Solved: Facebook Ads Budget Increase Worsens Results

Hi there,

Thanks for reaching out!

I had a look over the situation you described. It's a classic problem, and honestly, a good one to have. It means you've found something that works! But hitting that scaling wall where more money equals worse results is incredibly frustrating. I see it all the time. People think scaling is just a case of turning up the budget dial, but it's a lot more nuanced than that.

The good news is you're not limited at all. The reason your attempts have failed isn't because your product or ad is flawed, it's because the method of scaling you're using is actually working against Facebook's algorithm. To spend £1,000 or £10,000 a day, you need to shift from thinking about a single campaign to building a full-funnel advertising 'machine'. I'll walk you through some of the thinking and structures we use to get past this exact hurdle.

We'll need to look at why your current scaling method is failing...

Right, let's get straight to the heart of it. Why does performance tank when you increase the budget? You've found a little pocket of gold with your £80/day campaign. At that spend level, Meta's algorithm has gotten very efficient at finding the best, most likely-to-convert people within your broad audience. It's stable, it's predictable, and it's happy.

When you duplicate that campaign and whack the budget up to £200 or £400, two things go wrong:

1. You Reset the 'Learning Phase': Every new campaign or ad set has to go through a learning phase. It needs about 50 conversions in a week to properly understand who your best customers are. When you duplicate your successful campaign, you're not duplicating the learnings. You're starting a brand new campaign from scratch. The new campaign has no idea what made the old one work; it's just got a pile of cash (£400/day) and an instruction to spend it fast. This often leads to it targeting lower-quality users just to hit the budget, hence the poor results.

2. You Shock the System: A sudden jump from £80 to £400 is a massive shock to the algorithm. It panics. Its primary goal becomes spending the money, not spending it wisely. It'll enter auctions it would normally avoid and show your ad to people further away from your ideal customer profile. It's like telling a skilled salesperson who normally handles 5 high-quality leads a day to suddenly handle 50 leads of any quality. Their conversion rate will plumet. Gradual increases are much more effective. A 20% increase every few days on a *winning* campaign is the standard advice for a reason. It allows the algorithm to adjust gracefully.

The idea that you can just keep increasing the budget on one ad set forever is a myth. This is called 'vertical scaling'. It has a ceiling. At some point, you saturate the most valuable part of your audience and your costs skyrocket. The people spending thousands per day aren't doing it through one magic ad set. They are doing it through 'horizontal scaling'—which means building a wider, more robust structure that allows you to spend more across multiple campaigns and audiences simultaneously.

I'd say you need to build a proper advertising funnel...

This is the real secret to scaling. Instead of one campaign trying to do everything, you create a system of campaigns that work together, each with a specific job. This is the classic Top-of-Funnel (ToFu), Middle-of-Funnel (MoFu), and Bottom-of-Funnel (BoFu) approach. It gives you control, clarity, and multiple 'levers' to pull to increase spend.

Here’s what that looks like in practice:

Top of Funnel (ToFu) - Finding New People (Prospecting)
This is where your current successful campaign lives. Its job is to introduce your brand to people who have never heard of you. But to scale, you can't rely on just one. You need to build out a whole 'prospecting' division. This is where you test new audiences and creatives to find more winning combinations.

• Audience 1: Your Current Winner. Keep your broad, Advantage+ campaign running. It's your control, your benchmark. Don't touch it, just let it do its thing.
• Audience 2: Interest Stacking. You said interests don't work, but they often fail due to poor structure. Try grouping 5-10 highly relevant interests together. If you sell high-end coffee gear, you might group interests like 'James Hoffmann', 'Specialty Coffee Association', 'Fellow Products', and 'Aeropress'. Create several of these themed groups and test them against each other.
• Audience 3: High-Value Lookalikes. Once you have enough data (at least 1000 purchasers is ideal), create a Lookalike audience based on your customer list. This is often pure gold. You can also create lookalikes from people who've Added to Cart or Initiated Checkout. These are your most powerful prospecting tools beyond 'broad'.

Middle of Funnel (MoFu) - Warming Them Up (Engagement)
This is a step many people miss. Not everyone who sees your ToFu ad is ready to buy. MoFu's job is to stay in front of people who've shown some interest but haven't taken that final step. You're building trust and reminding them you exist. The budget here is usually smaller.

• Targeting: People who have engaged with your Facebook or Instagram page in the last 30 days, or people who have watched 50% of one of your video ads.
• The Ad: This isn't a hard sell. It could be customer testimonials, a behind-the-scenes look at how your products are made, or content that showcases the product in use. You're building a relationship, not just asking for a sale.

Bottom of Funnel (BoFu) - Closing the Deal (Retargeting)
This is your highest-performing campaign. It targets people who are on the verge of buying and just need a final nudge. This is where you make your money back.

• Targeting: People who have Added to Cart or Initiated Checkout in the last 7-14 days (but haven't purchased). This audience is red-hot.
• The Ad: You can be more direct here. Remind them what they left in their cart. Maybe offer a small incentive like free shipping if that's viable for you. Use Dynamic Product Ads (DPA) to show them the exact products they looked at. This is usually incredibly effective.

