Why your Meta ads break when you scale them and what to do about it

Dahna Borg – Bright Red Marketing (00:00)

Advantage+ has sort of thrown a spanner in the works of how scaling can happen. Obviously, Meta is really pushing us to be in this Advantage+ ecosystem — one campaign, one audience, all of your ad creative in that one campaign. If budget does want to increase because everything’s going well, what happens if you touch it? You are now resetting everything. That one campaign that’s been driving all of your sales has a wobble. That’s really scary.

Dahna Borg – Bright Red Marketing (00:26)

Hi and welcome to the Bright Minds of eCommerce podcast. I’m Dahna, founder of Bright Red Marketing, and I created this podcast because I wanted to bring you the best advice from Australian experts in eCommerce and eCommerce store owners. If you’re wanting relatable stories and actionable advice, as well as the latest Facebook advertising strategies, you’re in the right place. So let’s get into today’s episode.

Dahna Borg – Bright Red Marketing (00:46)

Hi, and welcome to the Bright Minds of eCommerce podcast. Today you’re here with me. I’m Dahna from Bright Red Marketing. Most people think that scaling means spending more, and it doesn’t. Scaling means spending smarter and more strategically, and oftentimes means spending more. At the end of the day, Meta has made that significantly harder than it used to be with the way they are encouraging us to set up accounts. So what I want to do in this episode is talk to you about the difference between the two ways that you can scale your account, my favourite, and then what you can do with these changes.

So firstly, there are two different ways that you can scale your account: horizontal scaling and vertical scaling. Now, if you’ve ever heard me speak before, you will know that I am a massive fan of horizontal scaling. The challenge is that horizontal scaling works better with your traditional testing structures. So when we used to do audiences — interest-based audiences, lookalike audiences — we really got into that audience testing. We still see that working really successfully for a number of clients, but we are definitely seeing a push to these more Advantage+ style campaigns where we have a lot less leverage. Unfortunately, having a lot less leverage means horizontal scaling became so much more difficult, which means we are stuck, in theory, with vertical scaling.

The problem with vertical scaling is, as you will have noticed if you’ve ever run a Meta ad before, anytime you touch anything at the campaign or audience level, things tend to get a bit wobbly. So what I’m going to do today is talk you through those two different kinds of scaling and my suggestions on how you can safely and more securely scale whether you are in an Advantage+ or a more traditional structure.

First things first, horizontal scaling essentially means that you are adding more campaigns and ad sets targeting slightly different audiences, as a way to spend more strategically without touching the campaigns that are already working. The reason this is my preference, and always has been my preference, is because of that last part — the thing that’s already working doesn’t get touched. I’m not a massive believer in the learning phase that Meta likes to harp on about. I find that most of our clients never quite get out of the learning phase, but there is still something to be said that if you are in that learning phase or you make changes, it can get a bit wobbly. So we want to try and avoid that as much as possible.

With a horizontal campaign, we are essentially looking at: how can I add additional audiences that are similar but different enough to the existing audience to test, and increase the budget of the new audience or campaign so that we are scaling and increasing spend without affecting the thing that’s working? If interest-based audiences are working really well, can you try a different interest-based audience? If lookalike audiences are working really well, can you try a different lookalike audience? That’s your horizontal scaling.

Vertical scaling is what most people think of when they think of scaling — increasing the budget of something that’s already working. The challenge with this, as I’ve already said, is that anytime you change that, it does disrupt things. There are so many people out there that have a magic number. I’ve heard anything from 10% to 12.5% to 30% to $50 in terms of amounts you can adjust without it disrupting Meta. I have never found any of these to be particularly foolproof. There are always wobbles — well, often wobbles. Let’s go with often, because there are times when just adding more budget does really work. Usually that’s a perfect storm sort of situation. The timing is right, the creative is right, the audience is right — that magic perfect storm where you can quite rapidly scale. Those are magical, mystical times and we do love those.

But vertical scaling is very much that: add more budget, hope for the best. Moving forward and long term, the best scenario and the thing we’re going to find working best is a combination of these two.

