Hi there,
Thanks for reaching out!
Happy to give you some initial thoughts on your campaign structure. It's a common question, and the simple answer is no, you shouldn't duplicate the entire campaign. But honestly, that's just scratching the surface. The real issue, and what separates campaigns that burn cash from those that actually turn a profit, is building a proper testing framework from the ground up. Most people new to ads try to test everything at once and end up learning nothing. We're going to avoid that and set up a system that gives you clear winners you can actually scale.
TLDR;
- Don't duplicate your campaign to test landing pages. It creates audience overlap and splits your budget, making it harder for the algorithm to learn. You should test LPs at the ad level, within a single ad set.
- Your proposed structure is trying to test too many things at once (creatives and LPs across three different audience types). This is inefficient and expensive. We need to simplify and isolate variables.
- The most important piece of advice is to rethink your audience strategy. Don't treat Warm, Lookalike, and Cold audiences the same. Start with your warm (BoFu) audience to find a winning combination first, as they are most likely to convert.
- Profitable advertising isn't about finding one magic bullet, it's about a systematic process of validating your offer, creative, and landing page with your highest-intent audience before you try to scale with colder traffic.
- This letter includes a flowchart to visualise a better testing structure and an interactive ROAS calculator to help you understand your numbers.
We'll need to look at your campaign structure...
Right, let's get the immediate question out of the way. Duplicating the entire campaign is a bad idea, you're right to be concerned. When you run two campaigns with identical (or very similar) targeting, they end up in the same auction, bidding against each other. This can drive up your costs and make your results messy and unreliable. Meta's system is designed to learn, and when you split its learning between two identical campaigns, you're just making its job harder and more expensive for you. It's like asking two different people to build the same house without talking to each other – you'll get wasted effort and a rubbish outcome.
The 'textbook' way to test two landing pages is within a single campaign. You'd have your ad set (e.g., your Warm audience), and inside that ad set, you'd create ads for each combination. So, for every one of your four creatives, you'd have two ads:
- -> Creative 1 -> Landing Page A
- -> Creative 1 -> Landing Page B
- -> Creative 2 -> Landing Page A
- -> Creative 2 -> Landing Page B
...and so on. This keeps everything in one ad set, under one budget, allowing the algorithm to optimise delivery towards the best-performing combination of creative and landing page.
But here’s the contrarian view, based on what we see in countless accounts we audit: this is still a terrible idea for someone just starting out. You're still trying to test too many things. With 4 creatives and 2 landing pages, you have 8 different ad variations in each ad set. Across three ad sets, that's 24 ads. Unless you have a massive budget, you're spreading your spend so thin that no single ad will get enough data to give you a statistically significant winner. You'll end up turning off ads based on gut feeling rather than data, which is just gambling.
The goal isn't to test everything at once. It's to find a single winning combination and then scale it. We need a simpler, more methodical approach. Here’s a look at a better way to think about structuring your tests.
Your Initial Idea (Inefficient)
(Targeting Warm, LAL, Cold) -> LP A
(Targeting Warm, LAL, Cold) -> LP B
The 'Textbook' Method (Still Flawed)
(Creative 1/LP A, Creative 1/LP B, etc.)
A More Profitable Approach
I'd say you need to rethink your audience prioritisation...
This brings me to the much bigger issue. The fundamental flaw in your plan isn't the landing page test; it's treating your Warm, Lookalike, and Cold audiences as equals. They are worlds apart in terms of intent, and lumping them into one testing strategy is a recipe for confusing data and wasted money. As a consultant, this is one of the most common mistakes I see – people get excited about reaching new customers and forget to properly monetise the ones who already know them.
You need to think like a funnel. I've laid out a typical audience prioritisation below. The idea is to work from the bottom of the funnel (BoFu) upwards. These are your hottest prospects, the low-hanging fruit. Your goal is to find a profitable combination of ad and landing page here first, with a small budget, before you even think about scaling to colder, less predictable audiences.
BoFu (Bottom of Funnel) - Your Starting Point
These are your 'Warm' audiences. People who have visited your website, added a product to their cart, initiated checkout, or engaged with your page. They know who you are. They've shown interest. They are, by far, the most likely group to convert.
