Every brand with a multi-platform advertising budget faces the same question: where should the next dollar go? Amazon, Walmart, Target, Google, Meta, TikTok, and emerging platforms all compete for the same finite advertising budget. Retail media mix is the strategic framework for answering that question, not with gut instinct but with data on where each dollar works hardest.
The stakes are high. For mid-market CPG brands ($5M to $500M revenue), advertising typically represents 10-20% of revenue. A 10% improvement in allocation across platforms can mean millions in incremental profit. Yet most brands still allocate budgets based on historical precedent ("we've always spent 60% on Amazon") rather than performance-based allocation.
Why Retail Media Mix Matters Now
Three shifts make retail media mix strategy more important than it was even two years ago:
Retail media has exploded
Retail media ad spending in the US reached $62 billion in 2025 (eMarketer), up from $45 billion in 2023. Amazon accounts for roughly 75% of that, but Walmart Connect, Target Roundel, Instacart, and Kroger Precision Marketing are all growing fast. More platforms means more allocation decisions.
AI campaigns make each platform a "black box"
Amazon Performance+, Meta Advantage+, and Google Performance Max all use AI to optimize within their own ecosystems. They're good at maximizing their own platform's results. They don't tell you whether shifting $10K from Amazon to Meta would improve your total return. That cross-platform optimization is what retail media mix strategy addresses.
Attribution is getting harder
Each platform reports its own results and takes credit for conversions that other platforms influenced. Without a unified measurement framework, you can't compare Amazon ROAS to Meta ROAS to Google ROAS because each uses different attribution windows, models, and data sources. Multi-touch attribution helps, but even MTA has limitations in the post-privacy era.
Budget Allocation Framework for Mid-Market CPG
There's no universal formula, but there are data-informed starting points. This framework is designed for CPG brands selling through both e-commerce and retail channels.
| Channel | Typical Allocation Range | Primary Goal | Measurement Focus |
|---|---|---|---|
| Amazon Sponsored (Search) | 30-45% | Capture demand at the point of purchase | ACoS, contribution margin, organic rank lift |
| Amazon DSP | 10-20% | Awareness, retargeting, category conquesting | New-to-brand %, ROAS, purchase rate |
| Meta (Advantage+ and manual) | 15-25% | Demand generation, brand awareness, DTC sales | Blended CPA, contribution profit, conversion lift |
| Google (Search + PMax) | 10-20% | Search capture, Shopping, YouTube awareness | ROAS, CPA, search impression share |
| Other Retail Media (Walmart, Target, Instacart) | 5-15% | Retailer-specific visibility and sales | Retailer ROAS, share of shelf, retailer relationship value |
| TikTok / Emerging | 5-10% | Trend-driven awareness, younger demographics | CPM, engagement rate, brand lift |
These are starting points, not rules. A brand selling primarily through Amazon will skew higher on Amazon channels. A brand with strong DTC will invest more in Meta and Google. The goal is to establish a baseline and then optimize based on incremental testing.
How to Optimize Your Retail Media Mix
1. Establish a unified measurement framework
You can't optimize what you can't compare. Set up consistent measurement across all platforms: same attribution window (ideally 7-day click, 1-day view), same revenue definition (net revenue after returns, not gross), and same margin inputs. Tools like Measured, Rockerbox, or Northbeam help unify cross-platform measurement.
2. Run incrementality tests
Platform-reported ROAS is inflated because every platform takes credit for the same conversions. Run geo-holdout tests (turn off a platform in specific markets and measure the impact on total sales) or use conversion lift studies to understand each platform's true incremental contribution. This is the single most important investment in media mix optimization.
3. Shift budget at the margin
Don't make drastic reallocation changes. Move 5-10% of budget from the lowest-performing channel to the next-highest-performing channel. Measure for 4-6 weeks. Repeat. This gradual approach prevents disruption to platform learning algorithms and lets you isolate the impact of each shift.
4. Separate brand and non-brand performance
Branded search on Amazon or Google inflates ROAS because those shoppers were already looking for you. When comparing platform performance for allocation decisions, focus on non-brand metrics. Your "real" efficiency on each platform is measured by how well it captures net-new demand.
Example: Rebalancing the Retail Media Mix
A beauty brand was spending 85% of their retail media budget on Amazon and 15% on Google Shopping, with no presence on Walmart Connect, Target's Roundel, or Instacart Ads. After analyzing their sales data, they found that 22% of their online revenue came from Walmart and Target combined. They rebalanced: 60% Amazon, 15% Walmart Connect, 10% Google Shopping, 10% Instacart, 5% Roundel. The result was a 19% improvement in blended ROAS over six months, driven mainly by lower competition on the newer platforms. This pattern matches the broader trend: eMarketer reports that while Amazon holds approximately 75% of US retail media ad spend, Walmart Connect grew 27% YoY to approximately $4.4 billion in 2025 — signaling that advertisers are diversifying away from Amazon-only strategies.
Common Allocation Mistakes
- Over-concentrating on Amazon. Amazon's first-party data makes its ROAS look better than other platforms, which encourages brands to shift more budget there. But Amazon ROAS includes branded search (people who already decided to buy you). The incremental return often decreases sharply after 30-40% of total spend.
- Ignoring meta-level effects. Meta and TikTok campaigns drive demand that converts on Amazon. If you cut Meta spend and Amazon ROAS drops a month later, you've discovered that Meta was feeding your Amazon funnel. Attribution systems don't always capture these cross-platform effects.
- Setting platform budgets in January and not revisiting until December. Consumer behavior shifts throughout the year. Retail media mix should be reviewed monthly and adjusted quarterly at minimum.
- Comparing raw ROAS across platforms. A 4x ROAS on Amazon Sponsored Products is not the same as a 4x ROAS on Meta. Different attribution windows, different definitions of "conversion," and different incrementality make raw comparisons misleading.
Frequently Asked Questions
How often should I rebalance my retail media mix?
Review allocation monthly and make adjustments quarterly. Major rebalancing (more than 15% shift) should only happen after incrementality testing, not based on platform-reported metrics alone. Seasonal shifts (Q4 holiday, back-to-school, etc.) may require temporary allocation changes.
Should I use a media mix modeling (MMM) tool?
For brands spending $500K+ per year on advertising across 3+ platforms, yes. MMM tools (Google Meridian, Meta Robyn, or commercial options like Measured) use statistical models to estimate each channel's incremental contribution. For smaller budgets, simpler approaches like geo-holdout tests and manual analysis work fine.
What's the minimum budget to run a meaningful multi-platform strategy?
Roughly $15K-$25K per month across all platforms. Below that, you're better off concentrating on 1-2 platforms where your audience is most active rather than spreading thin across five platforms with insufficient budget for any of them to optimize properly.
Getting your retail media mix right is the difference between spending efficiently and wasting 20-30% of your ad budget on the wrong platforms. Our Paid Media team runs cross-platform allocation analysis for mid-market CPG brands. Let's review your current media mix.