Now, as I mentioned before, Advantage+ has sort of thrown a spanner in the works of how this scaling can happen. Obviously, Meta is really pushing us to be in this Advantage+ ecosystem — one campaign, one audience, all of your ad creative in that one campaign. You can see where I’m going with this. If that happens and your budget does want to increase because everything’s going well, what happens if you touch it? You are now resetting everything. That one campaign that’s been driving all of your sales has a wobble. That’s really scary — whether you’re spending $50 a day or $5,000 a day. That wobble is always terrifying.

So what I want to talk through today is some options for scaling if you are running Advantage+ campaigns that I think are a little bit safer and a little bit more stable. Before I get into that, let’s go through what I would suggest if you’re running manual campaigns.

If you’ve got a manual targeted campaign — lookalike, interest-based, those sorts of things — the thing I would have a look at is your interest stacks. If a particular interest-based audience is working really well, I want you to have a think about why that particular audience is working. I am still a firm believer in interest-based audiences because they work, but what’s not working is your old-school audiences. So if you’ve got an interest-based audience with two or three generic interests in there, that’s what I would call a really basic interest stack.

Maybe you sell yoga mats and you’re targeting people who go to the gym and who like yoga. The people engaging with your brand probably like yoga, probably go to the gym — cool, we knew that. What we don’t know is what Facebook doesn’t know. Meta’s probably got a pretty good idea that the people interested in your brand like yoga, but your brand hopefully is a little more complicated than that. There’s a particular niche or window in that yoga mat space that’s a bit more interesting. Maybe your yoga mat is made from some fancy new material — maybe it’s vegan.

So then what you’re going to do is go, okay, these people also have an interest in a vegan eating lifestyle. Maybe your yoga mats have book quotes on them, so you want to target people who like yoga but are also in the bookish space. Maybe you’ve got a yoga mat that’s for travellers or photographers. We call these our weird and wacky audiences. It’s not just that my audience likes my product because they like the thing I’m selling — it’s specifically why my product. I think this is where a lot of businesses struggle because they just sell a yoga mat. You’ve got to really think about what your business does, who it’s for, and what situations they’re in.

We do this a lot with our fashion clients. We’re not targeting the people buying things from Shein and Temu. We’re going after high-end brands. So it’s not as simple as targeting other high-end brands — it’s about asking, what do people who have the money to buy your luxury dresses like? Do they like travelling? Are they having long wine lunches? What else is going on in their world? Because Facebook is really good at looking at your website and going, cool, she sells dresses. What it doesn’t necessarily do as well is know the exact demographic you’re going after. Facebook will go, cool, fashion, done. There might not be much point running an interest-based audience that just says “fashion,” but there might be a point to going, okay, people who are into fashion but maybe they’re also into interior design — really playing with that.

So if you are running manual campaigns, trying some different interest stacks is where you’re going to see the most success. Running just a couple of interests and throwing them in one ad set is not the way it works anymore. What you’re trying to do is give Facebook information it doesn’t already have. That’s where a well-constructed interest-based audience is still working quite successfully for many brands.

Obviously, if your lookalike audiences are working really well, looking at different lookalike audiences, separating out your cold and warm audiences, and adding exclusions can also be a really good way of adding additional audiences to use for your scaling.

So now that we’ve talked about that, this is where we get into the good stuff — how you can actually start to scale your Advantage+ and take back some of that control. The point of an Advantage+ campaign is to give all of that control to Meta. You’re saying: this is my pixel, this is my website, this is my creative — use that creative as your targeting. Now, if you are a really big brand, Meta is going to have a much better idea of what a potential customer looks like for you. But those are really, really big brands. And if you’re listening to this podcast, you’re probably not the CEO of Nike. If you are the CEO of Nike, hi, welcome — it’s nice to have you here. But most of you are going to be running smaller brands. We’re building businesses for our families, building businesses for our lifestyles, and we’ve got good goals, but we’re not trying to take over the world.