You probably should understand your customer's true value...

Here's the next big mindset shift. The reason you're scared to scale is because you're likely judging success on a day-to-day ROAS (Return On Ad Spend). When you increase the budget and the ROAS drops from, say, 5x to 2x, you panic and pull back. But what if a 2x ROAS is actually wildly profitable in the long run?

The only way to know this is to stop thinking about a single purchase and start thinking about Customer Lifetime Value (LTV). How much profit does a new customer generate for you over their entire relationship with your business? Once you know this number, it changes everything.

Let's do some quick maths. We need three figures:

1. Average Revenue Per Account (ARPA): The average monthly spend of a customer. For ecommerce, you might calculate this as (Total Revenue in a Year / Number of Unique Customers) / 12.
2. Gross Margin %: Your profit margin after the cost of goods is taken out.
3. Monthly Churn Rate %: The percentage of customers you lose each month. This is trickier for ecommerce, but you can estimate it. A simple way is to see what percentage of customers who bought 6 months ago have not bought again since.

The calculation is: LTV = (ARPA * Gross Margin %) / Monthly Churn Rate

This tells you the total *profit* you can expect from a new customer. Now you can make an informed decision on how much you're willing to pay to acquire one. A healthy rule of thumb is a 3:1 LTV to CAC (Customer Acquisition Cost) ratio. This means you can comfortably spend up to one-third of your LTV to acquire a customer and still have a very healthy business.

Play around with the calculator below. You might be surprised at how high a CPA you can actually afford.

Suddenly, if you know you can afford to pay £200 for a customer, seeing a £100 CPA when you increase the budget doesn't seem so bad, does it? It looks like a bargain. This is the maths that allows brands to confidently spend thousands a day. They're not gambling; they've done the calculations.

You'll need a systematic testing process...

Your statement, "I cant use other audience, I cant use any creatives, I cant use interests," is the biggest roadblock right now. I understand why you feel that way—your tests have failed so far. But it's almost certainly not true. It's more likely that the *way* you've tested has been the problem.

To scale, you need to build an engine that constantly finds new winning ads and new winning audiences. Your one successful campaign is the 'control'. The goal is to find new 'challengers' that can beat it, or at least perform well enough to run alongside it.

Creative Testing
Don't just make random new ads. Test with a purpose.

1. Test Angles: Before you even think about images or videos, think about the message. What problem does your product solve? For an ecommerce store, you could test:
   • An angle focused on the quality and craftsmanship of your products.
   • An angle focused on a specific problem it solves (e.g., "Tired of generic, mass-produced gifts?").
   • A social proof angle, using customer reviews and user-generated content (UGC). We've seen UGC work wonders for SaaS clients, and it's even more powerful for e-commerce.
2. Test Formats: Once you have a winning message, test it in different formats.
   • High-quality static image.
   • A simple video showing the product in use.
   • A carousel ad showing different products or different features of one product.

Audience Testing
This goes back to the funnel structure. Your ToFu campaigns should be a battlefield where new audiences compete. A good way to structure this is with a Campaign Budget Optimisation (CBO) campaign.

Create one 'Prospecting Test' campaign with CBO turned on. Inside it, create 3-4 different ad sets.

• Ad Set 1: Your winning Broad audience (as a benchmark).
• Ad Set 2: A Lookalike of your Purchasers (1%).
• Ad Set 3: A stack of your best-guess interests.
• Ad Set 4: Another Lookalike (e.g., Added to Cart 2%).

Use your best 2-3 ads (creatives) in all of the ad sets. Set a healthy budget for the campaign (£100-£200/day) and let it run for 4-7 days. CBO will automatically shift the budget towards the ad sets that are performing best. After a week, you can turn off the losers and replace them with new challengers. This is horizontal scaling in action. You are constantly searching for new pockets of customers.

This is the main advice I have for you:

Scaling isn't about finding one secret setting and pouring money into it. It's about building a robust system that can handle more pressure. It requires a shift in mindset from short-term tactics to long-term strategy, underpinned by a solid understanding of your business's economics. I've detailed my main recommendations for you below:

I know this is a lot to take in. Moving from a single campaign to a multi-layered machine like this takes time, expertise, and a rigorous testing methodology. It's a significant step up in complexity. Many businesses find that this is the point where bringing in a specialist can save a lot of money and time, by implementing these proven structures correctly from the start and managing the ongoing optimisation process.

This isn't just theory. I remember we worked with a women's apparel brand that faced a similar scaling challenge. By implementing this exact full-funnel structure and a disciplined testing process on Meta Ads, we were able to generate a 691% return on their ad spend. It's this strategic approach that truly unlocks growth.

We do this day-in, day-out for e-commerce and SaaS businesses. If you'd like to have a chat and walk through your account on a screen share, we offer a completely free, no-obligation initial consultation where we can give you some more specific pointers.

Hope this helps!

Regards,

Team @ Lukas Holschuh