How to Build an Ecommerce Paid Media Strategy That Actually Scales
Most ecommerce brands treat paid media as a tap they can turn on for instant revenue. Spend more, sell more. The problem: at some point, every campaign hits a ceiling where extra spend stops producing extra sales, and brands that haven’t built the foundations underneath their ads watch ROAS collapse. This guide walks through how to build a paid media strategy that scales properly across Meta, Google Shopping, and TikTok, where to allocate budget, and what to fix before you increase ad spend at all.
A scalable ecommerce paid media strategy starts with one principle: ads amplify what already works, they don’t fix what doesn’t. Before increasing spend, fix the conversion rate of your site, the lifetime value of your customers, and the email retention loop that turns first-time buyers into repeat ones. Then build a full-funnel structure: Meta for prospecting and remarketing, Google Shopping for high-intent demand capture, TikTok for awareness and creative testing. Allocate 50-60% to prospecting, 25-30% to remarketing, and 15-20% to brand and creative testing. Track contribution margin, not just ROAS, and rebalance quarterly based on actual contribution to revenue, not the platform’s attribution.
Why Most Paid Media Strategies Fail
Most ecommerce brands we audit at 5MS aren’t failing at paid media because their targeting is wrong or their creative is bad. They’re failing because they’ve treated paid as a self-contained channel, disconnected from the website it points to, the emails it should be triggering, and the customer experience it’s selling.
The pattern is consistent. A brand launches Meta and Google ads. Early results look great (low CAC, strong ROAS, sales coming in). They scale spend. Performance softens. They try new creative, new audiences, new campaign types. It stabilises briefly, then dips again. Within 12 months, the cost per acquisition has doubled, ROAS has halved, and every new pound of spend produces diminishing returns.
This isn’t a paid media problem. It’s a full-funnel problem disguised as one.
The four reasons paid media stops scaling
- Conversion rate is the bottleneck. If your site converts at 1.2%, no amount of cheaper traffic will fix it. The first thing more spend reveals is how broken the funnel below the ads is.
- Lifetime value is too low. If your average customer buys once and never returns, every acquisition has to pay for itself in one purchase. That cap on CAC limits how far you can scale.
- Attribution is misleading. Meta and Google both claim credit for the same sale. Following platform-reported ROAS leads to over-investment in channels that just claim the conversion, not channels that drive it.
- The brand has no demand layer. Brands with strong organic search, email lists, and social presence get cheaper paid results because branded search lifts. Brands without these foundations are paying full price for every click.
The Foundations That Have to Come First
Before increasing ad spend, fix the three things that determine whether paid media will scale or stall. Brands that skip this stage waste 30-50% of their ad spend within 6 months. Brands that invest here first typically see 2-3x improvement in blended ROAS within the same period.
Fix site conversion rate (CRO)
A site converting at 1.5% with £1.50 CPC needs £100 in spend to generate one sale. The same site at 3% needs £50. Doubling your conversion rate halves your acquisition cost, with no change to ad performance at all. Audit and fix product page load times, mobile checkout friction, trust signals, and add-to-cart flow before spending more on traffic.
Build the email and SMS retention loop
If your average customer order value is £80 and your CAC is £40, you’ve made £40 on first purchase. If your email retention pushes that customer to a second order within 60 days, you’ve made £120. The difference between these two outcomes is whether paid media can scale. Pre-purchase: abandoned cart, welcome series, browse abandonment. Post-purchase: thank you, review request, replenishment, win-back at 60/90/120 days.
Set up proper measurement
Platform-reported ROAS is the wrong number to optimise on. You need contribution margin (revenue minus product cost, fulfilment, ad spend, returns) across the full customer journey. Set up server-side tracking (Conversions API for Meta, enhanced conversions for Google), connect your CRM or Shopify data to the platforms, and start tracking blended ROAS across all channels, not platform-reported figures.
The Role of Each Channel
The biggest mistake ecommerce brands make is treating Meta, Google, and TikTok as interchangeable channels competing for the same ROAS. They aren’t. Each serves a different role in the customer journey, and a strategy that scales recognises this.
Demand creation + remarketing
Meta excels at finding customers who don’t know they want your product yet (prospecting) and converting those who do (remarketing). The strength is creative-led testing at scale and best-in-class warm-audience conversion.
