It’s a common problem. Meta advertisers, in search of the perfect combination of advertising strategies, overcomplicate things and make it worse. Several factors contribute to this problem.
First, we assume that “complicated” means “better” and “sophisticated.”
How could an Advantage+ Shopping Campaign with no targeting inputs and one ad set perform better than my execution of a 10-step strategy with multiple campaigns, ad sets, and audience segmentations?
Second, our answer to fixing a campaign that isn’t working as we’d like is to tweak, add, and add some more. Duplicate the campaign, create a new ad set, try a new optimization, or target a new group.
It’s not that the simplest strategy is always the best. But, less complicated strategies provide more clarity. By simplifying, you aren’t driving up your costs by competing against yourself or restricting the algorithm. If something isn’t working as well as you’d like, the focus is on ad copy and creative, your offer, and your website.
Look, I was once the king of complicated strategies. My favorite thing to do was create complex Evergreen Campaigns that used 10 ad sets to move a very small number of people through an ad funnel. But, things have changed.
Some advertisers are stuck in the past. Others are frustrated with their results and are trying everything. And we listen to far too many “gurus” with complicated strategies that sound good because of their complexity.
Limit Unnecessary Complexity
Don’t take every recommendation in this post as gospel. There are too many factors that contribute to performance to decide that one human’s advice is best for everyone.
I’m not always right. Sometimes there’s nuance. And even my advice below should be taken as a general approach. I don’t want you to always do what I suggest. I just want you to think about these things.
Your main takeaway should be this: Make a conscious effort to limit unnecessary complexity.
“Complex” is rarely helpful. By adding more variables, you make it more difficult to understand what is working and what isn’t. You’re watering down your results. It’s difficult to know what you need to change to get things back on the right track.
Complexity can be created by adding campaigns or ad sets. It can result from micromanaging targeting or placements. It can even be found in testing your ad creative.
It’s not that you should never do these things. But, before you do, ask yourself whether it’s necessary. Is it truly helpful?
There are always exceptions to what I recommend below. It could be due to large budgets, specific company goals, or unique circumstances. I get it. Sometimes you can’t avoid it. But, understand why “simple” is often better.
Let’s discuss the main things that you can simplify…
1. Campaigns
I host regular one-on-ones and help clients think through their advertising. One of the first things I often see is an Ads Manager cluttered with a whole bunch of campaigns.
Campaigns for sales, leads, traffic, engagement, and awareness. Multiple campaigns for a single objective.
There are only two objectives that I would recommend are required for virtually every business:
- Sales
- Leads
If we really want to simplify things, an Advantage+ Shopping Campaign is often the best way to run a sales campaign. No targeting and one ad set. All of the focus is on your ads.
Regardless, focus on sales and leads — or on conversions of some kind. Everything else is extra and needs a good business reason for doing it. Top of funnel objectives are rarely worth the money because Meta doesn’t have a way to optimize for quality traffic or engagement. Instead, you’ll typically get a bunch of empty clicks.
Save the money you were going to use on those top of funnel campaigns and push them towards leads or sales. You will build awareness, engagement, and traffic incidentally with those campaigns.
Also make sure you actually need the multiple sales or leads campaigns before you create them. If you have specific business goals, multiple campaigns can be difficult to get around.
Just remember that the more campaigns you create, the more ad sets you create. And that can eventually become problematic, in the form of Auction Overlap, which can drive up your costs.
2. Ad Sets
This is connected to limiting your number of campaigns, but also not. If you create 20 campaigns, that’s at least 20 ad sets. That, by itself, could be a problem.
But, you could also have two campaigns that each house 10 ad sets. Maybe this is an exaggeration, but advertisers do it. In most cases, it’s completely unnecessary.
If you create multiple ad sets to segment your audience, for example, you are contributing to Audience Fragmentation. This makes your ad spend less efficient.
You can’t always avoid creating that extra ad set. But, whenever possible, aim to consolidate.
3. Targeting
Since targeting is the primary motivator for advertisers who create multiple ad sets, this is a good transition.
Targeting may be the best example of how advertisers overcomplicate things. While it made sense in the past, it almost never does now.
If you’re optimizing for some sort of conversion, you should use Advantage+ Audience (assuming you haven’t created an Advantage+ Shopping Campaign). Provide some audience suggestions and allow the algorithm to do its thing.
