Advertisers: CPMs and CPCs Don't Matter

I've spent almost $1 billion advertising on FB since '08 and this is something I see advertisers keep getting wrong all the time:

🛑 Stop focusing on FB CPMs (and CPCs)! They don't matter that much.

❌ Turning off ads due to high CPCs or CPMs can be turning off top performers.

‍Before we dive in: this is for *conversion-focused* advertisers

If you're focused on reach/awareness, then yes, of course, CPMs matter. DUH!

If you're *only* focused on reach/awareness, you should consider optimizing to other actions.

Conventional ad wisdom says more impressions = more clicks = more conversions.

❌ It doesn't work that way on FB.

‍More impressions don't always drive more clicks. More clicks don't always drive more conversions. 🤯

When observing/comparing placement performance, there's a loose correlation between CPM/CPC and CPA.

There are many variables that define FB ad delivery: age, geo, placement (mobile vs desktop, iPhone vs Android, feed vs story, FB vs IG), etc.

Let's focus on *placements*

‍Each placement has different inventory, competition, and relevance to your ads/brand/product.

Users are/act in different ways on each placement too.

If you're not manually targeting each of these placements individually, you're letting Facebook optimize them automatically.

🥅 Your goal as an advertiser should be to get the lowest cost conversion (or highest ROAS/MER), not cheaper impressions or clicks.

Facebook does that for you automatically by optimizing across placements, regardless of CPM or CPC.

You can't know or control what inventory is available or how competitive each placement is in real-time.

FB does.

FB determines where, when, and who they show your ads to based on relevance, ad quality, inventory, competition, etc.

FB shifts placement delivery according to how efficiently your ads succeed at getting conversions on each placement. Your CPM, CPC will shift from that.

Some expensive CPC placements may have better CPA while cheaper CPC placements may have worse CPA.

Right column CPMs are super low, do you want to advertise more there? No, you don't.

FB could deliver your entire budget there if it wanted. FB knows that while it can get you cheap CPMs from the right column, it won't get a lot of click conversions.

When you say "my CPMs are high this month", you're trying to say that your costs have gone up outside of your control due to competition.

I get it.

You don't know why, so you make up stories to explain why CPMs changed.

Instead, just focus on your CPA.

We make guesses:

Maybe users are spending more time on FB properties, increasing ad inventory. Maybe demand for your product changed due to seasonality. Maybe FB changes their algorithm. Maybe a mega advertiser activated a giant campaign, creating more competition 🤷

🔬If you're looking at your ads through a microscope, you might notice the amount spent on these placements shift, but you'll never have full transparency into *why* it shifted.

You also won't have transparency into if/how/when FB changes the available inventory on placements.

Example: On IG Stories, FB now shows two back-to-back ads between stories from profiles you follow. It used to be only one ad impression. That changed a few months ago.

FB increased the deliverable inventory there to combat increased competition.

The costs of advertising are ever-rising over time 📈

This is not new and it's never going to get easier.

Blaming CPMs doesn't help.

You still need to hit your business goals.

Focus on sharpening your ad spear 🔱

I can hear some people saying

"But Barry, how do you know any of this? Conversion data isn't in breakdowns since early-2021."

That's true.

I've seen how the performance varied in the past.

I also understand that FB allocates more $ where there is scalable efficiency.

Even if you could look at the conversion breakdown data, the lowest CPA or highest ROAS are misleading.

They don't tell the whole story because you don't have transparency into available inventory.

Google: "Facebook Breakdown Effect" ^

Pay attention to what gets the most $$

So, what should you do?

👉 Focus on what matters: conversions, sales, revenue.

Not impressions or clicks.

You can use other metrics as a guide/comparison, but if you don't understand the underlying meaning/causes, you'll make bad/wrong decisions.

👉 Spot check your breakdowns to understand where and how much FB is delivering to different placements.

This doesn't need to be daily, but once in a while can help you better understand why/how your individual tests or your broader efforts fail/succeed.

👉 Keep using a variety of ads!

If you only use images or only use videos, you're limiting your chances for success.

Videos and images deliver differently to different users on different placements.

There isn't a silver bullet "winning" ad type.

❌ Don't separate your ad sets to try to control/manage these differences.

FB's optimization does this faster and more effectively than you ever could.

You can't outsmart FB. It's not 2015 anymore.

You can't automate this better than FB either.

🛑 Stop pausing ads due to "high" CPMs or CPCs.

If you have an ad that is resonating and scaling more on IG feed than your other ads, it's going to have a higher CPC. If you turn it off due to the high CPC, you might be killing your best performer.

🛑 Stop thinking and talking about FB CPMs going up or down! They don't matter!

I understand that by talking about "CPM" you're trying to isolate it as an external factor but that's not valuable.

Also remember that CPM reflects your ad performance, not a raw auction CPM

🚮 Remove CPM from your default Ads Manager column view.

If someone asks you about your CPMs, tell them "they don't matter because there are too many variables" and send them this blog!

If you focus on conversions, don't compare CPMs on FB to Snap or TikTok.

🧠 Remember that aside from direct-attributable click conversions, the power of an impression varies on each placement too.

If a user views an IG Story ad without clicking, that might leave a larger lasting impact than one in FB mobile feed.

I'm not advocating for view-based conversion attribution, but reminding y'all that we don't advertise in a vacuum and our conversion ads have a bigger/longer impact than we measure.

🏁You've reached the end!