A couple of weeks back, I shared The Master Brief on my current recommendations related to Meta advertising today. This week, I’m going to take a different approach.
While The Master Brief is an AI-generated summary, written in the third person, of my recommendations based on my most recent content, I’m going to write this post from scratch. No AI involvement, which is what you can normally expect from my blog posts.
You will certainly find some common themes between the two posts, but the format, tone, and way I explain these concepts will be different. I hope that would be obvious, since AI isn’t good enough to completely replace us yet.
Today, I want to lean heavily into the words and phrases I regularly repeat when giving out my recommendations related to Meta advertising. These will serve as the foundation of this post.
If you’ve read my blog posts, watched my videos, listened to my podcast, or attended any of our member strategy sessions, you’ve heard me say and read me write these things. I’ve found that this repetition is helpful for highlighting what you should and shouldn’t do in the current environment as an advertiser.
Each section is headlined by a quote you’ve likely heard or read from me before:
- “Magical strategies do not exist”
- “What problem are you solving?”
- “Prioritize a simplified approach to campaign construction”
- “The algorithm is literal”
- “Let go of the illusion of control”
- “Start and finish with your ads”
- “Evaluate performance in aggregate”
- “Let meaningful results be your guide”
They serve as a great foundation for a productive approach to Meta advertising that strips away the nonsense so that you can focus on what matters most.
Let’s get to it…
1. Magical Strategies Do Not Exist
The sooner you understand this, the sooner you can focus on productive tweaks that positively impact performance. There is no such thing as a Meta ads strategy that will universally work for everyone. And no strategy exists that will be the reason you get good results.
Advertisers commonly fall for the latest branded strategy as the solution to their problems. But these snake oil claims are typically traps for buying a new course or driving up views on a video.
You can technically get good results with any strategy. But the strategy isn’t the reason for your good results. And, more often than not, the strategy will limit your potential.
Most of these strategies thrive on complexity. Lots of campaigns and ad sets with manual settings and customizations. If it’s complicated, it sounds more sophisticated.
2. What Problem Are You Solving?
The nothingness of these strategies tends to be exposed when you repeatedly ask a simple question.
You created multiple campaigns. Why? What is the problem, proven in data, that this solves? And what does the data say that proves it is the solution?
You created multiple ad sets. Why? What problem does it solve?
You restricted by age, gender, remarketing audiences, or another targeting input. Why? What problems are you solving? If a problem exists, would a value rule be a better solution?
You turned off a placement. What was the problem that motivated this? Would a value rule be sufficient?
You turned off all of the Advantage+ Creative enhancements. What evidence do you have that these were hurting results?
It’s not that there is never a problem that is solved by adding complexity. Instead, don’t add complexity for complexity’s sake, without clear reasoning and proof that doing so is the solution.
And this decision needs to be based on smart, specific data. “Audience Network is junk” isn’t a reason to remove it universally. “Our audience is men aged 25-44” isn’t a reason to remove women and age groups outside of that range. “These enhancements look terrible” is rarely a good reason to turn them all off.
Make sure you have data-backed reasons behind the actions you take.
3. Prioritize a Simplified Approach to Campaign Construction
My primary complaint with these magical strategies that thrive on campaign complexity is that they intentionally water down your budget. Advertisers need to understand that the more they segment the budget that they have, the less efficient it becomes.
As a rule of thumb, it’s generally recommended that you generate at least 50 optimized actions per week per ad set (or campaign with Advantage+ Campaign Budget on). This helps you exit the learning phase, though the more important accomplishment is getting enough volume so that performance is stable and predictable.
Segmenting that budget multiple times will do the opposite. You are creating a situation where you have multiple campaigns or ad sets with small dedicated budgets, none of which is capable of driving the volume of results necessary for you to achieve stability.
In a perfect world, you’ll create a single sales campaign with one ad set that consolidates your budget toward your primary goal. This gives you the best opportunity to exit learning and achieve stable results.
Of course, we don’t always live in a perfect world. You may have reasons for creating campaigns or ad sets for multiple business goals, product lines, or locations. But even if this is unavoidable, it’s important that you understand the potential negative impact of watering down your budget.
Value the simplest approach possible. Minimize the number of campaigns and ad sets you create. Have very clear business reasons for adding complexity when you do.
This also goes for adjustments to Meta’s default settings. Prioritize the broadest targeting using Advantage+ Placements and Meta’s Advantage+ Creative enhancements. This doesn’t mean these settings are universally perfect. If you go away from defaults, make sure that there is a very good reason for doing so, backed in data.
