Don't Do What I Tell You to Do

I know this sounds a little crazy, but let me explain. I do not want you to reflexively adopt every recommendation that I make.

I’ll have people who come to me, frustrated, saying, “I know you said not to bother with detailed targeting, so I haven’t tried it.” Or, “You said to limit the number of ad sets, so I’m only using one.”

Taking this approach is not what I want at all. I want to give you enough information so you can make your own decisions.

My entire goal with my content, whether it’s my podcast, blog, or videos, is to help you understand how things work. That’s the foundation for everything else. It’s all the stuff that gets confused most easily.

I’ll tell you about the new features and what Meta says they do. I’ll share my tests, my results, and what I’ve learned. I’ll break down Meta’s own best practices to help clarify what they actually want you to do. And when there’s uncertainty, I make educated guesses by connecting the dots of what we know to make sense of what we don’t.

I do believe there’s a general approach and philosophy that are ideal. I promote a simplified approach to advertising for many reasons. Complexity is often unnecessary and can even make results worse. And the way things work now, complexity often doesn’t make much sense anyway. A simplified approach also helps isolate what’s working and what isn’t.

The bottom line with my recommendations is that there are always exceptions. Very few universal rules exist, if any. Yes, I think you should limit the number of ad sets, and you can often get away with using only one in a campaign. But there are situations where multiple ad sets may be necessary or even beneficial.

Yes, I have serious doubts that targeting inputs like detailed targeting and lookalikes do much when optimizing for conversions. They’re treated as suggestions, and my tests have shown they’re often ignored. The algorithm is going to reach the people it wants to reach. But if you’ve seen good results with those inputs, great. Maybe it’s correlation and not causation. Or maybe you’re experiencing an exception that I hadn’t considered.

If I reach my goal, you’ll gain a much deeper understanding of how things work. And when that happens, you won’t need my guidelines or recommendations. They may be a starting point, but you’ll be able to quickly sort out whether you’re dealing with exceptions.

Once you understand how things work and run your own tests, you’ll start developing your own guidelines. They may not even be the same as mine. And you know what? That’s great. Whether you’re listening to me or another experienced advertiser, use our advice as a starting point.

What worries me is when people hear a new strategy from someone else and copy it step by step, word for word, as if it were gospel. That’s when you get into trouble. You assume every part of that strategy is solid, but it may not be. There could be guesses or inaccuracies in the approach. And if you don’t question it, you’re only going to make mistakes.

Here’s the bottom of the glass. Your goal as an advertiser should be to fully understand how everything works. That goal may never be realized. It’s my entire focus, and there are still things about Meta advertising that are complete mysteries to me.

But when you have deep knowledge of how most of these things work, it’s extremely valuable. When you run into something that you don’t understand, you can use what you do understand to piece it together. It also makes it much easier to spot the flawed conspiracy theories I see floating around all the time.

So do not take what I or anyone else says as gospel. Don’t copy it word for word. If I’ve earned your trust, use it as a starting point. But ultimately, you need to determine the best approach for your situation. You need to run your own tests. You need to define your own best practices.

When you build processes based on knowledge and experience, rather than repeating what someone else told you, you’ll be better equipped to explain your strategy and make adjustments when things don’t go as expected.

And that is true knowledge.