Why one plan beats four budgets
Zero Fluff Marketing builds one media plan across Meta, Google, TikTok and CTV for DTC ecommerce brands, then buys it hands-on.
When each platform is run on its own, each one optimizes for its own dashboard. Meta claims the sale, Google claims it again through branded search, and retargeting claims it a third time. Add up the platform ROAS and you get more revenue than Shopify shows. Each one is counting the same customer.
A single plan fixes the incentive. Every channel is judged on the job it was given, and the total is judged on what actually hits the bank: blended MER, new customer CAC and contribution margin after ad spend.
How we build the plan
1. Start from the unit economics. Before any channel talk, we work out what a new customer is worth on the first order and over 90 days, after product cost, shipping, discounts and returns. That gives the CAC ceiling every channel has to clear.
2. Give every dollar one job. Reach dollars (CTV, YouTube, TikTok, Snapchat) find people who have never heard of you. Acquisition dollars (Meta Advantage+ and prospecting, TikTok, X, creator ads) turn attention into first orders. Capture dollars (Google Search, Shopping, Performance Max, retargeting, Criteo) close the demand the rest of the plan created. Each is judged first on the one number that matches its job.
3. Fund the engine first, on the next dollar. We plan on marginal CAC, not average CAC. Each channel has a spend curve, and the channel that brings in new customers most profitably gets funded until the next new customer costs more than the ceiling. For most DTC brands that is Meta. Only then does money go to reach, because reach without a capture engine just makes someone else’s search ads cheaper.
For the full step by step, see how to split a DTC ad budget.
4. Protect a test budget. Five to ten percent of spend goes to new channels, formats and audiences, with a pass or fail line agreed in advance. The next growth lever gets found before the current one plateaus, not after.
5. Plan the calendar. Launches, promotions and seasonal peaks get their own budget curve, so spend ramps up before demand does instead of chasing it.
How we read the market
Every month, before the budget is re-cut, we check the same signals, and each one has a trigger:
| Signal | Where it comes from | What it can trigger |
|---|---|---|
| CPM and CPC trends by platform | Your accounts | Meta CPMs up 20% with a flat conversion rate shifts test budget toward search and TikTok |
| Branded search volume | Google Ads, Search Console | Rising brand demand with flat sales points to a capture problem, not a reach problem |
| Competitors bidding on your name | Google auction insights | Brand search budget grows only while conquesting is real |
| Competitor creative | Meta Ad Library, TikTok Creative Center | New angles to brief and test |
| New formats and campaign types | Platform releases | A capped test from the test budget, never your core spend |
When something proves out in a test, it graduates into the plan.
How the plan is measured
Three tools, each with its own job:
- Platform attribution runs campaigns day to day. It is fast and useful for choosing between ads and audiences, and poor for deciding budgets.
- Lift tests decide whether a channel gets more money. Platform conversion lift studies and matched-market geo tests, each with a minimum detectable lift set before it starts. Retargeting, branded search and upper-funnel video get tested first, because platform reporting over credits them most.
- Above roughly $1M a year in spend, a media mix model sets the split. An open source model (Meridian or Robyn), refreshed quarterly and calibrated with the lift tests.
All of it is read against Shopify: blended MER (net revenue divided by total paid media spend), new customer CAC, and contribution margin after ad spend. Post-purchase surveys (“how did you hear about us?”) fill the gaps clicks cannot see, especially for CTV and creators. More on MER vs ROAS.
What you get each month
A one-page plan, a short weekly note (what moved, why, and what we are doing about it), and a monthly reallocation with the reasoning written down. No 40-slide decks.