Demand capture
Google captures high-intent buyers actively searching for what you sell. The highest-ROAS channel for most brands because it converts existing demand rather than creating it. Essential, but doesn’t scale infinitely.
Awareness + creative
TikTok works for top-of-funnel awareness and as a creative testing ground. Conversion attribution is weaker than Meta or Google, but it influences searches and visits that show up on other channels (the “TikTok effect”).
Branded + competitor
Bid on your own brand to defend against competitors stealing branded clicks, plus selective competitor bids on terms where you genuinely outperform. Lower volume than Shopping but high-converting.
The strategic insight: Google captures the demand other channels create. If you only invest in Google Shopping, you’re harvesting a market you didn’t plant. If you only invest in Meta and TikTok, you’re creating demand competitors are harvesting. The strategy that scales runs all three in concert, with each channel doing what it does best.
Meta Ads for Ecommerce
Meta (Facebook and Instagram) is the workhorse of ecommerce paid media. The platform’s automation has improved dramatically with Advantage+ Shopping Campaigns (ASC), and for most ecommerce brands it’s the largest single channel.
Campaign structure that scales
Meta now favours consolidated campaign structures over the old “10 ad sets with different audiences” approach. The structure we recommend at 5MS for most ecommerce brands:
- 1 Advantage+ Shopping Campaign for top-of-funnel prospecting (broad targeting, AI-driven optimisation, 70-80% of prospecting budget)
- 1 retargeting campaign with 3-5 audience tiers (website visitors 30 days, ATC 14 days, viewed product 7 days, etc.)
- 1 catalogue dynamic product ads campaign targeting cart abandoners and product viewers with the specific products they saw
- 1 creative testing campaign with a small budget (£100-£300/day) running new creative weekly
Creative is the variable that matters
Within Meta’s auction system, audience targeting has been commoditised by their machine learning. The variable you can still control is creative quality and volume. Brands scaling Meta successfully are producing 20-50 new creatives per month and rotating them based on performance. The brands stuck at the same scale for years are running the same 5 creatives forever.
Creative testing framework
- Run 3-5 new creatives per week into the creative testing campaign
- After 7-14 days, measure CTR, hook rate (3-second views), and CPA
- Move winning creatives into the Advantage+ Shopping Campaign and remarketing campaigns
- Repeat. Creative fatigue happens within 4-8 weeks even for top performers.
Google Shopping Strategy
Google Shopping captures demand at the bottom of the funnel where buyers are typing specific product searches into Google. The platform now defaults to Performance Max (PMax) campaigns, which combine Shopping, Search, Display, YouTube and Discovery into one AI-managed campaign type.
Performance Max done properly
PMax is powerful when set up correctly and a black box that wastes spend when it isn’t. The keys to making it work:
- Feed quality is everything. Product titles, descriptions, images, and structured data in your feed are the only signals you’re giving Google. Generic titles (“Black T-Shirt”) rank for nothing. Specific, keyword-rich titles (“Men’s Black Organic Cotton T-Shirt UK Made”) win.
- Use asset groups by product theme. Don’t lump everything into one asset group. Group by product category, margin tier, or seasonality so you can manage them independently.
- Audience signals matter. Even though PMax is AI-managed, audience signals (your customer list, in-market segments, custom segments) materially affect performance. Always add them.
- Negative keyword lists. Add brand exclusions and category exclusions to stop PMax cannibalising your branded search or wasting budget on irrelevant searches.
- Set realistic ROAS targets. Setting tROAS too high kills volume; too low wastes budget. Most ecommerce brands should target 3-5x ROAS for PMax, depending on margin.
Standard Shopping campaigns still have a place
Some brands run standard Shopping alongside Performance Max for products where they want more manual control (high-margin SKUs, new product launches, premium tier products). This works particularly well when you have product-level performance data showing certain SKUs need different bidding logic than the rest of your catalogue.
Brand search bidding
Always bid on your own brand. Competitors will if you don’t, and brand searches convert at 8-15% typically. The £1-2 per click is essentially insurance against losing your highest-intent traffic to a competitor’s bid.
