the foundation of how things work is debatable and murky and confusing. So if you do not have a strong foundation in fact related to how things work, guess what. It is even murkier and you are prone to believe just about anything.

Let me give you some examples.

Targeting. It does not matter what you believe about the effectiveness of algorithmic targeting. There are certain things that are part of the foundation of fact. Age, gender, custom audiences, lookalike audiences, and detailed targeting are used only as suggestions by default.

You can always restrict by age and gender if you want. That is a fact. Whether that is necessary is theory.

You can also restrict by custom audiences if you want, but Meta will prioritize remarketing audiences if you do not target them explicitly. Meta says this, and you can prove it with audience segments in sales campaigns. Whether remarketing is useful is theory.

Detailed targeting and lookalikes are only used as suggestions in some cases, and in others you can restrict to those audiences if you want. Whether it is possible is determined by your performance goal. That is provable fact. Whether using these as suggestions or restricting them when you can is theory.

Attribution. Meta counts a conversion by default when someone clicks your ad and converts within seven days or views your ad without clicking and converts within one day. This is a fact.

The click does not need to be an outbound click on a link. That is a provable fact that I discovered within the past year because Meta never spelled it out. Whether you should use seven day click or one day click or one day view or value is theory.

Meta is not making up conversions that do not exist. If you get conversions that did not happen, there is a reason. Check all conversions versus first conversions. One person can perform multiple conversions, and that can lead to mismatched information.

We also know the results column is for the event you want, but not necessarily the product you are selling. That can lead to mismatched data too.

There are numerous reasons an event might fire when it should not. You may have set up the event wrong. Maybe it sends on the wrong step. Maybe you did not deduplicate your pixel and API events properly. Or maybe someone on your team is testing or building a confirmation page and the event fires after they are shown an ad.

When you have a foundation in facts, you know there are many explanations for inflated or deflated results. When you know how things work, you should have an internal checklist of things to check.

When you do not have this foundation, the easy reaction is to blame Meta for inflating your numbers or making up conversions. Your results can absolutely be misleading if you do not know how to uncover the layers. But that is why this base understanding is so important.

So here is the bottom of the glass. Whether we are talking about targeting or attribution or ad distribution or Andromeda or something else, it is easy to get lost and confused by conflicting information when you do not have a base understanding of how these things work.

This is your responsibility as an advertiser. Read Meta’s documentation so you know how things work. And yes, you can listen to this podcast or consume my content too. When you have that foundation, any theory that comes from experience suddenly has a whole lot more meaning.