4. The Algorithm Is Literal
Understanding how Meta’s ad delivery algorithm works is fundamental to an efficient approach to campaign construction.
Once you know the algorithm’s strengths and weaknesses, you’ll understand both when the simplest approach is most effective and where potential problems come up that will need to be solved. You’ll also know where you can waste money if you’re not careful.
“The algorithm is literal” means this: When you define your goal action, Meta’s entire focus is helping to get you more of that thing within your budget. This can be both good and bad.
It can be bad when optimizing for top-of-funnel actions like link clicks, landing page views, ThruPlay views, and post engagement. When you use a performance goal that maximizes the number of any of these things, that will be Meta’s entire focus with your budget. It doesn’t matter to Meta whether these people will ultimately buy from you.
Here are some examples…
If you set a performance goal to maximize the number of link clicks or landing page views, expect to get a lot of low-quality clicks. Meta doesn’t care whether these people do anything on your website because you only said that you wanted the click. Meta can find these clicks from weaknesses in placements or from people who seem to click on everything.
If you set a performance goal to maximize the number of ThruPlay views, expect to get a lot of people who watch your video and do nothing else. Your logical assumption is that this will improve brand awareness and that the people who watch for 15 seconds or more care about your product. But Meta can exploit weaknesses like placements that force people to watch your video. They may not be watching for 15 seconds because they are interested.
If you set a performance goal to maximize the number of leads, Meta doesn’t care whether the leads convert. The only focus is on getting you as many leads as possible, and Meta may be able to exploit weaknesses found in demographic groups or locations where people are most likely to complete forms.
You need to be aware of these potential weaknesses when it comes to the algorithm. But such weaknesses are far less likely when your goal action is a purchase.
When you define your performance goal to maximize the number of conversions where the conversion event is a purchase, Meta’s focus will be getting you more purchases within your budget. If you’re unable to get purchases, it’s because your ads and landing page aren’t converting them.
Meta’s not wasting money on low-quality clicks because those clicks won’t help achieve your goal. There isn’t a weakness to exploit in placements or demographics or location that can get you more purchases. Meta will show your ads to people most likely to purchase. Your ads take over the responsibility from there.
Of course, if the performance goal is to maximize the number of conversions where the event is a purchase, you could still run into a potential problem. Maybe you’re getting low-value customers who make only a single small purchase. This is possible because Meta’s only goal is getting you more purchases.
But you can correct that problem with an adjustment. If you’re not getting an acceptable Return On Ad Spend because Meta’s getting you low-value customers, you can instead define your performance goal to maximize the value of conversions. Meta will then focus on getting you a higher ROAS.
The bottom line in all of these scenarios is that Meta is trying to get you the action that you want, as defined by your performance goal. Depending on the goal action, that can lead to low-quality results.
But you will be most aligned with the algorithm when using a performance goal that maximizes the number or value of conversions where the conversion event is a purchase.
This is also why you should prioritize dedicating most, if not all, of your budget toward purchase-related performance goals. Leads can be acceptable if you’re not able to generate profitable results from purchases with your budget. But even then, you should be aware of the potential lead quality issue so that you can address it before it becomes wasted ad spend.
5. Let Go of the Illusion of Control
One of the biggest mistakes advertisers can make in the current environment is attempting to control things that they either can’t or don’t need to control. If you don’t know what you’re doing, you’re bound to hurt results in the process.
You don’t need to create multiple ad sets targeting different lookalike audiences and detailed targeting groups. These are only audience suggestions for most performance goals now, so you’re creating multiple ad sets that can reach the same people.
You don’t need to create a separate remarketing campaign or ad set to show Meta the people most likely to convert. Meta already prioritizes these people by default, and you can prove it with audience segments and sales campaigns.
You don’t need to turn off the Audience Network placement when using a performance goal that maximizes the number or value of conversions. While it’s the source of cheap and low-quality traffic when optimizing for link clicks or landing page views, Meta won’t lean into this placement if your goal is more purchases or leads.
You don’t need to isolate the best primary text, headline, creative, or even ad. Creative diversification is the new targeting lever that will help you reach many groups of people with your ads. A successful ad set will consist of many effective combinations of text and creative, not just a handful. Your job is to provide the diverse assets that Meta can combine to find the right audience.
You no longer have the control you once enjoyed. Most control you think you have is an illusion. Trying to force your control now is rarely an effective strategy.
Once you embrace your lack of control and your place in this process, you’ll be able to get out of the way so that you can help guide your ads to good results.
6. Start and Finish With Your Ads
Once you understand that campaign performance isn’t tied to some convoluted strategy, you can focus on what drives results: Your ads.