-> Audiences: Added to cart (last 30 days), Initiated Checkout (last 30 days), Website Visitors (last 30 days).
-> Your Goal Here: Get them back to finish the purchase. Remind them what they were looking at. This should be your most profitable ad set, hands down. You validate your creative and offer here first.
MoFu (Middle of Funnel) - The Next Step
These are people who have shown some interest but aren't as committed as your BoFu audience. They might have watched a percentage of your videos or engaged with an Instagram post. They're aware of you but haven't taken that crucial step of visiting your site.
-> Audiences: 50% Video Viewers, Instagram Engagers, Facebook Page Engagers.
-> Your Goal Here: Nudge them towards your website. Show them more about your product's benefits and build more trust.
ToFu (Top of Funnel) - For Scaling Only
This is where your Lookalikes and Cold interest-based audiences live. These people have likely never heard of you. This is the most expensive and difficult part of advertising. You should only pour significant budget here *after* you have a proven, profitable system running with your BoFu audiences.
-> Audiences: Lookalikes of your best customers (e.g., a 1% Lookalike of people who have purchased), and detailed targeting (interests, behaviours).
-> Your Goal Here: Introduce your brand and solution to a new audience, using the ad creative and landing page that you already proved works on your warm traffic.
Too many advertisers start at the top and wonder why their costs are so high. It's because they're trying to convince complete strangers. Start by converting the people who are already halfway there. It’s cheaper, faster, and it’s how you build the foundation for a scalable campaign.
You probably should focus on the offer first...
Now, let's talk about those landing pages. The fact you're testing two is good – it shows you're thinking about optimisation. But the biggest lever you can pull, far more important than the colour of a button, is the offer itself. A great offer on a decent landing page will always beat a weak offer on a perfectly designed one. I've seen so many campaigns fail not because of targeting or creative, but because what they were asking people to do just wasn't compelling enough.
Before you get lost in testing landing page layouts, ask yourself: is the core offer on both pages fundamentally strong? Does it solve an urgent, expensive, and frustrating problem for your target audience? If you sell handcrafted jewellery, for example, the problem isn't just "I need a necklace". It might be "I need a unique, meaningful gift for an anniversary that shows I put thought into it, and I need it to feel special, not mass-produced." Your landing page needs to speak to *that* pain.
One powerful copywriting framework we use for this is Before-After-Bridge.
- Before: Describe their current world, full of the problem. "Tired of generic gifts that get forgotten in a drawer?"
- After: Paint a picture of the world after they use your product. "Imagine their face when they unwrap a one-of-a-kind piece, handcrafted just for them, a story they can wear every day."
- Bridge: Position your product as the bridge to get them there. "Our Celestial Collection is that bridge. Each piece is unique, telling a story of craftsmanship and care."
This is so much more powerful than just listing features. You’re selling a transformation, not a product. Make sure both your landing pages have this kind of persuasive, problem-focused copy. If they don't, no amount of A/B testing will save them. We've seen first-hand how a powerful offer can drive incredible results; I remember one eCommerce campaign for a subscription box client where we achieved a 1000% Return On Ad Spend on Meta Ads.
You'll need a solid testing methodology...
Okay, let's put this all together into an actionable plan. This is a simplified version of the methodology we use for new clients. It's designed to be capital-efficient and to give you clear, unambiguous data to build upon. Forget testing everything at once. We're going to build this machine piece by piece.
Phase 1: Validate Your Creative (Budget: ~20% of total)
- Campaign: Create one new campaign with the "Sales" objective. Call it "[PROD] - Retargeting - CBO". Use Campaign Budget Optimisation (CBO).
- Ad Set: Inside this campaign, create one ad set. Call it "BoFu - Website Visitors 30d". Target people who have visited your website in the last 30 days, but exclude purchasers. This is your highest-intent audience.
- Landing Page: Pick ONE of your two landing pages. Your best guess. We need to keep variables to a minimum for now.
- Ads: Create your four different ads, all pointing to that single landing page.
- Run & Analyse: Let this run until you've got meaningful data (e.g., at least 1,000 impressions per ad, or until one ad clearly outperforms the others on metrics like CTR and Cost Per Purchase). Your goal here is simple: find your single best-performing creative.
