Why does Meta Ads BREAK when you scale?

Key Takeaways

If your meta ads work in the beginning, and the moment you try to scale and push more budgets, it breaks, then this video is for you. In our experience after auditing hundreds of meta ads accounts, in 95% of the cases, there are just six failure modes that are the reason that advertisers are not able to scale their campaigns. So, that's why in this video, I'm going to introduce all those six failure modes, tell you what's happening, what are the symptoms to diagnose that this is the failure mode that's causing you not to scale, and what are the solutions for each of these six failure modes. The first one is the most common, which is the concept of warm audience pockets. You have to understand one thing, when you launch a meta ads campaign, meta doesn't start from scratch. It already knows that what kind of audiences are engaging with advertisers in your industry, in your niche, and your competitors. And they know which are the audience which are close to buying. So, for most advertisers and industries, meta creates kind of this hierarchy of audiences, placing placing the audiences with the highest probability to buy at top. So, when you launch a campaign, ideally with a lower budget, meta initially goes with the easy mode wins. These are the highest readiness buyers, and that's why your CPA initially looks magical. But, when you try to scale or let the campaign run for a few days, in most cases, what happens is meta runs out of these warm audiences. And the frequency rises, and meta is forced to basically expand to other audiences and explore other audiences specific to your business. And that is when you will see that performance drops after a few days or when you scale. But, the question is, how do you know this is the reason for you not being able to scale? Now, what we have seen is, when your daily budget is between $50 to $200, ideally, this is when you will see great results. But, the moment you cross $400 or $500, that is when you will see you won't be able to even get sales or conversions as much as you used to get with $200 budgets. Because meta goes crazy in expanding to new audiences and once you see this problem, you will see that if you have retargeting campaigns live, they will still do fine, but the prospecting is which will suffer. And one thing you have to remember is if you do not make any changes at all, no budgets, nothing and still she see you have this issue then this is exactly what is happening to your account. And one of the important markers in this one is you will see that your frequency of your campaigns will decrease suddenly and your CTR will drop a lot. In certain cases, a lot of companies in their retargeting promote certain offers, which is like a discount. And overall, the prospecting and retargeting blended CPA is good. In those cases, you can scale that. You don't have to worry about prospecting is bad. Look at the blended because overall driving your business. And one more thing to remember in this case is that it is not your offer or it's not you as an advertiser who was generating results before and you're not able to scale. Actually, the performance starts after Meta moves to the cold audiences. Those were ready to buy, other advertisers had worked on them, Meta had understood on other advertisers budgets that these are the users who might buy the product or service. So, it really is not your offer or your creative that did the job. So, once Meta moves to cold audiences, that is when you will actually know whether your marketing strategy overall is working or not. This leads us to our failure mode two, which is learning never stabilize. And I see this a lot with new advertisers. People who have been doing Meta for a while, they know this, but new people, this is the trap they fall in generally. Now, these people think that they're optimizing, but what they're all they're doing is basically resetting the learning system of Meta. See, you need to understand one thing, Meta algorithm is very strong, but the only caveat is it needs stability to learn who converts to your business. And I have seen people who start touching from day one everything about the campaign, the budgets, targeting, bids, creatives, conversion events, and a lot of so they have a campaign they're like, "No, it's not working on day two. Let's move to optimizing towards add to carts. Now, let's move to sales. Now, let's move to LTV. Now, let's move to ROAS." Some people think that, "Okay, if I spend $1,000 or $2,000 at a campaign, that's enough for the campaign to learn." Actually, it does not depend on the budget alone. The most important factor in determining whether Meta has learned anything is that each ad set or campaign needs to have at least 50 conversions in a week. That is when Meta basically says, "Okay, we have enough data to learn." So, it depends on the conversions, actually. Now, the biggest symptom of this being your fail- failure mode is that your ad sets and campaigns will always be stuck in either learning or learning limited. You can check that in your ad manager dashboard. The other symptom you will see is if you check weekly or daily reports, your CPA will swing a lot. Some days you will have, let's say, a CPA of $100, someday you will have 23. Some days there will be no conversions at all. If that is happening, that and along with the learning and learning limited, that means this is the issue. You're frequently making changes to the campaign, which you should not. Now, you might ask me, "Okay, so what do we do in this case?" See, there are two rules you have to remember in this case. So, the first one is consolidate your spends, which means if you have 12 ad groups and six campaigns, make sure you have only one campaign and two ad groups or two campaigns or four ad groups, so that you basically hit this number, 50 conversions every week. That is rule number one. Rule number two is whatever you are optimizing, remember always, at one point you only have to basically change one parameter. You cannot go and optimize, change your ROAS target, and switch all your creatives at the same time. Even if it works or doesn't work, you will not have any learning. You wouldn't know what caused what. So, always you have to