TikTok Ads for Ecommerce
TikTok is the most misunderstood paid media channel for ecommerce. Brands either dismiss it as “for younger audiences only” or expect it to deliver Meta-level direct response. Neither approach works.
The reality of TikTok for ecommerce
TikTok ads work for top-of-funnel awareness, brand building, and creative-led demand creation. Last-click attribution typically undersells TikTok by 30-60% because viewers see the ad on TikTok, then later search Google or visit direct. The “TikTok effect” shows up in branded search lift, direct traffic, and organic social, not in TikTok’s own ROAS reports.
When TikTok works
- Products that demo well on video (anything visual, anything with a transformation, anything physical with sensory appeal)
- Brands willing to produce native-feel creative (not polished brand ads, but content that fits the platform)
- Mid-funnel and remarketing campaigns (TikTok remarketing performs strongly because of high engagement signals)
- Brands with margin and pricing structure that can sustain longer payback periods
When TikTok struggles
- High-consideration purchases over £200 (less impulse-driven)
- B2B or professional ecommerce (wrong audience attention mode)
- Brands trying to repurpose Meta or Google creative without adapting (it doesn’t work)
- Tight last-click attribution budgets (TikTok’s contribution shows up across other channels)
The 5MS approach
For most ecommerce brands, we recommend running TikTok at 10-20% of total paid spend as a creative testing and awareness channel, with the explicit understanding that its contribution will show up as branded search lift and direct traffic increases, not pure TikTok-attributed sales. Brands that measure properly (using marketing mix modelling or post-purchase surveys) consistently find TikTok contributes 1.5-3x more than its attribution claims.
Budget Allocation Framework
The right budget split depends on your brand stage, margin structure, and existing demand. Here’s how we typically allocate budget at 5MS across three common ecommerce stages.
| Channel | Early stage (£5k-£20k/month) | Growth stage (£20k-£100k/month) | Scale stage (£100k+/month) |
|---|---|---|---|
| Meta prospecting | 35-45% | 40-50% | 35-45% |
| Meta remarketing | 10-15% | 10-15% | 10-15% |
| Google Shopping / PMax | 30-40% | 25-35% | 20-30% |
| Google branded search | 5-10% | 3-5% | 2-3% |
| TikTok Ads | 5-10% | 10-15% | 10-20% |
| YouTube / other | 0-5% | 5-10% | 10-15% |
Why the split changes by stage
- Early stage leans into Google Shopping because it captures existing demand at the highest ROAS. The brand is establishing what works before investing in demand creation.
- Growth stage shifts toward Meta prospecting because Google demand has been largely captured and the brand needs to create new demand to keep growing.
- Scale stage diversifies further into TikTok, YouTube, and other channels because Meta and Google have reached saturation and incremental scale requires new audience surfaces.
The split that doesn’t work: 100% on one channel
Brands that run only Meta hit a creative-fatigue ceiling. Brands that run only Google can’t grow past existing demand. Brands that run only TikTok rarely achieve direct response scale. Concentration risk is real. Diversification isn’t just a portfolio principle, it’s a structural requirement to keep scaling.
Measuring What Actually Matters
The metrics that move scaled ecommerce brands are different from the metrics that move starting-out brands.
Stop optimising on platform-reported ROAS
Meta and Google both claim credit for the same conversions. Adding their reported ROAS together routinely overstates actual revenue by 30-60%. A campaign showing 5x ROAS in Meta might actually be delivering 2.5x when measured against real revenue contribution.
Start optimising on these instead
- Blended ROAS: total revenue divided by total ad spend across all channels. This is the only ROAS number that matches your actual business.
- Contribution margin per channel: revenue minus product cost, fulfilment, ad spend, returns. The number that tells you whether scaling a channel actually makes you money.
- New customer CAC vs CLV: are you acquiring customers at a cost that pays back within 12 months including their future repeat orders?
- Marketing efficiency ratio (MER): total revenue divided by total marketing spend. Some scaling brands target 3-4x MER as the central scale metric.
- Branded search lift: are TikTok and top-of-funnel Meta driving more people to search your brand on Google? If yes, your awareness layer is working.
