You May Be Surprised Who Converts

You may be surprised who converts.

There’s a classic mistake advertisers make that is completely understandable.

It starts during the planning process of a campaign. We’re thinking about who our product is for, who our customer is, and we want to be sure Meta knows that, too.

We assume the delivery algorithm needs help, so we start refining.

We sell men’s clothes, so there’s no need to target women. So we restrict by gender.

Our target demographic is men between 35 and 54, so we restrict by age.

We want to make sure that Meta doesn’t waste any money on people outside of those groups to keep our costs down.

And then what happens?

Your costs are higher than expected. This is partially because you’ve limited the audience pool, so your CPM costs go up.

But it’s also because you aren’t even giving Meta a chance to find people outside of those demographic groups.

You assume they won’t buy, but you don’t know that.

The reality is that women may be just as likely to buy clothes for their male partner as a man is to buy for themselves. But you’re not giving them a chance to buy it.

And even if they don’t buy it, it may be a woman who sees your ad and generates initial interest before she passes the information on.

The same goes for age.

You may see that your target demographic is 35 to 54, but why did you decide on those specific ages?

Are 33 and 34 simply too young to appreciate your clothes? Are 55 and 56 too old?

You aren’t even giving them the opportunity to buy.

And this also ignores the problem with gift giving. What if someone outside of your target demographic is looking for a shirt for a man in that target age group?

They won’t see your ad.

You may be surprised by who converts.

And when you force your assumptions onto ad distribution, you are bound to hurt results.

What you don’t understand is that Meta wants you to get good results, too. If you’re optimizing for purchases, Meta’s goal is to get you as many purchases or as much value as possible, depending on your goal.

Meta will deliver ads to both men and women and people of all age groups, but it won’t be done evenly.

The goal is to get you more purchases, so it’s unlikely that Meta will waste a high percentage of your budget on groups who simply don’t convert.

So I encourage you to start with the broadest demographic groups for age and gender, unless you’re selling age-restricted goods.

If you’re selling something that people under a certain age can’t legally buy, you need to prevent them from seeing your ads.

Of course, this recommendation is mostly for when you’re optimizing for a purchase.

The reason is all about how optimization for delivery works. If your performance goal is to maximize the number or value of conversions where the conversion event is a purchase, that will be Meta’s focus.

It’s unlikely you’ll waste a bunch of money on groups of people who don’t convert. This is especially the case if you’re spending enough to generate data that Meta can learn from.

The problem is when optimizing for top-of-funnel actions, especially clicks, post engagement, and video views.

If you’re selling men’s clothing and your goal is one of these actions and not a purchase, Meta doesn’t care if these people don’t buy.

You said you only want a click or video view, and your ads will be delivered to the people to make sure that happens.

Even if you’re optimizing for an action other than a purchase, first confirm that there’s a problem with how your ads are delivered.

And if there’s a problem to be solved, use value rules instead to lower bids on the problem demographics.

This could apply to purchases if you’ve determined that certain demographics lead to more lifetime value than others. That’s information that you have but Meta doesn’t.

But again, make sure you have the budget and conversion volume that make such a restriction worthwhile.

Finally, avoid the trap of attempting to refine your audience based on results.

You use the breakdown by age and see that the worst Cost Per Conversion comes from people under 35.

Or the best ROAS is from men aged 45 to 54.

This is especially dangerous when you don’t have nearly enough data to make the good and bad performance predictable.

But you start restricting your audience based on these results, thinking it will optimize performance.

And it almost never does.

And if it does for a short time, it won’t last.

So here’s the bottom of the glass.

While it may be counterintuitive, resist your instincts.

When optimizing for a purchase, you and Meta are on the same page regarding what you want.

Let Meta find your customers, and let your ads do the heavy lifting.

When you make assumptions about your target demographic, you never give Meta a chance to find anyone who might be outside of it.

And even if you’re right that a certain age group is your strongest concentration, you are preventing customers from other groups that might surprise you.

Resisting this urge to restrict by demographic group is one of the key points in my Master Brief.

Read it at jonloomer.com/master.