be a bit patient if you're a new advertiser. You have to change one by one, and that is when basically any experiment you will have a learning. This creative works, this type of creative doesn't work. ROAS at this point works, this doesn't work. Now that we are talking about scaling our marketing efforts, I want to take a minute and talk about my best recommendation for email marketing and automation tool, which is Omnisend. I've been using Omnisend for a while and recommend to all my clients who are starting with email automation or cases migrate to make better use of this channel. Thing I like most about Omnisend is it's quite intuitive and very user-friendly. Advertisers who use Omnisend make a ROAS of 6,800% on average, and thanks to the tons of automations and templates which are ready to use within the platform. Omnisend connects to all major e-commerce platforms with just one click. If you want to migrate to Omnisend, they have live support available 24/7 to make the migration process a breeze. For customers that qualify for a customer success manager, personal onboarding and dedicated migration support are also available. Once you sign up with Omnisend, you get access to helpful blogs, webinars, case studies, and strategies to make you an expert in email marketing automation. And you'll get a complete free trial until you have 250 paying customers if you sign up using the link in the description below. Also, you'll get 30% discount for the first 3 months if you want to go with the paid plan using the link in the description below. Now, this leads us to our failure mode three, which is creative cannot survive the expansion. Now, don't confuse it with the first failure mode we discussed. This is very specific to creative. Now, I'll try to explain this with an example. In a lot of cases, what we have seen is when we have small budgets and we see one creative doing very good, and when we scale, actually this becomes one of the worst performers, and one creative that was not doing that great at lower spends does better at higher spends. This goes back to our first point. Initially, Meta reaches out to warm audiences. Now, you might have a creative that does superb with warm audience at lower budgets. When you scale, because Meta is now going to colder audiences, this specific creative, which was doing good in warm audience, is not doing good at cold audiences. You had at lower spends, you had this creative doing good. When you try to scale, you move this creative to the scaling campaign or you just increase the budget, pause the low performers, and then you are like, why this creative not doing great? Now, the symptoms for this is very easy to predict. When the frequency has started dropping, the CTR of this creative will drop a lot. The CVR will drop as well. And the CP, of course, will increase. So, the solution for this is pretty simple. You need to create a creative system. What I mean by creative system is you need to explore new angles. You need to understand new primary text that you think will be engaging for this particular persona and this particular avatar. A lot of people think that scaling is just vertical, which means we just increase the budget and we will scale. No, it's not like that. Scaling is horizontal as well. Because you were targeting one avatar and you thought the other avatar, the audience is very small. We don't need a specific creative for that. You basically broaden your creative spectrum, which will also help you to scale as well. You will never know which one basically generates good results for you. Even if it's a very tiny avatar within your audience which you never basically thought would be able to scale. Now, this leads us to our fourth failure mode, which is a lot common and it is unfortunate that it's disheartening for a lot of people because they think they cannot do anything about it. Which is your price tag does not have a headroom. If you're selling an ebook for $9, which is amazing. It's a great offer. It's a great offering. It's a great value for $9. And anybody who buys it, they love it. And your creative is good as well. However, you have to understand one thing. Price tag is very important because no matter how good your offer is, there is a CPM rate Meta will charge from you all the time. And the moment you scale, it means you have to go to more audiences, more premium placements. The CPM in most cases increases once you basically cross that sweet spot for your product or your you as an advertiser. And after that, the CPM prices become so much that no matter how high conversion rate you have, you cannot scale. So, at that point Meta will say, "Boss, we know your product is amazing, but leave we need $7 to sell your product at this scale. Okay, if you want to be within $200, $300 daily budget, no problem. We'll get you at $4, $5. But now you want to sell it thousands of times every day, then the prices will increase." Now, the symptoms for this is remember in the previous points we discussed that the frequency when the frequency basically reduces, the CTR decreases, and all. In this case, you will not see that. You'll just see CPA inflation. And you will see CPM inflation. You will see the CPM rates going up, and you'll see the CPM CPA rates going up. Otherwise, CTR it will not be impacted much, the conversion rate it will not be impacted much. Now, the solution for this is basically not within the ad manager. There's nothing you can do within the ad manager. What you have to fix is the economics of your product. You somehow have to increase the lifetime value or the product value. Now, I recently did a video with Carlos where basically he discussed that he was losing money until he was selling his product for $200, pre-recorded courses. And then what he he just made one change. He made it $19.99. He selected a niche audience, old people in America. Before that, he was targeting a lot of countries. And then he included like a one-to-one sessions in that as well. He made the product $2,000, and now he's profitable. He's at 4x ROAS probably. So, you have to come up with something. For example, our courses, we were not that profitable when we were individually advertising our courses. Then what we did is we created a bundle like a lifetime