When you’re that kind of business, Meta doesn’t have the level and calibre of data it has for someone like Nike. So we need to rely on Meta to give us that information, but it doesn’t always have the best information. That’s where we go, okay, let’s play with this a little bit and see what we can do. A straight-up Advantage+ sometimes works really well, sometimes it does some things we don’t necessarily like, and sometimes it’s not spending the money where we want it — which is where the scaling conversation comes in.

So if you’ve got a traditional Advantage+ set-up — one campaign, one audience, a whole bunch of ads — and you’re thinking, this is going amazingly, I want to scale this — my suggestion is: don’t touch that one. Let’s expand. Let’s filter. Let’s see what we can do instead.

The easiest ones to start with are age and gender splits, and country and region splits. What I mean by that is if you sell a product to multiple age ranges or two different genders, actually separate that campaign out into the segment that’s not getting as much spend. For instance, if you sell to both men and women and you’ve got your Advantage+ running, and you notice it’s only spending 10% of your budget on men but you actually have a decent line of men’s products and you make good money on that — leave that first campaign alone, duplicate it, and make this new one specifically for men. Add in the gender filter, make sure all of the creative specifically targets men. Show off the men’s products, use male models, really heavily focus it.

What that allows you to do is: the campaign that’s doing really well doesn’t get touched, but you’re taking that untapped market and saying, look, I know Meta knows this exists, but we’re going to do this really well and do it much better. So we’re putting more budget behind it and we’re going to do the creative very differently. You’ve now been able to scale in a sustainable way without breaking the thing that’s already bringing in all the money.

The same thing applies for age. If you’ve got a product range with a wide age spread — maybe some products that skew younger, some that skew older, or different aged models — you can start to pull these out. When you look at your breakdown data, you might see that your Advantage+ campaign is spending really heavily on an older demographic, but you really want to grow that younger demographic base as well. If the younger audience isn’t getting the spend you’d like, pull that one out and build a dedicated campaign: here are all of the youngest models, here are the products the younger demographics are buying.

Your country and region splits are also really interesting. If you are scaling internationally or can sell to different countries, I highly recommend that each country has its own separate Advantage+. Even just across Western countries, the language is different — American spelling, Australian spelling — and people respond best when content is in their own spelling and style. But also, the way different countries respond to marketing is wildly different. We have a client that targets both Australia and the US and the way their funnel content works is completely flipped between the two.

In Australia, their UGC does wildly better at the top of funnel for a cold audience, whereas static graphics perform better in the middle of funnel. In America, it’s completely flipped. That exact breakdown might not be universal, but the concept holds. Being able to split those countries out in your Advantage+ campaigns gives you that control back — control of where they’re seeing content, what kind of content, spelling, tone. Americans are generally more receptive to aggressive marketing, whereas Australians need you to be a lot more chill, a little more casual, and have a bit more fun with it. So the creative is going to be a little bit different country to country, and having that country split gives you the ability to set up a stable base.

That is the thing we are trying to do with any horizontal scaling — set up a stable base. We don’t want to have all of our eggs in literally one campaign basket. We talk about not having all of your eggs in the Meta basket. We really don’t want to have all of our eggs in one Meta basket that relies on one campaign. If that one campaign has any problems, if you try to scale it and everything falls apart, you’re in big trouble.

One campaign is not a stable base. Two campaigns is not a particularly stable base. We don’t start getting stable until we get to three or four. But because Meta is pushing us towards Advantage+, it’s getting harder to navigate that. So being able to create some of these different splits starts to give you a little more stability, so that you can have those campaigns scale as much as they can, and then start to vertically scale some of them without the whole thing crashing if one has a wobble or doesn’t perform at the new budget.

The other really interesting option is a product line split. If you’ve got different products or ranges on your website — and this won’t apply to everyone — you can start splitting those out. This doesn’t particularly work in fashion the way you might think. We don’t want a campaign on blouses, a different one on skirts, and a different one on trousers. But you could split by collection. If you sold handbags and luggage, for example, those are two different things you’re going to market differently. It can be really interesting to split those product lines out. The key thing to be careful of here is that you have enough creative to back it up — you don’t want to split things out and not have enough unique creative to promote each individual line.

