Solved: How To Increase ROAS From x2 to x4 on Meta Ads?

Hi there,

Thanks for reaching out! I had a look over the situation you described. Being stuck just below your break-even ROAS is a really common and frustrating place to be, so you're not alone in this. I've seen it with quite a few eCommerce brands we've worked with. You build something that clearly has potential, people are buying, but making the ads properly profitable is the final hurdle. It's often just a few key tweaks away from tipping over into real growth.

Based on what you've said, I'm happy to give you some initial thoughts and guidance from my experience. This is all based on the hundreds of campaigns we've run for businesses, many of them in the eCommerce space just like yours. My advice is always brutally honest and to the point, no fluff. The goal is to get that ROAS from x2 up to where it needs to be, ideally hitting that x4 mark.

We'll need to look at your overall campaign structure...

Right, first things first. Your current setup – one CBO campaign with two ad sets for prospecting – is a decent starting point. It's simple, it gets the ads running, and it's brought you sales. But to get to the next level of profitability, that structure just isn't sophisticated enough. It's like trying to win a Grand Prix with a standard road car. It'll get you around the track, but you won't be setting any records. To really start squeezing more revenue out of every pound or dollar you spend, you need a structure that treats different types of customers differently.

What I mean by that is a funnel-based approach. This is standard practice for how we build out accounts for our clients, including eCommerce ones that have seen huge returns. I remember working with a women's apparel brand, for example, where we implemented this strategy and achieved a 691% return on ad spend using Meta ads. The core idea is simple: not everyone who sees your ad is at the same stage of their buying journey. Someone who has never heard of your brand needs a very different message to someone who was just on your site and added a hoodie to their cart before getting distracted.

Your current campaign is treating everyone the same. It's all Top of Funnel (ToFu), or 'prospecting'. You're just trying to find new people out there in the big wide world of the US who might be interested. This is the most expensive and difficult part of advertising. Where you're missing a massive, massive opportunity is with the Middle and Bottom of the Funnel (MoFu and BoFu).

Here’s a quick breakdown:

-> Top of Funnel (ToFu): This is your cold audience. People who don't know you exist. Your goal here is introduction and interest. This is what your current campaign is doing. You show them your cool designs and hope they click.

-> Middle of Funnel (MoFu): This is your warm audience. These are people who've shown some interest. They've visited your website, watched one of your videos, engaged with your Instagram page, but they haven't bought anything or even added a product to their cart. They know who you are. The goal here isn't to introduce the brand, it's to re-engage them and pull them deeper into your world. They need a nudge, not a full-on sales pitch from scratch.

-> Bottom of Funnel (BoFu): This is your hot audience. These are the people who are on the verge of buying. They've added items to their cart, they've initiated the checkout process, maybe they even added their payment info. These are the hottest leads you have, and converting them should be your absolute top priority. They are so close to giving you money.

So, instead of one big CBO campaign, I'd build it out into three separate campaigns, one for each stage of the funnel. Why? Because it lets you control the budget and the messaging precisely. You might find that you can spend just £10 a day on your BoFu campaign and it brings in a 15x ROAS, because you're just converting people who were already 99% of the way there. That BoFu campaign then becomes your profit engine, and it helps to fund the more expensive ToFu campaign that's out there finding new customers. Your MoFu campaign acts as the bridge, stopping the leads you've already paid for from going cold. This seperation is one of the most fundamental things we do to scale accounts. Without it, you're just throwing all your messaging into one pot and hoping for the best, which is exactly why you're seeing an 'average' ROAS of around 2.

I'd say you need to completely rethink your targeting...

This is probably the single biggest lever you have to pull right now. You've mentioned that you tried narrow interest targeting and it didn't work, and that broad is doing... okay. When I audit Meta accounts, this is the most common issue I see. People either go way too broad without a mature enough pixel, or they try 'narrow' targeting but pick the wrong interests entirely. It's less about narrow vs broad and more about relevant vs irrelevant.

Let's take your situation. A clothing brand for a specific US state. Your finding that people outside the state buy it is a brilliant insight – it's about state pride and identity, not just location. But targeting the entire US with 'broad' is asking the Meta algorithm to do a huge amount of heavy lifting. It has to sift through 200 million+ people to find the small pocket that has an affinity for, let's say, Ohio, and also wants to buy a t-shirt about it. Your pixel has been running for six months, which is good, but to double your ROAS you need to help it out.

So, here’s how I would approach your targeting, broken down by the funnel we just talked about.

For your new ToFu (Prospecting) Campaign:

This is where you'll test different cold audiences to find pockets of customers. Forget broad for a minute. We need to build some audiences that are packed with your ideal customer. You need to get into their heads. Who is this person who lives in California but still has enough pride for their home state of Wisconsin to buy a hoodie? What else do they like?