membership, increase the price tag of course, and now our ROAS has improved a lot. So, this is something you will have to do on the product side. This leads us to our fifth failure mode, which is you don't have incrementality. You basically have credit. Now, this one hurts the most because in this case, Meta looks profitable, but actually what's happening is there's no business impact from no matter how much you're spending on your Meta. Now, this happens a lot with businesses who already have organic or other sources apart from their digital advertising generating results for them, whether it's sales, leads, or whatever. For example, our website already generates a lot of score sales from our YouTube traffic or search volume, organic volume, organic search, and sources like that. Now, when we actually start Meta, there are two things that happen. One, there are people who come from other sources and Meta basically takes the attribution. Or, there are a lot of people who came to our website from organic sources or they know about us from YouTube and Meta just targeted them and they converted, but the actual work was done from our organic content and other sources. It's just Meta is just pushing the last few uh inches and calling it a day. If we have, let's say for example, 50 or 100 core sales every day coming organically from all the sources. When we start Meta ads campaign and Meta tells us, "Oh, I'm getting 15 conversions every day." Now, due to fluctuations, because even from other sources, our website some days has 85 core sales, some days 105, and this basically blends in and we start believing that, "Okay, Meta is getting 15 conversions every day." When we scale it to work the budget that we are getting 50 conversions from Meta, Meta starts breaking because it cannot take the attribution of so many 50% of the organic sources. And even if it takes, the problem is we will be able to see, "Okay, our lowest is 80. If Meta is saying 50, we're pushing budgets for 50 conversions, we should at least have 130. That's like the minimum we should have. Okay, 120, but we still hover between 80 to 100, and that's when we realize that Meta might or might not be taking the attributions, but there is no business impact, and that's when it starts to break. But, you have to understand that it's not that you are not able to scale. Actually, at lower scales, Meta was just taking attributions from your organic and other sources. It never worked, actually. So, there's no problem that we are not able to scale. You need to create a different strategy, basically. So, how to fix this? The first thing is whenever you are running Meta Ads and scaling, whenever you are trying to understand new creative, new strategies working or not, you have to rely a lot on the overall uplift. Okay, before running this campaign, what was the daily average sales, leads, whatever, and after running the campaign, what's happening? When we increase the budget, is there an uplift in the overall on the business impact side, from your CRM or whatever? That is the first thing you have to do. The second thing is you have to run incrementality test, which meaning that any new source, Meta or any Google Ads, anything you run, you have to basically see if the sales that if the platform is showing, is it incremental or not? There are a lot of ways to do incrementality test. For example, if you have dedicated technical account manager at Meta or Google, they can help you to set it up on your ad account. Or, on your side, you can do different kind of incrementality test, which is like the geo test or time-bound test. I have video on my channel explaining all that. It's super easy to do. Basically, the test should answer your question, did the overall increase after running ads on Meta or Google? This leads us to our failure mode six, which is your bidding controls are choking scale. A lot of advertisers set very tight guardrails, which means let's say they are profitable at 1.2x ROAS, and at lower scales, they have been getting 1.9 or two, they set it at 1.8, roughly. But, you have to remember, there's a rule of thumb. In 99% of the cases, every advertiser has a sweet spot, which is generally close to the lower budgets. After that sweet spot, when you increase your daily spends, in 99% of the cases, with the increase in budgets, the lower ROAS will dip. In some cases, small, let's say from 1.8 to 1.7, in some cases, harsh, 1.8 to 1.5. You have to understand that. And a lot of advertisers, what they try to do is they try to scale with just the budgets, or in certain cases, the creatives as well. They do everything that needs to be done, but they do not lower their ROAS targets, for example, and they think they'll they'll be able to get it at higher scales as well. It's exactly like the ebook we discussed. At after a point, no matter how good your product is, the efficiency will decrease. That happens in this case as well. And then they're wondering why this is happening. We analyze a lot of accounts, the creative is great, offer is great, landing page great. Everything is great, but they don't understand the fact that when you increase media spends, efficiency is decreased. That's how these algorithms are designed. That's how these behave. That's my experience from 13 years. Now, the symptoms for this, the first thing you will see is your daily budget is not spending. If you have a daily budget of $500, let's say, sometimes the campaign spends only 150, 200, or 250. You'll have limited delivery and very sudden volume drops. On one day, you will spend $600, on other day, you'll just spend $120. Your daily reach will kind of stay same, but your CPMs will drastically increase because delivery is constrained. Now, you might ask me, how do we fix this? See, treat bidding like a controlled experiment, not a safety blanket. If you're under-delivering, loosen the constraint, or test a less restrictive strategy temporarily. And you have to keep doing this until you're spending unprofitably. After that, you can put the guardrails back. One thing you have to understand is you cannot demand maximum volume and maximum efficiency at the same time. Scaling requires trade-offs. And that's all for this video. Thank you so much, and I will see you in the next one.