I really like a product line split. I think it’s one of the safest ways of scaling, especially if you’ve got diverse product ranges and enough content for each of them. It can be a really great way to grow your business safely. And if a product line sells out or goes seasonal, you’re not turning everything off — you can start to play with the pieces of the puzzle. I really, really like that one.

The other one worth mentioning is something a lot of people are talking about at the moment — separating by avatars. Now, I have historically not been a massive fan of avatars because I find that most people who talk about them kind of miss the boat. So I’m going to be really clear: if you’re going to test by avatars, they need to be wildly different people. This is not “Susie likes yoga on Sundays and Laura likes Pilates on Tuesdays.” Those are not two different avatars. We need something like John who does yoga after knocking off from his trade business versus Susie who goes to Pilates on Tuesdays — two wildly different people. If you are going to test by avatar, keep that in mind. It has to be wildly different people. Similar to the product line split, you also need enough creative to back it up and content to sustain it regularly. But I do really like where that’s going and I think it’s going to be applicable for bigger brands as they’re scaling.

Just to clarify, the different ways of scaling your Advantage+ horizontally are: age and gender splits, country splits, product line splits, and avatar splits — though be careful with that last one. These are some really nice, healthy ways to start scaling your Advantage+ ecosystem horizontally, which gives you a really solid foundation to start increasing vertically.

Now, if you start to do this, Meta will start to get a bit cranky with you. You are breaking the rules. You’re not playing their game. They’ll start talking to you about things like budget fragmentation and competing against yourself. But what I want you to know is that Facebook is trying to get you to simplify things as much as possible. You know your business, you know your results, you know what your targets are. If you do this and you’re getting results, don’t worry about it. If you do this and you’re not getting results, that’s when you use that information, make some decisions, do some testing, and try something else.

I’m not saying the only way to scale is to add product line splits, and if it doesn’t work, that’s it, there’s nothing more to talk about. Everything I ever do and everything I ever talk about is testing. Test it, see what works, test something else, see what works. We’re always testing to find what’s working best for your business.

Now, I know you’re going to ask about budget. My suggestion and rule of thumb here is that the budget for your new horizontal campaign should be equal to or slightly more than the current campaign. So if you’re spending $100 a day on your general Advantage+ campaign and you want to try a horizontal scale into a different country, a gender split, a product split, or an avatar split — start it at at least the same, if not a little bit more. The idea is to scale. If you’re spending $100 a day and it’s going beautifully, don’t start the next one at $20 a day. That’s not the point. The point is to grow the business and scale. If you do that first test and it’s working really well but the frequency is high, then you can start to pull it back a little bit. But generally speaking, when we’re having these conversations it’s because things are going really well and we want to scale. The only caveat I’ll make to that is: scaling is for when things are going really well. We don’t want to start spending more money if things are not working. Keep that in mind.

When it comes to vertically scaling, there’s no real rule of thumb. You’ll hear lots of people say 10% here, 12% there. We tend to move things in small increments, depending on how much the account is spending. Give it a couple of days, let it wobble, see what it’s doing. Once the wobble settles, do it again. There is no hard and fast rule. Anyone who tells you there is — I don’t believe them. I’m sorry. I’ve been doing this a very long time. Sometimes there’s a wobble, sometimes there’s not. Sometimes the wobble lasts a day, sometimes it lasts three weeks. Sometimes the wobble is irreconcilable and we need to pull it all back. So keep in mind what’s happening in your account, don’t worry about everyone else’s, take money you can afford to spend, make sure it’s profitable for you, and you’ll be happy days.

At the end of the day, Meta is trying to take back so much of that control, and our job as advertisers and business owners is to find the places where control is still important, still makes a difference, and gives us something back. There’s no point trying to take control from things that are irrelevant or don’t actually help us, but scaling safely and sustainably is definitely something worth taking some of that control back for. So happy scaling, and I hope you learned something.

Dahna Borg – Bright Red Marketing (21:32)

Thank you for listening to the Bright Minds of eCommerce podcast. As always, you can find the show notes on our website at brightredmarketing.com.au. Just look for the podcast page. Thanks for listening.