How I Actually Approach Targeting in 2026

Here’s how to approach targeting in 2026.

Last week, I shared how Meta should eliminate audience suggestions from the advertiser’s toolbox. It’s a pointless option that’s unlikely to do anything, and it’s certainly not worth the confusion it causes.

So you may wonder how I approach targeting now and what I recommend. That’s going to be the focus today.

First, it should go without saying that I don’t bother with audience suggestions. I’ve seen no indication that they do anything. They can’t hurt, but there’s also no reason to see them as an important part of your strategy.

You could make the argument that MAYBE they’d be useful for new businesses or new accounts where Meta doesn’t have data, but that’s purely a hypothetical.

I do not restrict by age or gender.

Now, I understand that there are times when there can be problems related to these demographics settings. But that’s almost never when optimizing for a purchase. Because the algorithm is literal, Meta will show your ads to people most likely to perform the action you want.

And when you optimize for something other than a purchase, you can bet that Meta will find cheap actions beyond your typical customer demographics.

If this sounds like I’m backtracking on my stand related to not restricting by age or gender, I’m not. If there’s a problem to be solved — and that problem would be because I have information that Meta doesn’t related to customer lifetime value or lead quality and it’s demographic specific, I don’t solve it by restricting by age or gender. That would be the old way.

Now I apply a value rule to bid less on the problematic demographic group.

Next, I do not use detailed targeting or lookalike audiences.

Both of these targeting options go back more than a decade when there was value in using them. But Meta has so much data now — both historic and realtime data — that often makes these audiences obsolete.

And even if you’re convinced that detailed targeting and lookalike audiences could have value, there’s a problem. They are ONLY used as suggestions now for the most commonly used performance goals.

You can’t restrict by detailed targeting or lookalike audiences when optimizing for conversions or leads or really anything else I’d care about.

Next, I almost never restrict by custom audiences. In other words, remarketing.

This is something I used to do almost exclusively, but times have changed. I know that Meta now prioritizes my remarketing audiences without providing any inputs at all. I can prove and monitor this by defining my audience segments and using breakdowns.

Meta already spends 20 to 25% of my budget on remarketing, so why do I need to isolate these people?

There are rare exceptions when I’ll isolate a remarketing audience, but it’s exceedingly rare.

I don’t get cute with location targeting. I target the countries I want to reach. Unless I can get significant volume from splitting them up, I combine similarly priced countries into the same ad set.

I don’t perform breakdowns by region and focus only on certain states. I focus primarily on the US, Canada, UK, and Australia. Occasionally, I’ll throw in a few more countries.

And finally, I’ll exclude custom audiences when I don’t want to reach the people who already performed the action I’m promoting.

This isn’t true prospecting where some advertisers will exclude all custom audiences. I’m just trying to limit waste on people who can’t buy twice.

But I also know custom audience exclusions aren’t perfect, and I will reach some people I’m trying to exclude. And that’s just not something I lose sleep over.

So, let’s summarize.

I do not use audience suggestions. I do not restrict by age or gender. If there’s a demographic problem to be solved, I’ll use value rules. I do not use detailed targeting or lookalikes, mainly because they are almost always suggestions now.

I rarely restrict by custom audiences because remarketing happens naturally now, and I can prove it with the breakdown by audience segments.

I target groups of countries in a single ad set. And I exclude custom audiences when I want to prevent people from seeing my ads who already bought the thing I’m promoting.

It’s about as hands-off as you can get.

So here’s the bottom of the glass.

Your targeting strategy is not the key to your advertising success. In fact, your obsession with targeting control may be hurting results.

In most cases, the only settings I touch are locations and exclusions, which are audience controls. And when I run ads for leads, I’ll use a value rule for age adjustment.

Other than that, I don’t touch a thing, and it’s great not having to worry about it.

For more on my recommended approach to targeting, I’ve published a free mini-course. It’s called Modern Targeting, and it consists of eight lessons that walk you through what changed, how things work, and how you should approach targeting now.

Subscribe for free at jonloomer.com/targeting.