One of my favorite things to do is run Facebook (or Meta, whatever) ads experiments. I’m in the middle of one now that I want to tell you about.
Maybe you are seeing the ads now? Maybe. Let me explain what you might see and how you might qualify.
[SIDE NOTE: Here’s an experiment I ran way back in the day that was a lot of fun.]
The Concept
Like all of my experiments, the point is never to generate amazing results, but sometimes that’s a byproduct. The primary goal is to try something new and see what happens.
The first motivator is to use ads in a way that most people aren’t using them. Just like my initial experiment in 2015 (linked earlier), Facebook ads don’t have to suck. Sometimes, they can surface exclusive content. Sometimes, people will actually look forward to seeing them. If you’re lucky, they’ll even complain if they don’t see them (this happened!).
A whole lot has changed since that original experiment in 2015. First, I could never replicate it because the costs were SO MUCH lower. The final ad in that experiment generated 634 webinar registrants on an ad spend of $13.64 while reaching only 740 people. You do the math. It was crazy nuts.
Needless to say, I’m not expecting that now. Instead, I’m looking to reward loyal readers of my website while also incentivizing additional engagement.
Unlike the first experiment, I’m not going to create an opt-in based on a simple click. Instead, I’m going to straight-up show exclusive content to those who engaged with me in the past. I’m going to create a series of lessons. And each lesson will be served to an audience that is more engaged than the last.
In addition to rewarding and incentivizing engagement, I want to test how small an audience can be to reach it. This, too, has clearly changed since 2015.
Top-Line Details
So, how this campaign is structured is pretty complex. And honestly, those details make up the meat of the lessons (so, why would I disclose them here?). But, let’s provide some top-line details…
I’m going to create a number of lessons. Each lesson explains how I am running this experiment. The experiment is constantly evolving based on what happens, so I’m creating the lessons as I go.
The first lesson was sent to those who viewed at least two pages of my website during the past 30 days. After creating that lesson, I got nervous that Facebook would stop serving lessons too quickly, so I expanded the net to 180 days going forward.
As of this moment, I’ve created the following lessons (target audience in parentheses):
- Lesson 1: Experiment Targeting (Viewed 2+ pages of my website last 30 days)
- Lesson 2: Gamification (3+ last 180 days)
- Lesson 3: Exclusions (4+ 180 days)
- Lesson 4: Country Targeting (5+ 180 days)
- Lesson 5: How I Handle iOS Devices (6+ 180 days)
Topics that are on my list to cover:
- How I’m Optimizing the Campaign
- How I Use Frequency Capping
- Placements I Use
- Ad Creation
- As the Audience Shrinks
Basic Targeting Plan
Okay, not to give away the game from the first lesson, but I am going to target using a little-used method that anyone with a pixel on their website can use.
When you create a website custom audience, select “PageView” under “From your events.”
Then click “Refine by” and select “Aggregated Value.”
By default, you can now create an audience of people who viewed at least 2 pages of your website during the past 30 days (unless you change the retention).
I wrote a blog post recently on a bunch of ways that you can target quality website visitors, and this was one of them.
In the initial list of lessons, I targeted people who viewed at least 2, 3, 4, 5, and 6 pages of my website, increasing that threshold for each lesson. I plan to continue that approach for each additional lesson.
The Goal
Look, let’s be obvious about the initial goal. I want to reward and encourage engagement. Keep clicking pages of my website and you’ll be more likely to continue seeing these ads. Everybody wins.
But I also want to test what Facebook is doing these days. Back when I conducted the initial experiment, Facebook had no problem serving ads to HYPER small audiences. That hasn’t been my experience lately.
But, what is the threshold? Is there one? How small is too small?
According to Facebook, the upcoming audience sizes are as follows:
- 7+ page views (180 days): 3,600 – 3,700
- 8+ page views (180 days): 2,900 – 3,100
- 9+ page views (180 days): 2,400 – 2,500
- 10+ page views (180 days): 2,100 – 2,200
(Yes, Facebook uses ranges now for audience sizes).
What’s interesting is that these ranges are not in line with what Google Analytics reports. I know, that’s generally expected, but my “All Website Visitors” audience for 180 days from Facebook matches up very closely to GA.
Anyway, we’ll keep going up from there. I can’t tell you what the audience sizes are after 10+ because I just created them and Facebook needs to take the requisite 473 days to generate audience sizes (close the sarcasm font).
What to Look For
If you’re relatively active on my website, you should see my ads. Hell, by viewing this blog post, you really only need to view one more page to make sure you’re in the initial audience (DO IT).
You’re looking for videos from me. The first lesson can be viewed either by watching the entire video in the ad or clicking the link in it.
It looks like this…
Each subsequent lesson looks similar, but with a couple of differences:
1. It will be clear how many pages you’ve viewed to unlock that lesson, and
2. The video will only be a 15-second “short” introducing the lesson.
From lesson 2 on down, you’ll need to click the link to view the lesson. By doing so, you’ll be excluded from viewing that ad again (which is good for you and me!).
Let Me Know
I’m still learning a lot from this lesson, and I’m looking forward to seeing where it goes. Have you seen these ads yet? What do you think?
Let me know in the comments below!