I usually group interests into themes to keep the testing clean. You'd have a separate ad set for each theme. For example, if your state was Texas:

• Theme 1: Pro Sports Fans. Interests: Dallas Cowboys, Houston Texans, San Antonio Spurs, Texas Rangers. People who follow these teams have a strong state identity.
• Theme 2: College/Alumni Pride. Interests: University of Texas at Austin, Texas A&M University, Baylor University. Another incredibly strong indicator of state affinity.
• Theme 3: Cultural/Lifestyle Interests. Interests: Whataburger (the fast food chain), Texas Monthly (magazine), George Strait (musician), BBQ, Country Music. These are cultural touchstones.
• Theme 4: Competitor Audiences. Interests: Other state-pride brands (if they're big enough to be an interest), Buc-ee's (the gas station), YETI coolers (a famous Texan brand).

The key here is specificity. The reason your previous narrow targeting might have failed is you might have targeted something like "T-Shirts". That's a terrible interest. It includes everyone who might like t-shirts, not people who would like your t-shirts. The interests must be a proxy for your specific customer persona. By testing these themes in different ad sets, you'll quickly see if one group (e.g. sports fans) is giving you a 4x ROAS while another is at 1.5x. Then you kill the loser and put more budget behind the winner. This is active optimisation.

Once you have more data, you move onto Lookalike Audiences. You've been running for six months, you should have enough data for this now. This is where you can really find gold. You're essentially telling Meta: "Here is a list of my best customers. Go and find me a million more people who look just like them". You need at least 100 people in your source audience, but the more the better. I would prioritise them like this:

1. Lookalike of Highest LTV Customers: If you can, export a list of your repeat buyers or highest spenders and upload it to Meta. A lookalike of this group is pure gold.
2. Lookalike of all Purchasers: Your entire customer list. A very strong signal.
3. Lookalike of Initiated Checkouts: People who were close to buying. Still a good signal.
4. Lookalike of Adds to Cart: A bit broader, but still valuable.

You should test these in seperate ad sets within your ToFu campaign. Start with a 1% lookalike in the US. This will be the audience most similar to your source. Then you can test a 1-3% or 3-5% lookalike, which will be larger but less precise. Again, you're looking for winning pockets of performance.

For your new MoFu & BoFu (Retargeting) Campaigns:

This is the untapped potential. This is where you'll get your highest ROAS and make your business consistently profitable. You are paying to get people to your site with your ToFu campaign; it is a travesty to just let them leave and never speak to them again. Setting up retargeting is non-negotiable.

MoFu (Warm Audience) Campaign:

The goal here is re-engagement. The audience for this campaign would be:

• All Website Visitors (in the last 30-60 days) - EXCLUDING anyone who purchased or added to cart.
• Instagram Engagers (last 90 days) - EXCLUDING purchasers.
• Facebook Page Engagers (last 90 days) - EXCLUDING purchasers.
• Video Viewers (50%+) (last 90 days) - EXCLUDING purchasers.

You can group these into one ad set to start with if your traffic is low. The ads you show this group should be different. Don't just show them the same prospecting ad. You could show them testimonials from happy customers, user-generated photos of people wearing your gear, or a carousel of your best-selling products. The message is: "Hey, remember us? We've got some cool stuff other people from [Your State] are loving."

BoFu (Hot Audience) Campaign:

This campaign will almost certainly have the highest ROAS. These people are ready to buy. Your audience here is:

• Added to Cart (in the last 7-14 days) - EXCLUDING purchasers.
• Initiated Checkout (in the last 7-14 days) - EXCLUDING purchasers.

The ad for this audience needs to be a direct punch. You should be using Dynamic Product Ads (DPA). This will automatically show them the exact products they left in their cart. The copy can be simple and urgent: "Did you forget something?" or "Still thinking it over? Your items are waiting". You can test adding a small incentive like "Complete your order now and get 10% off" or just reminding them "Free shipping is included!". This is about closing the deal. A well-run BoFu campaign can easily hit 10x+ ROAS and will fundamentally change the economics of your entire advertising operation.

You probably should get more aggressive with creative testing...

Now, let's talk about the ads themselves. You're asking the right questions here. Relying solely on catalog ads, especially for prospecting, is leaving a lot of performance on the table. Catalog ads are fantastic for BoFu retargeting, but for cold audiences (ToFu), they can be a bit... boring. They lack story and personality.

So to answer your questions:

"Should I add ads with custom creatives to compete alongside my catalog ads?"

Yes. 1000% yes. For your ToFu campaign, you should be testing a range of creatives. This is how you stop people scrolling. For a clothing brand, the best creative you can possibly have is User-Generated Content (UGC). This means photos and videos of real, normal customers wearing your clothes and looking happy. It's infinitely more powerful and trustworthy than a simple product mockup. Reach out to your past customers and offer them a discount on their next order in exchange for a photo or short video. People love to be featured.