There’s no need to create multiple ad sets to test the use of different audience suggestions. Those suggestions are unlikely to be all that impactful anyway. They’re just a starting point. Once the audience expands, those multiple ad sets will be nothing but overlap.
If you prefer original audiences over Advantage+ Audience due to the perception of additional control, keep in mind that your targeting inputs are often expanded:
- Advantage Detailed Targeting and Advantage Lookalike are automatically on and can’t be turned off when optimizing for conversions
- Advantage Detailed Targeting is automatically on and can’t be turned off when optimizing for link clicks or landing page views
The audience is often expanding anyway.
Also, don’t assume that expansion is bad and needs to be avoided. While eliminating expansion can lead to good temporary results, it’s not scalable. You can’t, for example, keep targeting your email list and website visitors while spending $100 per day and expect to get good results beyond a short window.
Remarketing is mostly dead. First, the algorithm is smart enough now that it will automatically target people based on your conversion history, pixel data, and prior engagement with your ads. This is even the case when you don’t provide targeting inputs with Advantage+ Shopping or Advantage+ Audience.
I will still use my remarketing custom audiences as suggestions for Advantage+ Audience. Even then, I don’t know how much it matters. But, it gives me peace of mind that it’s at least starting with that group.
Simplify your targeting. Embrace the fact that your targeting inputs are far less impactful than they were in the past. Stop obsessing over isolating the perfect combination of demographics, detailed targeting, and lookalike audiences. Ditch creating multiple ad sets for the purpose of audience segmentation.
Don’t lose any sleep over it. This is a good thing because it allows you to focus on your copy and creative.
4. Budget
All of these things are related.
The vast majority of advertisers have a finite budget. You can’t spend more than a certain amount per day or month.
And yet, you’re spreading that budget across a cluttered list of campaigns and ad sets — many of which are unnecessary. You complain about bad results and your inability to exit the learning phase. And the whole time, this problem is easily solvable.
Create fewer campaigns. Create fewer ad sets. But spend the same amount. Consolidate your budget that was spread across campaigns and ad sets into fewer targets.
This will give you the best chance of spending enough to help the algorithm learn and generate optimal results.
5. Performance Goals
Your performance goal may be the most important part of the campaign creation process.
I know, the ad copy and creative are incredibly important. But, great copy and creative may not overcome the wrong performance goal. If you use the right performance goal, mediocre ad copy and creative could still get you acceptable results.
The performance goal is exactly that: It defines what you are trying to accomplish. This helps Meta know how to deliver your ads and who should see them. It helps determine whether your ad set is working or underperforming and something needs to be corrected.
What’s crazy to me is that this shouldn’t be complicated, but advertisers love to complicate it.
As discussed earlier, your priority should be to optimize for conversions of some kind. You can set a performance goal to Maximize Conversions or Maximize Value of Conversions.
And then define what exact conversion type is most important to you. It could be purchases, leads, complete registrations, or potentially something else.
The algorithm will then focus on getting you those conversions. It wants to make you happy.
But, don’t get cute.
If you optimize for link clicks or landing page views, the algorithm will be focused on getting you link clicks or landing page views. They could be accidental clicks, bots (before detected), or people who click on everything. But, these people may not have any interest in your ad or your product.
If you optimize for ThruPlay, the algorithm will find ways to get people to watch at least 15 seconds of your video. That includes prioritizing placements where users are forced to watch video ads and can’t skip them. You assume these people cared about your video, so you create remarketing campaigns to target them. But, many didn’t care.
Keep it simple: Set a performance goal that defines exactly what you want.
This is the only way that you and Meta’s ad delivery algorithm will be on the same page. You can’t complain about low-quality traffic if you didn’t define you wanted high-quality traffic. You can’t complain about not getting purchases if you told the algorithm you wanted add to carts.
6. Bidding
The ad auction is dependent on three things:
- Your bid
- The likelihood that someone will engage with your ad
- Ad quality
Unlike the typical auction, your bid isn’t everything. The highest bidder doesn’t necessarily get the impression. And really, you don’t want that to be why you win the auction anyway.
If you don’t touch anything, Meta bids for you. In most cases, it’s using the Highest Volume bid strategy. Meta’s focus will be to get you the highest volume of optimized actions within your budget. If you optimize for Value, the Highest Value bid strategy is default.
Otherwise, you can use a Cost Per Result Goal, ROAS Goal, or Bid Cap.