When you take a simplified approach to campaign construction, you may have only one campaign and one ad set with minimal, if any, customizations to settings. And when you do, your entire focus should be on your ads.
Your initial campaign planning won’t be about how to assemble a complex web of campaigns and ad sets, targeting different groups of people. It will start with answering these important questions:
- Who are your primary customer personas?
- What are their common pain points?
- What are the solutions that are provided by your product?
- How will they feel when these problems are solved?
- How can you communicate these messages through text, images, and videos?
This is the foundation of your initial set of ads. They don’t need to be perfect. You also don’t need to create dozens of them to start. But your ads should reflect this problem/solution thought process.
If, after a few weeks, you aren’t getting the results you want, the answer is usually found in your ads. You’re not going to fix it by creating more campaigns and ad sets. The solution is finding the ad creative and messaging that resonate with your target audience. So create a uniquely different set of ads and keep trying.
Assuming you don’t have a tracking or landing page issue (both should be confirmed), the answer to better results is almost always found in creating better ads.
7. Evaluate Performance in Aggregate
One of the easiest traps that control-hungry advertisers can fall into is what Meta calls “The Breakdown Effect.” It happens when advertisers make assumptions about performance from a small sample size of results.
For example, you may see that Meta is spending a high percentage of your budget on one ad while spending a fraction on one that you’ve labeled a “high performer.” You assume that its low cost will remain consistent when scaled and that it will surpass the ad currently getting the budget. But Meta has already determined that this won’t be the case.
The same phenomenon is found when analyzing budget distribution to placements. You may see that one placement is doing the heavy lifting while another gets a smaller percentage of the budget at a lower CPA. You naturally assume that’s an inefficient way to distribute the budget. But Meta knows that costs will increase on what appears to be the higher-performing placement if given more budget.
This is why you should evaluate your results in aggregate. Do not obsess over the performance and budget distribution for individual primary text blocks, headlines, images, videos, or even ads. Distribution will ebb and flow. It will adjust dynamically.
Your only question should be this: Are results for the ad set good in aggregate?
If the overall performance of your ad set (or campaign if using Advantage+ Campaign Budget) is good, that’s all that matters. You’re unlikely to “optimize” results by isolating a few top-performing assets and forcing Meta to show them.
This is the question you should ask before making any change based on a perceived distribution problem. If your results are good in aggregate, there isn’t a problem to be solved. If the results are bad, it’s time to make changes.
Once you start viewing results from a macro rather than a micro level, you’ll make fewer and better decisions.
8. Let Meaningful Results Be Your Guide
This is certainly tied to the item above because, far too often, advertisers will make assessments about distribution based on results that aren’t meaningful.
They obsess over CTR, CPC, or CPM when the goal is sales. It’s not that these metrics aren’t useful, but they aren’t the primary metric that determines success.
But even when advertisers focus on conversion results like Cost Per Conversion or Return On Ad Spend, they’re prone to miss the meaning. A common mistake is taking conversion results at face value without understanding what went into those results.
For example:
- Is the analysis based on a volume of results that would reflect predictability?
- How many of the results are click-through, view-through, and engage-through?
- How many click-through conversions fall under 1-day click versus 7-day click?
- How many of the results does Meta believe are incremental?
- How many of the results were the first conversions vs. repeat conversions?
- How many of the results were from remarketing audiences vs. new audiences?
Answers to these questions will help you determine how meaningful your results are. And if they’re meaningful, they can lead you to your next step. Use the breakdowns by attribution setting and conversion count to reveal this information.
I get a lot of “Should I…” questions from advertisers that never have universal answers. My response is always some version of “let meaningful results be your guide.”
Should I increase my budget? Let meaningful results be your guide. Start with 1-day click results. Are they profitable? How profitable? Knowing that even these costs will go up, it should give you an idea of how much room there is to increase your budget.
I’m not getting 50 purchases per week, so should I optimize for leads instead? Let meaningful results be your guide. Are you still getting acceptable results when optimizing for purchases, even if the volume is low? What are the cost per closed lead and lead value when optimizing that way?
How often should I publish new ads? Let meaningful results be your guide. Expect ads to initially surge in performance as Meta first shows them more to remarketing audiences and people most likely to convert. Costs will likely stabilize for a prospecting-heavy audience after that. If costs aren’t good enough, create a new batch of ads. Otherwise, there’s no requirement to have a regular ad creation cycle.
Surface-level and small-sample results can cloud your judgment when looking for the most productive next steps. That’s why it’s critical to analyze meaningful results in aggregate before making any changes.
Your Turn
What questions do you have about an effective approach to Meta advertising today?
Let me know in the comments below!