You should also be testing other formats:

• Video Ads: A simple video showing off the quality of a hoodie, the detail in the embroidery on a hat. A person turning around in it, showing the fit. This answers so many questions a static image can't. We've seen UGC videos work wonders for SaaS clients, and for eCommerce it's even more direct and powerful.
• Lifestyle Images: Not just the product on a white background. Show the t-shirt being worn at a football tailgate. Show the hat on someone fishing at a lake in your state. Sell the vibe and the identity, not just the peice of cotton.
• Carousel Ads: But used creatively. Instead of just 5 products, make the first card a lifestyle image or a customer review graphic, then follow it with the relevant products. Tell a mini-story.

"Should I try multiple different catalog ads with different primary texts and run them against eachother?"

Yes. This is basic A/B testing and you should always be doing it. You've got a great product angle with state pride. Your copy should reflect that. Test different angles against each other in the same ad set.

• Angle 1: Nostalgia. "No matter where you are, a part of you is always back home in [State]. Wear your pride."
• Angle 2: The Perfect Gift. "Know someone who's missing [State]? The perfect gift for the homesick friend in your life."
• Angle 3: Quality/Features. "Super-soft hoodies and premium embroidered hats, designed for true [State] fans. Free shipping on all orders."
• Angle 4: Social Proof. "Join thousands of [State] natives and expats showing off their pride. See why our customers give us 5 stars."

You will be suprised at how a different headline or primary text can drastically change performance. Never assume you know which one will work best. Always test.

"Should I group multiple types of prodcuts together (hoodies, hats, T-Shirts) and run them in a single catalog ad?"

Yes, this is another great idea to test. These are called 'Product Sets'. You can create sets for 'Bestsellers', 'New Arrivals', 'Hats Collection', 'Hoodie Collection', or even themed sets like 'Game Day Gear' which includes hats and t-shirts. Test these different product sets against each other in your ToFu and MoFu campaigns to see which combination resonates most with new audiences.

The main takeaway for creative is to move beyond the basic Shopify catalog feed and start thinking like a proper brand. You need to create assets that connect emotionally with your audience and build trust. That's what gets people to stop, click, and ultimately buy.

You'll need to think about your offer and landing page...

Finally, it's important to remember that ads are only one half of the equation. Your ads can be perfect, but if they send people to a website that doesn't convert, you'll still have a low ROAS. You said you feel good about your landing pages, which is great, but in the quest to get from 2x to 4x ROAS, every single percentage point of conversion rate matters. A tiny improvement on your site can have a huge impact on profitability.

The Offer:

You mentioned your prices are slightly high but come with free shipping. This is a classic eCommerce conundrum. Have you ever tested the opposite? Lowering the product price and charging for shipping. For some customer psychologies, a £25 t-shirt + £5 shipping feels cheaper than a £30 t-shirt with free shipping, even though it's the same. It's a very simple test to run for a week or two and see what it does to your conversion rate and AOV (Average Order Value).

You should also consider other offers to increase AOV, which directly increases ROAS. For example:

• Bundles: "Buy any hoodie and get a hat for 30% off". This encourages people to buy more in one go.
• Tiered Discounts: "Spend £50 get 10% off, spend £100 get 20% off".
• First-Time Buyer Pop-up: An exit-intent pop-up offering 10% off their first order in exchange for their email can capture sales you would have otherwise lost.

The Store Itself:

Since this is a dropshipped business, trust is everything. People are wary of new online stores. Your website needs to scream "we are a legit, high-quality brand". Without seeing it, here are the things that matter most:

• Social Proof is King: Are you plastering customer reviews and ratings on your product pages? Do you have a gallery of customer photos (UGC) on your homepage? This is the single most effective way to build trust.
• High-Quality Product Photography: Mockups are okay to start, but your ROAS will climb if you can get photos of your clothing on real people. It makes the products tangible and desirable. Even if you just get a friend to model for an afternoon, it will be better than a generic mockup.
• Compelling Product Descriptions: Don't just list the features. Sell the feeling. Talk about the "unbelievably soft fleece" of the hoodie, or the "perfect fit of our classic dad hat". Re-iterate the state pride connection in the descriptions.
• About Us Page: Tell a story! Why did you start this brand? What does the state mean to you? People connect with stories, not faceless dropshipping stores.

Improving your website's conversion rate by even just 0.5% (say from 1.5% to 2.0%) is a 33% increase in sales from the same amount of traffic. That alone could be the difference between your 2x and 2.6x ROAS.

I know this is a lot to take in, but moving from breaking even to being really profitable requires this level of detail. It's about building a proper advertising machine, not just running a few ads. I've detailed my main recommendations for you in a table below to make it clearer.