A few days ago I visited a former client to help him and his team work through a problem with their Facebook ad targeting: They were in halfway into a Kickstarter campaign. They’d hit their relatively low external goal of $20,000, but they were an order of magnitude away from their real goal. Their ad performance was terrible.
I’m not going to embed any links to the campaign. I want to protect the privacy of a friend and client. And just as importantly I don’t want to get sued by their wholly incompetent advertising agency.
Slander, you say? No. The facts support my point of view.
“Let’s target everyone”
Within three minutes of looking at the audiences in my friend’s Facebook Business Manager account, one problem became glaringly obvious: The ads had (almost) no targeting. Why is “almost” in parentheses in the last sentence?
Let me explain.
If you create an ad on Facebook and target everyone in the United States, aged 18 and up, you end up with 230 million people. To be clear: This is every adult in the United States who has a Facebook profile set to allow ads.
The ad agency my friend hired was targeting 200 million of the 230 million American adults on Facebook.
(Almost) no targeting.
What went wrong?
To put it simply, the agency just kept adding interests, behaviors, and demographics to the target audience until it included nearly every American on Facebook. I’ll explain how to avoid this later in this article, but first I think it’s good to explain why this garbage heap of an audience is a bad idea.
Why it’s bad to target huge Facebook audiences.
Targeting 87% of American adults on Facebook is the equivalent of buying an ad during the Super Bowl. It’s great if you have a gargantuan budget and sell products that appeal to nearly everyone. Think Hyundai, Budweiser, Verizon, and Netflix. But as many a tech startup learned just before the dot com bust, buying Super Bowl ads is bad business if you’re not mass market.
Spreading the money too thin.
It gets worse: My friend’s one-month ad budget was $5,000. That’s plenty (really!) for most Kickstarter campaigns, as long as the ads are well targeted. But if they’re not, it spreads the money far too thinly to have any effect. Some ads may, by random chance, have reached people interested in his product, but they’re far more likely to reach the 90% of people who aren’t interested.
The results proved this out. After spending more than $2,000, only a handful of people bought the product because of Facebook ads.
Different audiences need different messages.
Let me point out a second problem with targeting “everyone”. Even if you do have mass appeal, different people are likely to respond to different advertising messages.
Some people will buy an entry-level sedan because it’s a good value and reliable. Others will buy the same car because of its aggressive styling and the fact that they can option it out for cheaper than a “near luxury car”. Facebook ads give you the ability to target both groups with the messages that are most likely to resonate with them. Why wouldn’t you take advantage of that?
But, they tried!
Here’s what’s really funny sad: The ad agency actually tried to target the right people! However, from their approach, it’s clear that they 1.) didn’t check the total audience size to see that they ended up targeting 87% of Americans on Facebook and 2.) didn’t understand a fundamental principle of Facebook ad targeting.
What is that fundamental principle? Facebook ad audiences aren’t ever-expanding circles.
Facebook ads are like Venn Diagrams.
What does good targeting look like in practice? The best way to show you is through a real example, so I picked a project at random from Kickstarter: It’s a fancy nitro brew coffee maker. (NOT my friend’s product.)
Circle 1: Coffee Drinkers
Let’s start with the obvious. We want people who are interested in coffee.
I’ll target people who Facebook’s algorithms think are interested in coffee (for whatever reason) and people who are interested in Starbucks, as it’s the largest coffee chain. It’s a good start, narrowing our audience from 230 million to 78 million.
(Adding additional coffee chains did not increase the audience size. I tried.)
Circle 2: Working Adults
Now I’ll draw the next circle in my Venn Diagram, limiting the age from 23 to 45. Why these? We want people old enough to have an apartment or house (not a dorm) where they can keep the contraption, but also within the most likely age range for Kickstarter buyers. (They’re still cool… like I barely am!)
This further reduces the audience to 46 million people. We’re down to 20% of the people on Facebook. Still too many. Let’s keep going.
Circle 3: Luxury Tech Buyers
This Nitro Coffee thing is kind of pricey, starting at $159 for the early bird special. A quick search of Amazon gives me the impression that most coffee makers only cost about $50. This one is AT LEAST three times the average price, putting it on par with high-end capsule coffee makers.
We need to reduce the audience further to focus on people who are likely to spend lots of money on gadgets. (No judgment here! I shelled out $200 for a coffee maker.)
I’ll target people who like luxury cars, luxury watches, luxury phones, and expensive kitchen gadgets. This brings us all the way down to 8.2 million people or 3% of Facebook users in the US. (I left out some exotic car brands like Lamborghini and Ferrari because I didn’t want to inadvertently target F1 fans.)
Circle 4: Crowdfunding Connoisseurs
Finally, let’s limit our target to people who like Kickstarter, Indiegogo or crowdfunding, thereby selecting the people who are already familiar with buying through these platforms.
This gives us a final audience size of 2 million people. That’s 0.8% of the American adults on Facebook and a much better starting point than 87%.
Why is this 0.8% audience so much better than a bigger one?
- We’ll use our money more efficiently to target people truly interested in our product.
- We can adapt the text and images we use in our ads to this specific group.
- People often need to see an ad (or different versions of an ad) multiple times before clicking.
- Targeting a smaller number of people means we can increase the frequency with which each individual sees our ads.
Will it work?
Only time will tell. My friend will be using the five audiences we developed together with a new campaign, launching soon.
If you want to see the final audience I came up with for the coffee maker, all the details are at the bottom of this post.
You can use Facebook’s ad interface to create audiences like this, but I prefer to use a tool called AdEspresso. I don’t get any kind of a referral fee for recommending them. I honestly like what they’ve done.
My company, Bright Shiny Robot, isn’t a full-service ad agency. That said, if your ads aren’t converting well, or if you’re not running ads yet and need some help getting started, feel free to get in touch. You might be surprised what we can accomplish together in a few hours.
YOUR AD TARGETS PEOPLE:
- Location – Living In: United States
- Age: 23 – 45
- Placements: on pages: News Feed on mobile devices, News Feed on desktop computers, Right column on desktop computers, Instant Article, Instream Video on mobile devices, Instream Video on desktop computers, Marketplace on mobile devices, Marketplace on desktop computers, Third-party Apps and Websites on mobile devices, Instagram Feed or Messenger Inbox
- Interest expansion: Off
- or Kickstarter
- or Indiegogo
- AND Behaviors(10)
- Facebook access (mobile): iPhone 8
- or Facebook access (mobile): iPhone 8 Plus
- or Facebook access (mobile): iPhone X
- or Facebook access (mobile): iPhone XR
- or Facebook access (mobile): iPhone XS
- or Facebook access (mobile): iPhone XS Max
- or Owns: iPhone 7
- or Owns: iPhone 7 Plus
- or Owns: Galaxy S9
- or Owns: Galaxy S9+
- or Interests(17)
- Luxury Watches
- or International Watch Company
- or Rolex
- or Sous-vide
- or BMW
- or Audi
- or Mercedes-Benz
- or Jaguar Cars
- or Jaguar Land Rover
- or Cadillac
- or Omega
- or Patek Philippe & Co.
- or Porsche
- or Hublot
- or TAG Heuer
- or Jaeger-LeCoultre
- or Lexus
- AND Demographics(3)
- Household income: top 10% of ZIP codes (US)
- or Household income: top 10%-25% of ZIP codes (US)
- or Household income: top 25%-50% of ZIP codes (US)
- AND Interests(2)
- or Starbucks