Last updated: November 12, 2026. Facts checked against Meta Business Help Centre and Google Meridian documentation in November 2026.
TL;DR
If you split a Facebook ads budget across multiple brands, divide it by headroom above each brand's break-even, not equally. Multiply each brand's ROAS by its contribution margin to get the headroom ratio. The brand with the highest ratio above 1.0 usually earns the next $1,000. Keep a capped test reserve and review weekly.
Quick answer
Split Meta ad budget between brands by headroom above each brand's break-even, not equally: a brand with headroom above its break-even ROAS gets the next thousand dollars, and a brand below its floor does not. Nova from AdAdvisor works from a separate profile for each business, with that brand's own break-even and the daily and monthly spend ceilings you set.
This guide is for owners of two to ten brands, each in its own Meta ad account, and answers one question: which brand gets the next $1,000. Moving budget inside one account is covered in AI budget reallocation for Meta ads.
AdAdvisor has 8+ years in media buying, more than $60M in managed ad spend and an ex-Meta engineer on the team. A common portfolio pattern: attention goes to the biggest brand, and a smaller brand with more headroom stays underfunded.
How should you split ad budget across several brands?
Rank brands by headroom above their own break-even ROAS and fund from the top. We call this Headroom-First Allocation: it treats a multi-brand ad budget as one portfolio with a separate floor for each brand. Equal splits and revenue-share splits are easier to run, but neither looks at what the next dollar is likely to earn in each brand.
| Rule | When it works | Risk |
|---|---|---|
| Equal split | Brands have similar margins, maturity and spend | Ignores different break-evens: headroom sits unused in one brand while another spends below its floor |
| Split by revenue share | Mature portfolio, every brand near saturation | Locks in the past: the biggest brand keeps growing its budget after its next dollar stops paying |
| Headroom above break-even (Headroom-First) | Each brand has a known contribution margin and a week of stable data | Only as good as margin data and tracking; noisy at very low conversion volume |
| Test reserve | New brands, offers and markets with no history | Turns into a second performance budget if it has no cap |
A brand at 3.8x ROAS can be a worse home for the next dollar than one at 2.6x if its margins are much thinner.
Drowning in Meta Ads?
Put your campaign on autopilot with Nova.
Read moreWhy equal splits starve the best brand
An equal split caps your strongest brand at the same budget as your weakest one. The brand with the most headroom hits its ceiling while profitable demand is still available, and the brand below break-even keeps spending money it is likely losing.
The damage is easy to miss: the capped brand reports a healthy ROAS, so nobody asks what it would have done with more.
Revenue-share splits reward the brand that was biggest last quarter, not the one whose next dollar earns the most. Google's Meridian documentation makes the same point for marketing channels: when marginal ROI is much lower than average ROI, the channel "is beginning to saturate", and the channels with the highest marginal ROI are the best place for additional funds. Treat each brand as a channel and the logic carries over.
How do you compare brands with different margins?
Convert every brand's ROAS into a headroom ratio: ROAS multiplied by contribution margin. A ratio of 1.0 means the brand is at break-even. Above 1.0 it is earning contribution profit on ad spend; below 1.0 it is likely losing money on each ad dollar.
Break-even ROAS is 1 divided by contribution margin, meaning revenue minus every variable cost of an order (product, shipping, payment fees, discounts, returns), before ad spend. For the full method, see the break-even ROAS definition in the AdAdvisor docs and how to calculate break-even ROAS from contribution margin.
Formula
Headroom ratio = ROAS × contribution margin = ROAS ÷ break-even ROAS. 1.0 = break-even.
The headroom ratio minus one is the contribution profit each ad dollar returns on attributed revenue, which puts brands with different margins in the same unit.
Illustrative example, not client data:
| Brand | Contribution margin | Break-even ROAS | Recent ROAS | Headroom ratio | Contribution per ad dollar |
|---|---|---|---|---|---|
| A | 50% | 2.00 | 2.6x | 1.30 | +$0.30 |
| B | 30% | 3.33 | 3.8x | 1.14 | +$0.14 |
| C | 60% | 1.67 | 1.5x | 0.90 | -$0.10 |
Brand B has the highest ROAS and still ranks second; Brand C has the best margin and is likely losing money. The next $1,000 probably goes to Brand A, and Brand C gets a diagnosis first.
Where you have a recent budget step, compute the ratio on the marginal ROAS of that step: extra revenue divided by extra spend, comparing the week after the step with the week before. The seven-day average is the fallback, and it tends to overstate headroom in a brand that is starting to saturate. This before-and-after read is a rough proxy for what marketing mix models such as Meta's open-source Robyn estimate from fitted response curves.
When should a brand get the next $1,000?
A brand earns the next $1,000 when its headroom ratio has stayed above 1.0, with the buffer you require, for at least seven days, its tracking passed the weekly check, and its ad sets are not stuck in learning for reasons money will not fix.
The seven-day minimum borrows from Meta's guidance for ROAS goal campaigns: day-to-day ROAS naturally fluctuates, so judge weekly averages and wait at least seven days after an adjustment before re-evaluating. Brands with long buying cycles or very few conversions may need longer.
Size the step so it does not reset what is working. Meta says budget changes "may or may not" restart the learning phase depending on their size: $100 to $101 is unlikely to, $100 to $1,000 may. A planned step once a week tends to be safer than daily nudges.
Check learning status before you add money. Meta says an ad set becomes "learning limited" when it is unlikely to get about 50 optimization events in the week after its last significant edit, and lists low budget, small audiences and high auction overlap among the causes.