Set up the data properly
None of this works without proper tracking. Server-side conversion tracking (Conversions API for Meta, enhanced conversions for Google), first-party data flowing into the platforms, CRM integration so you can see customer lifetime value not just first purchase. These aren’t optional anymore. Brands operating with broken iOS 14+ attribution and no server-side setup are flying blind and don’t know it.
How Paid Media Connects to Email and CRO
The brands that scale paid media don’t just scale paid media. They scale the surrounding system. Three connections matter most.
Paid media is the top of an integrated funnel
An ad gets a click. The landing page either converts the click or captures the email. The email turns the prospect into a customer over the next 30 days. The post-purchase email sequence turns the first-time customer into a repeat one. Without each of these layers, paid media performance is artificially low because you’re only counting first-purchase revenue from a system that should be producing months of follow-up purchases.
CRO is the multiplier on every ad pound
Every 1% improvement in conversion rate is a 100% improvement in CPA at the same spend. Brands that scale paid spend without continuous CRO see diminishing returns within 6-12 months because more traffic exposes more friction in the conversion path. CRO and paid media aren’t separate disciplines. They’re the same growth function with different tools.
Email is where the LTV grows
If your average customer buys once at £80 net margin, your maximum sustainable CAC is £80. If your email retention pushes that customer to 3 orders over 12 months at the same margin, your maximum sustainable CAC is £240. The brand with better email retention can outbid the brand with worse email retention on every keyword and every audience. This is how brands “find cheaper traffic” that competitors think doesn’t exist. They’ve made their traffic more valuable.
A 90-Day Plan to Scale Properly
If you’re starting from a position where paid media isn’t scaling and you want to fix it methodically, here’s the 90-day sequence we use at 5MS.
Days 1-30: Audit and foundations
Run a full audit of current paid performance, site conversion rate, email retention sequences, and tracking setup. Fix tracking (server-side, CAPI, enhanced conversions). Map the actual customer journey from first click to repeat purchase. Identify the biggest single bottleneck (usually conversion rate or email retention).
Days 31-60: Rebuild and restructure
Restructure paid campaigns into a clear prospecting / remarketing / brand split. Consolidate Meta into Advantage+ Shopping plus remarketing. Restructure Google PMax with proper asset groups and audience signals. Add TikTok creative testing as a small allocation. Fix the top 3 CRO issues identified in the audit. Build out missing email flows.
Days 61-90: Scale and optimise
Start incrementally scaling spend in 10-20% increments week-on-week as long as blended ROAS holds. Maintain a weekly creative testing programme of 3-5 new ads per week. Track new customer CAC vs CLV (12-month projection) as the central scale metric. Rebalance budget allocation based on actual contribution margin per channel.
After 90 days, the brand should be operating with proper measurement, a sustainable scale rate, a working creative testing loop, and clarity on which channels are actually driving growth vs claiming credit for it.
Key takeaways
- Paid media amplifies, it doesn’t fix. Get CRO, email retention, and measurement right before scaling spend
- Each channel has a role. Meta creates demand, Google captures it, TikTok builds awareness. Use them together
- Budget allocation depends on stage. Early stage favours Google; growth stage favours Meta; scale stage diversifies into TikTok and YouTube
- Platform-reported ROAS is misleading. Optimise on blended ROAS, contribution margin, and customer LTV instead
- Creative is the variable that matters most. Brands scaling Meta successfully run 20-50 new creatives per month
- Email and CRO are paid media multipliers. Better email retention raises sustainable CAC; better CRO halves it
- Concentration risk is real. No brand scales sustainably on a single channel. Diversification is structural, not optional
Talk to 5MS about your paid media performance
We help UK ecommerce brands scale paid media properly, with full-funnel CRO, email retention, and measurement built in. 15 years of UK ecommerce experience, 98% client retention, and the integrated approach that makes paid media actually work.
Book a strategy call See our servicesFrequently Asked Questions
Related reading from 5MS
- Magento and ecommerce growth strategies blog
- 5MS full-service ecommerce agency UK
- Talk to our paid media team
This article is updated periodically with current paid media platform changes, UK budget benchmarks, and refreshed scale frameworks based on the brands we work with at 5MS. Last reviewed against current Meta, Google, and TikTok advertising platforms.