Most e-commerce stores run Meta Ads. Few of them run them efficiently. Here is the campaign structure, budget logic and tracking setup that actually protects ROAS.
By Jack Goldsmith, Founder & Performance Marketer, Social Surge · 4 August 2026
What are the best strategies for optimising ad spend on popular social platforms?
Effective strategies for optimising ad spend on popular social platforms centre on four levers: consolidating campaigns so the algorithm receives enough conversion data, setting minimum daily budgets that clear the learning phase, testing creative concepts (not just cosmetic variations), and implementing server-side conversion tracking via the Meta Conversions API. Improving all four in a poorly structured account typically delivers meaningful ROAS gains within four to eight weeks, without increasing total spend.
Meta Ads (Facebook and Instagram) remain the dominant social advertising platform for UK e-commerce. The potential is real: demand creation at scale, full-funnel targeting, dynamic product ads that pull directly from your catalogue. But the potential for waste is equally real. In the accounts we take over, fragmented campaign structures, undersized budgets and broken tracking are the three things we fix first, every single time, before touching creative or audiences.
Meta's algorithm needs conversion data to improve. That is not an opinion; it is how the platform's machine learning operates. Each ad set enters a "learning phase" when it launches or resets, during which Meta is modelling which users are most likely to convert for your specific offer. To exit that phase and stabilise, Meta needs approximately 50 optimisation events per ad set per week, as Meta's Business Help documentation outlines.
The problem: most stores split their budget across too many ad sets. In the accounts we inherit, it is common to find 8-12 active ad sets, each receiving £15-25 per day. At a CPA of £40, a £20/day ad set generates roughly 0.5 conversions per day, or 3.5 per week. That is 7% of what the algorithm needs to exit learning. The campaigns never stabilise. ROAS swings by 40-60% week to week, not because the strategy is wrong, but because the structure starves the algorithm of signal.
The fix is consolidation. Fewer campaigns, fewer ad sets, more budget concentrated per campaign. A cleaner starting structure for most UK e-commerce stores looks like this: one prospecting campaign targeting cold audiences with your best-performing creative, one retargeting campaign targeting warm audiences (website visitors and catalogue viewers from the past 30 days), and if your purchase volume supports it, one Advantage+ Shopping Campaign running alongside for incremental reach.
Advantage+ Shopping Campaigns (ASC) are Meta's algorithm-led product: you feed in your catalogue, set a budget and a ROAS target, and Meta handles audience selection automatically. They work well when your account is generating at least 30-50 purchases per week and your conversion tracking is clean. Below that threshold, you typically need manual campaigns with tighter control. Running both (ASC for volume, a separate manual retargeting campaign for high-intent warm audiences) is a structure we use in a number of our Meta Ads accounts.
On Campaign Budget Optimisation (CBO) versus Ad Set Budget Optimisation (ABO): CBO, where you set budget at campaign level and let Meta distribute it across ad sets, generally outperforms ABO once your campaigns have enough data. ABO gives more granular control and is useful during structured creative tests where you need equal spend across variants.
The 50-conversions-per-week threshold is not just a nice target; it has direct implications for minimum viable budget. Work backwards from your CPA:
As a practical floor: if your total monthly Meta budget is below £500, consolidate everything into a single campaign. Spreading thin budgets across multiple campaigns is the fastest route to permanently unstable performance. Our guide on budget allocation for scaling e-commerce ads covers the Google and Meta split in more detail.
Once structure and budget are correct, creative becomes the primary performance variable. The mistake most stores make here is testing cosmetic variations: same concept, different colours, slightly different headline. That is not a test; it is noise. Meaningful creative tests compare different hooks, different formats (static image versus video versus carousel) and different value propositions.
In the enthusiast niches we focus on (cycling, fishing, outdoor), static product images frequently outperform video for direct-response purchase campaigns. Video has higher production cost and does not consistently win on conversion rate in product-focused categories where clarity of the product itself matters more than storytelling. Test it, but do not assume video is the answer.
Creative fatigue signals to watch for: frequency climbing above 2.5-3.0 on prospecting audiences, click-through rate declining week-on-week, and cost per mille (CPM) rising while conversion rate holds or falls. When you see that combination, refresh the creative before pausing the campaign. Pausing resets learning; fresh creative within the same campaign often revives performance without the reset cost.
Dynamic Creative Optimisation (DCO) is useful for testing at scale but can obscure what is working. If you use it, check asset-level breakdowns regularly to understand which headline and image combinations are driving conversions, not just clicks.