But, in most cases, don’t bother. You’re usually going to be disappointed. You’re not going to get magical results because you set a Cost Per Result Goal of $.01 and Meta unearths people willing to buy your product at a penny per purchase.
More often than not, your manual bidding will lead to spending less of your budget and getting fewer or worse results. It’s not that you should never try manual bidding. But, it should mostly be used as a last resort when you can’t get anything else to work.
7. Placements
If your only active ad sets are optimized for some sort of conversion, this is the easiest step possible. Do nothing. Keep your hands off and use Advantage+ Placements.
It’s not that there aren’t low-value placements. Audience Network is notorious for generating low-quality clicks and video views. But, if you’re optimizing for conversions, the algorithm knows about these pitfalls, too. You can bet that very little, if any, of your budget will be spent there.
Not, of course, unless that placement leads to conversions. And to be clear, impressions that don’t lead to a direct conversion can have value, too. One user may see three or five ads before finally converting. Some of those lower-performing (and lower-priced) placements may contribute.
Where you need to be careful is when optimizing for anything other than conversions. As we know, Audience Network leads to low-quality clicks. And since you can’t set a performance goal of high quality link clicks or landing page views, Meta will fill your results with those clicks if you set a performance goal of link clicks or landing page views.
A similar problem is found in Audience Network Rewarded Video when optimizing for ThruPlay. Third-party apps monetize themselves with Meta ads for this placement. People can watch videos in exchange for virtual currency that is used in the app. These people don’t care about your video.
Of course, there are other examples. But, this is another reason why optimizing for anything other than conversions is a complicated game of whack-a-mole. You need to do all you can to control quality, and that includes removing problematic placements.
But, again, that’s not an issue when optimizing for conversions. Keep it simple and use Advantage+ Placements.
8. Ad Copy and Creative
Ad copy and creative are super important. If they aren’t the most important part of your advertising, they’re at least in the discussion.
But, you don’t need to overdo this.
Meta says that there’s no benefit to creating more than six ads for a single ad set. And if your budget is low, even those six will chop up your budget to the point of making results mostly meaningless.
As I’m sure you know, the algorithm will pick one or a handful of those ads rather quickly and run with them. This isn’t because those ads were clearly more effective at a high level of certainty, it’s because the difference is negligible and the algorithm had to run with something.
Create multiple ads if you have multiple ad ideas. But, don’t feel you need to create six. And don’t obsess over the results and what they mean from small sample sizes.
Sometimes, it’s best to create two or three ads and run with them. Not getting great results? Fine. Create two or three more. It doesn’t matter that you restart the learning phase because you weren’t satisfied with your results anyway.
9. Testing
I don’t want to completely minimize testing because it can be helpful. But, I also see advertisers stuck in a constant cycle of A/B tests that barely move the needle.
Over-testing happens when you don’t trust anything that Meta does automatically. You feel the need to scientifically define absolute winning ads and optimizations.
But, the testing itself costs money. Performance is almost always worse when you force the algorithm to A/B split the audience. And you’re not guaranteed to get results that are statistically significant that would have made the test more productive than simply running the ads the old fashioned way.
Again, there are exceptions. If you’re going to run a long-term campaign, testing ads makes sense. And if you have big budgets, knock yourself out.
But, these low-budget tests to find winning creative are virtually meaningless. Just run the ads. Let the algorithm sort it out.
You can “test” without always needing to find a winner. Give the system multiple ads to work with. Utilize Dynamic Creative or the text variations feature.
Once again, complicating things with a test isn’t always the best path to profitability. Sometimes the simplest approach is the answer.
10. Reporting and Interpretation of Results
What’s a good CPC for this industry? Is this an acceptable CTR? Why is my CPM so high? How can I get it down?
Just stop…
Meta offers limitless metrics that can distract you. Many of them provide some value. But don’t obsess over the secondary metrics.
Keep it simple. Focus most on your goal action and the cost per goal action.
Not getting the cost per goal action that you’re wanting? The secondary metrics can help tell that story. Maybe your conversion rate is great, but the CPM is going up due to competition. Or maybe the CTR is lower than normal, indicating that you need to improve your offer to get people to click. Or those secondary metrics are all solid, but you aren’t getting conversions — so you shift your focus to the landing page.
Stop freaking out about every metric. They’re all part of the story. But, only a couple truly matter. The rest are window dressing.
Your Turn
How have you overcomplicated your ad strategy?
Let me know in the comments below!
