I stole a bunch of memes and ran them as Facebook ads

The first time I ran a Facebook ad was in 2014.

I was working on an innovation lab project for Toyota. Instead of marketing it to the millions of existing Toyota customers (too easy!), I had to find a brand new audience to prove this could make money.

Also, I couldn’t tell anyone Toyota was behind it.

So, we made a ‘coming soon’ landing page with an email sign up and ran some ads to it.

It. did. not. work.

In fact, it was so excruciating to deal with that I found it easier to recruit potential customers by lurking at gas stations and just cold pitching them myself. True story.

Fast forward to this year, it’s gotten A TON EASIER to run Facebook ads.

I wanted to see just how easy so I ran a two week experiment and today, I’m showing you my results.

Should you bother using Facebook ads to grow your newsletter? This issue will help you decide!

P.S. 👋 Welcome to all my new subscribers from Facebook and Instagram. I could have segmented you out of this email, but I thought you’d get a kick out of it. Can’t wait to find out what you think!

What you need to know about Facebook ads

Facebook ads are constantly changing because it’s Facebook’s pride and joy (and main source of revenue.) It’s not even worth writing you a “how to” guide because it will be irrelevant in 5 seconds.

If you have never run a Facebook ad before, buy a course. There are 10 million little settings inside the ads manager and it’s easier to have someone show you live.

I highly suggest Matt McGarry’s Write Grow Sell which covers a lot more than ads, but is also very thorough about ads, because McGarry is an ads guy. He walks you through his recommended set up and shares a ton of templates for your ad creative. A lot of my success with this week’s experiment I owe to him.

You can also find downloadable one hour “set up your Facebook ads” classes online. Ask around for recommendations.

If you have run Facebook ads before but they feel like a black hole of despair, I get it and I’m here to tell you that they have gotten a lot easier since my Toyota days.

Audience targeting has been by far the biggest improvement in the last couple years. You barely have to touch those settings now because Facebook’s machine learning does all the heavy lifting based on your ad copy/image copy.

Lastly, I’m sorry that I still call them Facebook ads. You can call them Meta ads if it helps you sleep at night.

Facebook ads run on all properties Facebook owns, including Instagram, Marketplace, Messenger and the Audience Network (other websites running display ads).

My Facebook ads experiment

🎯 Goal: Acquire new subscribers for less than $3 and keep paid acquisition at less than 30% of my overall list growth.

🤔 Prediction: I’ll acquire 33 new subscribers on an ad budget of $100.

🧪 Methods:

• Set up 2 ad sets: 1 that goes to my website, 1 that uses Instant Forms

• Keep audience simple: United States, ages 25-64, all genders

• Load in 10 images (same images on both ad sets)

• Load in the max amount of ad copy variations

• Set daily spent at $10 per day on each ad set

• Set total budget at $100

• Turn off low performers every couple days and add in new creative to see if I can beat the best rate

📊 Metrics:

• Return on ad spend

• Cost per new subscriber

• Number of new subscribers

• Open rate for new subscribers

• Unsubscribe rate for new subscribers

A few of my ad creatives 👇️ 

Oh yeah, about the stealing the memes thing. One thing I learned in McGarry’s class is how well memes work in ads. I didn’t really want to have to create 20+ memes from scratch for this test, so I scoured the internet for memes about journalism.

I also tried running some of my branded assets and some more utilitarian ads, but as I’m about to show you, those didn’t work for me.

My Facebook ad results

Overall results: ❌ Loser

I ended up spending $180 over the two weeks I ran these, because I knew I was going to write this issue and I wondered if I could dial in the creative better.

For $180, I got 54 new subscribers which is a cost per new subscriber of $3.30 but 2 have already left so that brings me up to $3.50. Both of which are above my target of $3.00 for a new subscriber.

• Some individual ads performed below $3.00. The lowest was $2.23.

• Facebook’s Instant Forms and my website performed about the same.

• Open rate for this cohort was 39% which is way worse than my other ads and my organic open rate.

• Unsubscribe rate was normal.

These were the top three performing ads based on cost per new subscriber:

This was the ad that got me the most new subscribers though. Cat memes are always a winner.

A few notes about my creative strategy here:

• I really did not overthink this. I stole the memes and plugged a block of text from my existing website into Claude to generate the ad copy.

• I did not bother to properly size ads for the hundred aspect ratios Facebook wants you to optimize for. I aimed for square and told Facebook to show it with negative space if it didn’t fit.