Phase 2: Validate Your Landing Page (Budget: ~10% of total)
- Pause Losing Ads: Go back into your BoFu ad set and turn off the three losing creatives from Phase 1.
- Duplicate the Winner: Take your single winning creative and duplicate it.
- Change the URL: In the duplicated ad, change the URL to point to your second landing page. Now you have two ads running: Winning Creative -> LP A, and Winning Creative -> LP B.
- Run & Analyse: Because you're only testing one variable (the landing page), it will be very clear which one converts better. You're looking for the highest conversion rate and lowest Cost Per Purchase. You've now found your winning combination: one creative and one landing page.
Phase 3: Scale to Colder Audiences (Budget: ~70% of total)
- New Campaign: Create a second campaign. Call it "[PROD] - Prospecting - CBO".
- Ad Sets: Inside this campaign, create two or three ad sets for your cold audiences. For example:
- Ad Set 1: "LAL 1% - Purchasers" (a lookalike of your past buyers).
- Ad Set 2: "Interests - Competitors" (targeting people interested in your direct competitors).
- Ads: In each of these new ad sets, create ONE ad using your winning creative and winning landing page from the previous phases.
- Run & Scale: Let CBO distribute the budget. Because you're using a combination you've already proven works with a warm audience, your chance of success with colder traffic is dramatically higher. You're not guessing anymore; you're scaling a known winner.
This methodical process removes the guesswork. It ensures you're spending the majority of your budget on what works, not on hopeful experiments. Understanding your numbers is paramount here. Use the calculator below to get a feel for how different metrics impact your bottom line.
I've detailed my main recommendations for you below:
I know that's a lot to take in, so here’s a summary table of the actionable strategy we've just walked through. This is the exact kind of foundational plan we'd implement to take the guesswork out of a new ad account and build it for profitability.
| Phase | Action | Rationale |
|---|---|---|
| Phase 1: Creative Validation | Run your 4 creatives to one landing page within a single ad set targeting your warmest BoFu audience (e.g., website visitors). | Isolates the creative as the only variable. Uses your highest-intent (and cheapest to convert) audience to find a winning ad quickly and efficiently. |
| Phase 2: Landing Page Validation | Take your single winning creative from Phase 1 and test it against your two different landing pages. | Now you're only testing the landing page. This gives you a clear, unambiguous winner and establishes your single best ad creative + landing page combination. |
| Phase 3: Scaling | Create a new prospecting campaign. Use your winning combination from Phase 2 to target your Cold and Lookalike audiences in seperate ad sets. | You are no longer guessing what might work on cold traffic. You are scaling a combination that has already been proven to convert, dramatically increasing your probability of success and reducing wasted ad spend. |
| Ongoing | Continually monitor performance. Once you have a stable, profitable system, you can start introducing new creative and audience tests systematically. | Advertising is never "set and forget". This framework provides a baseline from which you can run structured experiments to constantly improve performance over time. |
This structured approach is what truly separates professional media buying from the amateur "boost post" mentality. It might feel slower at the start, but it builds a solid, data-driven foundation that prevents you from burning through your budget on flawed assumptions. It's how you build a campaign that lasts.
And this is really just the begining. We haven't even touched on advanced strategies like audience layering, analysing your placement performance, crafting specific ad copy for different stages of the funnel, or setting up a robust tracking and reporting system. Each of those is a deep topic in its own right and can add another layer of profitability to your campaigns.
Running ads successfully is a skill, and it requires a level of expertise and experience to navigate properly. It's this deep experience that allowed us to, for instance, take a client's Cost Per Acquisition from £100 down to just £7, or generate $115k in revenue for a course creator in just six weeks. It comes from knowing which levers to pull and, just as importantly, in what order.
If you'd like to go over this strategy in the context of your specific business and ad account, we offer a completely free, no-obligation initial consultation. We can hop on a call, share screens, and I can give you some bespoke advice on how to best implement this kind of framework. It’s often the fastest way to get clarity and start seeing real results.
Regards,
Team @ Lukas Holschuh
