There are a few questions that I get every single day. They typically fall into one of these buckets…
- Which bidding method should I use?
- Should I optimize for X or Y?
- Should I use daily or lifetime budgets?
I’ve gotta be straight with you: These questions frustrate me.
The reason may not be what you’d think. It’s not because I think there’s an easy answer. It’s because the answer is often different from person to person and should be discovered by each individual advertiser.
I know people hate when I refuse to give black and white advice on advertising questions. I’m not going to come out and say you should always do this or never do that. If one thing always worked and one thing never worked, Facebook would simplify the options for you accordingly.
Finding what works and what doesn’t comes down to testing and experimenting. Let’s take a closer look at why so many advertisers don’t test, and how you can establish your own testing process…
Why Advertisers Don’t Test
Most advertisers are afraid to test. I find that there are three primary reasons why these questions are asked instead of doing the testing that it takes to uncover the answers:
1. Advertisers lack confidence.
The vast majority of advertisers fall into the beginner or beginner-to-intermediate categories. In the grand scheme of things, there are very few truly “advanced” Facebook advertisers.
As a result, Facebook ads are overwhelming to most. Should you use Power Editor? Yeah, that’s more complicated. Should you advertise like you do with Google ads? What are the rules? What can you and can’t you do?
Because of this lack of confidence, advertisers don’t trust that they’ll be able to figure out the answers on their own.
2. Advertisers aren’t sure what works and what doesn’t.
There is a persistent misconception that there are universal truths when it comes to advertising on Facebook. Surely perpetuated by snake oil marketers who claim “magical formulas” and simple templates for success.
So marketers, assuming this is easy, want to know the shortcuts. They don’t want to waste their time with things that don’t work. And they have no idea what works and what doesn’t.
3. Advertisers don’t want to waste money.
This may be the big one, and I understand it. Maybe you have a low budget. Or you’re advertising for clients and you don’t want to waste their money by testing.
Let me tell you, though: This is not an excuse. Even with the smallest of budgets (we’ll get to that later), you can test.
There is No Replacement for EXPERIENCE
I often hear from marketers who are overwhelmed with Power Editor or aren’t comfortable with running Facebook ads. There’s a very simple solution to this: Dive in and get some experience!
Don’t rely on the experiences of others to guide you. Don’t be afraid of failure. There is no failure when it comes to running Facebook ads — only lessons learned.
While the experiences of others can be helpful to set general expectations, they do not replace hands-on testing. You need to test yourself. You need to experiment. You need to find what works and doesn’t work for you.
Comfort and expertise will come with time — but only if you test yourself!
Here are a few things you should test…
1. Test Each Campaign Objective
One of the most common questions I get is whether advertisers should optimize for website clicks or conversions when looking for a conversion. My answer: It depends!
I strongly advise that you create at least one campaign for every campaign objective you may use. While there are plenty, you don’t necessarily need to test them all. If you aren’t planning on promoting an app or event, for example, don’t worry about those objectives.
Ideally, you’ll run several iterations of these campaigns. However, at minimum you should simply create the campaigns without uploading. At least you’ll get the experience of creating them to see how ad set and ad creation differs for each.
2. Test Each Bidding Option
Should you optimize for website clicks or impressions? Should you bid manually or allow Facebook to bid automatically? Should you get charged for the action or the impression? Are the answers different since the latest bidding changes?
I can’t tell you the answer to these questions. As I say over and over again, there aren’t universal truths to this stuff. What works for me may not work for you, and it depends on the situation.
So within each of the campaigns you’ve created, create separate ad sets for each bidding method. See how each one works for you!
3. Test Each Placement
I often hear people say that sidebar doesn’t work. This is a lazy generalization, and it often is far from the truth.
What is true is that it’s cheaper to reach people on the sidebar. It’s far cheaper, in fact. And even if you have a lower click or action rate there, the final cost per action may be lower, too.
But guess what? You won’t know for sure without trying it out yourself.
You could create multiple ad sets, one for each placement. Or you could simply use Facebook ad reports to break down the performance by placement. This would certainly be the most cost effective route, even if it doesn’t allow you to spread budget evenly.
Remember not to be distracted by peripheral stats. Don’t worry about how much it costs for impressions or what your click rate is. These things will absolutely vary greatly depending on placement. But ultimately, you should only care about your cost per desired action.
4. Test Scheduling Options
Should you use lifetime or daily budgets? Should you use dayparting?
I personally prefer daily, but I’m just one guy. You should find what works best for you!
Again, you’d need to create multiple ad sets for this to test it adequately. It doesn’t need to be incredibly scientific. Just get into the practice of trying out both to get a feel for what works and when.
5. Test Post Types
Should you use link posts? Image posts? Videos? Carousel ads? Drive to a landing page or use lead forms?
I’m a big believer in using a post type that is consistent with your objective. If I want traffic or conversions, I use link shares or carousel ads. If I want engagement, I use photos or videos.
Again, though, that’s me. I only have that opinion because I’ve tried things out and found what works for me.
Find what works for you!
6. Test Copy and Creative
Should you use one image or five image variations? Should you create multiple copy variations? How many will Facebook actually run?
By now, you should know the answer to this: It depends, and you won’t know what works for you until you experiment!
In general, my main rule of thumb is understanding your budget and required sample size to get meaningful results. If it costs $10 to get a desired action like a conversion and you have a budget of $5 per day, there’s no reason to have more than one ad.
But if you’re spending a few hundred dollars per day or if you are doing something that produces a larger sample size for a low cost (like website clicks or video views), you can start adding variations to the mix.
Just keep in mind that if you water down the results too much with too many variations, you won’t learn anything!
Budget LOW While You Learn
This may sound like a lot, and it may sound expensive. But do what works for you. And like I said earlier, you should at minimum go through the process of creating these campaigns without uploading them.
I don’t want you to waste money. However, you absolutely should establish a testing expense. The more and smarter you test, the more money you’ll make and save in the future.
Not having a budget is not an excuse. You must have a budget for this. After I was laid off four years ago, I established a $1 per day budget on Facebook ads. Sure, it took me longer to learn what worked and what didn’t, but at the same time I also had plenty of hands-on experiences!
You Won’t Figure It Out Overnight
If you haven’t tested before, know that you won’t figure this out overnight. And what you learn in your first tests may not apply to your next campaign as running ads is fluid and constantly changing.
Most importantly, though, is that you will be gaining experience. You will become more comfortable, and you will find the best possible processes for you.
What processes do you have for testing?
Let me know in the comments below!