Apple's App Tracking Transparency framework, rolled out from 2021 onwards, removed browser-level cookie tracking for users who opt out on iOS devices. For e-commerce stores with significant mobile traffic, this means the Meta Pixel alone misses a material proportion of conversions. In heavily iOS-skewed audiences, pixel-only accounts can under-report purchases by 20-40%, leading to campaigns being incorrectly paused as unprofitable when they are actually generating revenue.
The fix is the Meta Conversions API (CAPI): a server-side integration that sends conversion events directly from your web server to Meta, bypassing browser-level signal loss. Meta's Conversions API documentation explains the setup in full. For stores on Shopify, the native Meta channel integration includes a CAPI option that covers most common events without custom development. For custom platforms, a server-side implementation is required.
The diagnostic test: compare your Meta-reported ROAS against GA4 last-click attribution for the same date range. A consistent gap of more than 30-40% in either direction suggests a tracking problem rather than a genuine performance difference. Fix the tracking before drawing conclusions about campaign performance.
Once a campaign is performing consistently and your tracking is reliable, scaling is a matter of following a few non-negotiable rules. First, the 15-20% rule: increase budgets by no more than 15-20% every three to four days. Larger increases reset or destabilise the learning phase even in well-established campaigns. This is slower than most store owners want, but it is considerably faster than recovering from a ROAS collapse caused by an aggressive jump.
Second, horizontal scaling before vertical scaling. Before doubling the budget on your best campaign, test new creative concepts, new audience angles and new offers. Fresh creative can extend the life of a campaign at current budget far more efficiently than just spending more on a fatiguing setup.
Third, recognise saturation. Every account has a natural ceiling for its addressable audience size. When CPMs start climbing sharply and ROAS deteriorates despite stable creative and structure, you are likely saturating your Meta audience. At that point, incremental growth comes from adding Google Ads, not from pushing more budget into Meta. Our guide on retargeting strategies covers how to extend reach across both channels without overlap waste.
A common mistake among e-commerce stores is treating Meta and Google as competing channels where budget goes to one or the other. They are not competing; they serve fundamentally different functions in the buying journey.
Meta creates demand: it puts your product in front of people who were not searching for it, building awareness and triggering intent. Google captures demand: it converts people who are already searching for what you sell. Running only Meta means you create demand that your competitors on Google Shopping then capture. Running only Google means you are dependent on pre-existing search volume and have no mechanism for introducing your brand to new audiences.
The accounts that sustain the strongest long-term ROAS in our experience run both channels in a coordinated structure. Meta drives the top of the funnel; Google Shopping and Search close the intent-led traffic that Meta's brand-building generates. Our Google Ads management service is built specifically to complement Meta activity, so the two channels amplify rather than cannibalise each other.
If you want to understand whether your current social and search setup is structured to work together (or against each other), a free PPC audit will show you exactly where the leakage is happening.
There is no universal answer, but for Meta purchase campaigns the algorithm needs approximately 50 conversion events per ad set per week to exit the learning phase. If your average CPA is £40, that means spending at least £40 per day per active ad set just to generate enough data. For most UK e-commerce stores starting out, a realistic monthly budget is £500-£1,500 for Meta Ads, rising once CPA and ROAS targets are consistently hit.
Industry benchmarks typically sit between 2x and 4x ROAS for Meta Ads, though this varies considerably by niche, average order value and account structure. In the cycling and outdoor e-commerce accounts we manage, Meta Ads contribute to blended ROAS figures significantly above those averages when paired with well-structured Google Shopping campaigns and clean server-side conversion tracking.
The two biggest sources of wasted Meta spend are fragmented campaign structures (too many ad sets, each underfunded) and unreliable conversion tracking. Consolidate to fewer campaigns with higher daily budgets per campaign, implement the Conversions API alongside the Meta Pixel, exclude existing customers from prospecting ad sets, and cap frequency on retargeting audiences. These four fixes typically recover 20-30% of wasted spend in an account restructure.
Advantage+ Shopping Campaigns work well when your account has clean conversion data and at least 30-50 purchases per week, giving Meta's algorithm enough signal to optimise efficiently. Below that volume, manual campaigns with separate prospecting and retargeting ad sets give you more control and prevent the algorithm making poor decisions on limited data. Many well-structured accounts run both: Advantage+ Shopping for scale and a separate manual campaign to capture warm retargeting audiences.
Book a free PPC audit: full account review, wasted-spend analysis and a clear growth plan, delivered in plain English.
Book Your Free Audit