How to Scale Meta Ads Without Breaking ROAS

You can scale Meta ads without breaking ROAS if you consolidate the account first, feed it more creative across more formats, raise budgets in measured steps and judge the result on blended revenue from Shopify rather than in-platform ROAS. Most scaling attempts fail because spend goes up while creative and structure stay the same.

This is how I approach it, with Ghost as the main example.

What “not breaking ROAS” actually means

When you add spend, you reach people who are less likely to buy than the ones you were already reaching. Some drop in in-platform ROAS is normal. The question is not whether ROAS dips. It is whether the extra spend brings in enough extra revenue, at a healthy enough margin, to be worth it.

That is why I judge scaling on two numbers:

• Blended ROAS or MER. Total Shopify revenue divided by total ad spend across every channel.
• Contribution margin. Revenue after product costs, fulfilment and ad spend. This tells you whether growth is profitable.

In-platform ROAS is useful for comparing ads and ad sets inside Meta. It is a poor guide to whether the business is growing.

Step 1: Consolidate before you scale

Fragmented accounts are hard to scale. When budget is spread across many small campaigns and ad sets, each one gets too little data to optimise, audiences overlap and compete with each other, and a budget increase in one place is hard to measure.

On Ghost, a UK DTC ecommerce brand, one of the first jobs was consolidating fragmented Meta and Google campaigns into simple prospecting and retargeting structures. That made the account easier for Meta to optimise and easier for me to read.

A structure I often use:

• One main prospecting campaign with broad targeting
• One retargeting campaign for site visitors and engagers
• Dynamic product ads fed from the Shopify catalogue, where the range suits it

Fewer moving parts means every budget change has a clear effect you can measure.

Step 2: Increase creative volume and format range

Creative is the main lever on Meta now. Broad targeting means the ad itself does most of the work of finding the right person. If you double the budget without adding creative, you mostly just show the same ads to more people more often, and frequency climbs.

What I look for before scaling:

• Enough live ads per ad set that no single ad carries the whole budget
• A range of formats. Static, video, carousel and catalogue ads reach people differently
• A range of angles. Different reasons to buy, not just different crops of the same image
• A steady refresh rate, so new ads arrive before the old ones fatigue

Format range matters more than most brands think. On Mira Mikati, format-level reporting showed 73% of Meta spend was on static while video returned 4.7×. That kind of imbalance limits how far an account can scale, because one format is doing too much of the work.

Step 3: Measure against Shopify, not Ads Manager

Meta’s attribution counts sales that would have happened anyway and sales other channels also claim. As spend rises, that overlap often grows. If you scale on in-platform ROAS alone, you can end up spending more for revenue that is not new.

So I track a weekly reconciled view: spend from every ad platform alongside Shopify revenue for the same period. If spend goes up and Shopify revenue goes up in proportion, the scaling is working. If Meta reports more revenue but Shopify does not move, it is not.

That reconciled view also catches effects platform reporting misses. On Ghost, the weekly reconciled view exposed a Google Search uplift driven by Meta video. Meta was creating demand that showed up as branded and product searches on Google. Judged on in-platform numbers alone, that Meta spend would have looked weaker than it was.

Building that reconciled report by hand every week is tedious, and it is the kind of thing that gets skipped when things are busy. My sister business, Ad Ops Autopilot, builds dashboards that pull ad platform and Shopify data together automatically, so the blended view is there without the spreadsheet work.

Step 4: Raise budgets in measured steps

Large budget jumps cause two problems. They can push Meta back into a learning phase, and they make it hard to separate the effect of the increase from everything else going on.

How I handle it:

• Moderate steps. Raise spend gradually rather than doubling overnight.
• Wait between steps. Give each increase several days to settle before judging it.
• Watch blended numbers. Check MER and Shopify revenue, not just in-platform ROAS.
• Hold or pull back if blended efficiency drops beyond what the margin can absorb.
• Plan around seasonality. Scaling into a peak trading period behaves very differently from scaling in a quiet month.

There is no single correct step size for every account. It depends on how much spend the account already handles, how much creative is live and how stable conversion rates are.

Step 5: Do not forget retention

Scaling acquisition puts more first-time customers into the business. If they never buy again, every pound of acquisition spend has to pay back on the first order.

Repeat purchase changes the maths. When retargeting, email and SMS bring customers back, you can afford a lower first-order ROAS on prospecting and still grow profitably. On the ad side, I make sure:

• Retargeting is set up and not starved of budget
• Past purchasers are excluded or treated separately in prospecting, so you are not paying to reacquire existing customers
• Customer lists from Shopify feed into both exclusions and lookalike or signal audiences

What this looked like on Soto Gardens

In 2025, Soto Gardens spent £120,212 and generated £515,415 in tracked revenue, a 4.29× blended ROAS. Purchases were up 109% year on year, at an average order value of around £197, and the store recorded £1.35M in gross revenue. Meta’s peak ROAS reached 7.66× in February 2025.

Alongside the Meta work, I added Shopping and Performance Max to a Google account that had been search only, and tested Microsoft Ads as a low-cost incremental channel. The full case study is on the Soto Gardens results page.

The common thread was structure first, creative volume second, and every decision checked against Shopify revenue rather than platform reporting.

Want to scale without guessing?

If you want to spend more on Meta but are not sure whether the account is ready, or whether the extra spend would be profitable, I can look at it for you. Request a free account audit and I will tell you what needs to change before you raise the budget.

Questions people ask

How much should I increase my Meta budget at a time?

Make moderate steps and give each one time to settle before the next. Large jumps tend to push spend into weaker auctions and reset learning, which makes it hard to tell whether the extra money is working.

Why does ROAS drop when I scale Meta ads?

More spend means reaching people who are less likely to buy, and often showing the same creative to them more often. Without more creative and a clean structure, efficiency falls as budget rises.

What is MER and why use it instead of ROAS?

MER, or marketing efficiency ratio, is total store revenue divided by total ad spend. It uses Shopify revenue rather than platform-reported numbers, so it shows whether extra spend is actually growing the business.

Should I consolidate my Meta campaigns before scaling?

Usually, yes. Fewer, larger campaigns and ad sets give Meta more data per ad set and make budget increases easier to manage and measure.