If low budget is the cause, the next $1,000 may be exactly what the brand needs. If the audience is too small, more money likely buys frequency, not customers. Also check the floor: Meta sets minimum budgets that vary by country, objective, optimization event and currency, and notes that purchase optimization may need more budget than easier events. A brand funded below what its event needs cannot be judged fairly.
Nova from AdAdvisor is built to sit next to this decision. She works out break-even with you for each business and judges every campaign against it, and each business on Nova connects its own ad account. Which brand gets the next $1,000 stays your call.
How much should stay in reserve for tests?
There is no universally validated test-reserve percentage. Splits like 70/20/10 are portfolio heuristics, not a Meta recommendation or a research finding, so treat any number you see as a starting point, not a rule.
Size the reserve from the downside instead: how much can you lose in a month on ideas that fail without pushing any established brand below its break-even? New brands, markets and offers draw from it; once something clears break-even for a full week, it graduates into the main allocation.
Cap the reserve in Meta itself, because a daily budget is not a hard cap: Meta may spend up to 75% over a daily budget on some days, though not more than seven times it in a week. An ad account spending limit is the maximum you choose to spend across all campaigns in that account, and ads pause when it is reached. Meta can also readjust its own daily spending limit for a new ad account, so a brand-new account may spend less than planned at first.
The weekly portfolio review (60 minutes)
Run one 60-minute review a week in a fixed order: anomalies first, then headroom, then the allocation decision, then creative. The order matters because an allocation made on broken data moves money in the wrong direction. Weekly is a sensible starting cadence; very high-spend portfolios may need a mid-week anomaly check.
| Step | Minutes | What you check | Output |
|---|---|---|---|
| 1. Anomalies | 15 | Pixel and Conversions API events, sudden spend or ROAS jumps, disapproved ads, spending limits hit | A brand whose tracking failed sits out the allocation until it is fixed |
| 2. Headroom | 15 | Each brand's seven-day headroom ratio and the marginal read on its last budget step | A ranked list of brands |
| 3. Allocation decision | 15 | Who gets the next $1,000, who holds, who drops | Updated budgets and spending limits, with the reason written down |
| 4. Creative | 15 | Whether each newly funded brand has enough fresh ads to absorb the money | Briefs, so extra budget does not turn into extra frequency |
With Nova running each brand's account, this review usually takes about 10 minutes. By the time you sit down, she has already checked pixel and Conversions API health, flagged campaigns below each brand's break-even and briefed new creative before winners burn out. What is left for you is reading her summaries and making the allocation call.
This differs from an agency's weekly rhythm, where the review covers client accounts and reporting, as in the AI Meta ads workflow for agencies. With your own brands, the allocation decision belongs to the owner because it is a P&L decision, not a reporting task.
What the weekly review should not decide
The weekly portfolio review decides how much each brand gets, not how each brand spends it. Keep three kinds of decisions out of the hour.
Moving budget between ad sets inside one brand is covered in when to move budget between ad sets. Changing a brand's contribution margin belongs to finance, with real cost data. And a creative test younger than a week is likely still noise.
What this approach cannot do
Headroom-First Allocation is a decision rule, not a guarantee. It is only as good as each brand's tracking, margin data and attribution.
Separate ad accounts do not guarantee separate auctions. Meta notes that running ads from separate ad accounts "may not help you avoid overlap", because its delivery system may identify several accounts as the same advertiser. Brands that share audiences or assets may limit each other's delivery.
Attribution is incomplete. Meta's default attribution setting is 7-day click, 1-day view and 1-day engagement, and some results are modeled when data is partial. Brands with longer buying cycles can look weaker than they are, and platform ROAS is not the same as incremental profit. For a deeper look at measurement gaps, see multi-touch attribution for Meta ads.
Break-even depends on margin data: stale return or shipping costs distort the ratio. And it ignores lifetime value, so a subscription brand may justify spending below first-order break-even.
FAQ
Summary
Measure every brand against its own floor. Give the next $1,000 to the brand whose headroom ratio has stayed highest above 1.0 for a week, cap a test reserve and start each weekly review with tracking. If your brands sell on Shopify, Nova can hold each brand's account to its own break-even every day, inside the ceilings you set.
Sources
- Meta Business Help Centre: About learning limited
- Meta Business Help Centre: Significant edits and learning phase
- Meta Business Help Centre: About ROAS goal
- Meta Business Help Centre: About budgets
- Meta Business Help Centre: About daily spending limits set by Meta
- Meta Business Help Centre: Understand auction overlap
- Meta Business Help Centre: Best practices for minimum budgets
- Meta Business Help Centre: Results and default attribution setting
- Google Meridian: ROI, mROI and response curves
- Meta Robyn package documentation (CRAN)
Related reading
Analytics & Reporting
Break-Even ROAS vs Target ROAS: How to Set ROAS Tiers for Meta Ads
Break-even ROAS is your floor; target ROAS is where profit starts. See the difference, how to set a target above break-even, and how to build ROAS tiers for scaling Meta ads.
Read more
Performance Optimization
AI Budget Reallocation for Meta Ads: Reallocate Budget Automatically
AI can monitor Meta Ads ROAS signals and reallocate budget automatically. Here's how rule-based, conversational AI, and autonomous agents each handle it differently.
Read more
AI & Automation
AI Media Buying Governance and Guardrails: A Control Framework
The control system that lets an autonomous ad agent run Meta ads without running unsupervised: spend caps, approval-first, audit trails, escalation, and a kill switch.
Read more
Performance Optimization
Meta Ads Setup Series (Part 3 of 5): How to Set Up a Meta Ad Account the Right Way
Your ad account settings are permanent, so getting them right the first time matters. This guide walks you through every decision from currency selection to spending limits and naming conventions.
Read more