• I only did static ads. McGarry recommends trying some video ads. I just didn’t feel like it for this little test.

• Set up took me 1 hour and then I spent a total of 3 hours on separate days turning off low performers and adding new memes in.

• Why not market specific issues of the newsletter? McGarry told me not to! I asked him this question during the class because I’ve talked to local publishers who’ve had trouble with ads. He recommended advertising the entire newsletter, not just one issue, for better retention of new subscribers.

Did this ad pay me back?

Yes.

My revenue per new subscriber is $6.70. That’s because I make money through both sponsorships and subscriptions.

So even at $3.50 for new subscriber, I made close to double that back.

• Cost per new subscriber = $3.50

• Revenue per new subscriber = $6.70

• Return on ad spend = 191%

Compare this to my beehiiv boosts results and my creator ad results.

What’s next for my Facebooks ads experiment?

I shut it off.

It’s more annoying to manage than my other two ad options and I’m not actually spending that much on paid growth right now, so Facebook’s bottom of my list.

However, I did want to know how other newsletter writers were getting these below $1 per new subscriber ad results.

So I took to Reddit and Twitter to ask and boy, did the growth bros come running!

Some actually helpful answers I got:

• Facebook ads may be cheaper for local news (over national, international or interest-based news.) Brian Skinner who writes Marin Buzz, a local newsletter, reported that he’s getting new subscribers at $0.23. He said he thinks the hyper-local target makes a difference in the cost.

• Instant forms worked great for Tance Hughes who writes The Ruralpreneur. Tance’s cost per new subscriber is $0.49 and he’s about to launch his own Facebook ads course about his set up so get on his waitlist here.

• $2-4 is more typical, according to Louis Nicholls who runs a mediocre newsletter plug-in I don’t recommend called Sparkloop.

• $1-3 should be my aim according to Matt McGarry who I trust more than Louis.

But, I can get new subscribers from beehiiv at $2.75 where most of that money goes to another writer or I can sponsor other writers directly so unless I can get Facebook down reliably in the $1-2 range, it’s not a good spend for me.

How to try Facebook ads for yourself

It’s actually NOT AS HARD AS YOU THINK to run Facebook ads, but it is worth getting a leg up from people who already run them and luckily, those people are everywhere.

Don’t bother looking for an ads manager or an agency to help you unless your ad budget is at least $3k/month. The rest of us have to learn it ourselves 🙃 

1. Get a course! Take McGarry’s course or get on the list for Tance Hughes’ course about to launch soon. You’ll need this to get set up quickly. It’s worth it, trust me.

2. Set up your ad account, your Facebook page and your tracking pixel.

3. Create your first campaign. Inside your campaign, you’ll have Ad Sets (groups of ads) and Ads (individual ads).

4. In your Ad Set, set your conversion goal. I recommend trying both website leads and Instant Forms. That would be two different Ad Sets because an Ad Set can only have one goal. You’re aiming for new subscribers (leads), not traffic.

5. Make sure your conversion goal is hooked up to an actual thing that happens on your website (unless you’re using Instant Forms)

6. Set your budget and schedule. I ran $10 a day on each ad set.

7. Adjust your audience target. I stopped sweating this altogether and mainly made sure the country was set correctly because Facebook did seem to be able to figure out who to show this to. If you’re a local publisher, I would filter to your city or state.

8. Load in the creatives! McGarry recommends trying 20+ different ones which is why I ran so many. The more you can add here, the faster you’ll get results because Facebook tests them against each other for you.

9. Hit publish and check back tomorrow!

Tips for running Facebook ads

• Take a course. Did I already say that?

• Check out the Facebook ads library. You can see what ads other publishers are running. Like here’s CNN, the NYTimes and the Philly Inquirer. Ads that have run the longest are ads that are working for them.

• Use memes.

• Use quick keywords. Facebook is reading these as is your actual reader. Mine were “independent” “journalism” “newsletter” “business.”

• Monitor your ads on first 3 days of running, shut off ad creative that doesn’t work and add in new creative options for Facebook to test.

Would love to hear your experience if you try Facebook ads or if you’re already running them. Send me a note by replying to this email.

This wraps up our paid ad series. Find all three issues here including my beehiiv boosts and newsletter sponsorship experiments.

Next month, we’re getting into what to paywall. Stay tuned!

Get The Paid Sub Playbook