ROAS Floor vs ROI Threshold for Scale Decisions
A ROAS floor tells you when to stop bidding; an ROI threshold tells you when to stop scaling. Here's why they're different numbers and which decision each one governs.
ROAS Floor vs ROI Threshold for Scale Decisions
A ROAS floor is a revenue-per-ad-euro minimum; an ROI threshold is a margin-per-euro-after-payback minimum. They govern different decisions.
A ROAS floor is the revenue-to-ad-spend ratio below which a channel or campaign stops being worth bidding on. It ignores contribution margin, fulfilment cost, and how long the customer takes to pay back. An ROI threshold is the return-on-investment ratio a channel must clear on a fully-loaded, payback-adjusted basis before you commit incremental budget to scale it.
Most pacing arguments happen because growth teams optimise to the ROAS floor and finance approves budget against an ROI threshold. The two numbers rarely equal each other, and the gap between them is where scale decisions actually live.
The distinction matters because a channel can beat its ROAS floor every day of the quarter and still destroy contribution margin once you account for COGS, shipping, returns, and the twelve weeks it takes new customers to repay their acquisition cost. That's not a reporting glitch — it's the definition of the two metrics doing its job.
Think of ROAS as a revenue signal and ROI as a profit signal. ROAS is fast, ad-platform-native, and useful for in-flight bid decisions. ROI is slower, needs finance-owned inputs (margin, payback window, blended CAC), and is the number that should gate whether you push a working channel from €50k to €150k monthly spend.
Typical ROAS floor vs ROI threshold by store profile (illustrative ranges)
| Store profile | Contribution margin | ROAS floor | ROI threshold to scale |
|---|---|---|---|
| Apparel, one-shot purchase | 35-45% | 2.5x - 3.0x | 1.4x - 1.6x |
| Beauty, repeat but non-subscription | 50-60% | 2.0x - 2.5x | 1.5x - 1.8x |
| Subscription DTC (coffee, supplements) | 55-65% | 1.6x - 2.0x | 1.8x - 2.2x |
| Consumer electronics | 20-30% | 4.0x - 5.0x | 1.2x - 1.4x |
| Home & furniture, high AOV | 30-40% | 3.0x - 4.0x | 1.3x - 1.5x |
Notice the pattern: high-margin subscription brands can tolerate a low ROAS floor (they'll recover it on repeat orders) but should hold a stricter ROI threshold because scaling the wrong way locks in unprofitable cohorts for months. Low-margin categories are the opposite — the ROAS floor has to be aggressive because there's almost no margin to recover.
Why the two numbers diverge
ROAS is a top-line ratio: revenue ÷ ad spend. It doesn't know your gross margin, your return rate, your fulfilment cost per order, or the discount code the shopper stacked at checkout. On a €80 apparel order at 40% contribution margin, a 3.0x ROAS produces €32 of gross profit against €26.67 of ad spend — a razor-thin 1.2x contribution ROI.
ROI, when finance calculates it, bakes in the full P&L stack and often a payback window. A 9-month payback brand asks: after nine months of repeat orders from this cohort, is contribution margin greater than acquisition cost by the required hurdle? That's a fundamentally different question than 'did today's Meta campaign return more revenue than it spent?'
The most common misalignment
Growth reports 'we're hitting a 2.8x ROAS, we should scale' while finance sees a 1.1x ROI on the same cohort and blocks the budget increase. Both are right. The ROAS floor was set for in-flight bidding; the ROI threshold governs the scale decision. Trying to reconcile them with a single number is what causes the weekly pacing meeting to run 45 minutes long.
Which number governs which decision
Use the ROAS floor as your bid cap. It's the number the media buyer checks daily, the threshold that triggers pausing an ad set, and the input that goes into Meta's or Google's target-ROAS bidding. It should be tuned quarterly as CPMs, seasonality, and product mix shift.
Use the ROI threshold as your scale gate. It's the number finance signs off, the hurdle a channel must clear on a fully-loaded basis before you release incremental budget, and it should stay stable across quarters — because it reflects your cost of capital and target contribution, not this week's auction dynamics. When MER (media efficiency ratio) enters the picture at a portfolio level, it can override both, but that's a separate blended check.
Meta prospecting: clearing the ROAS floor while failing the ROI scale test
Reported ROAS (vs 2.5x floor)
Contribution ROI (vs 1.5x threshold)
Frequently asked questions
A ROAS floor is a revenue-to-ad-spend minimum used for in-flight bidding — it ignores margin. An ROI threshold is a fully-loaded, margin- and payback-adjusted return that a channel must clear before you approve incremental scale budget.
You can, but you'll either over-scale unprofitable channels (if you only use ROAS) or under-invest in fast-moving auctions where finance's slow ROI signal lags real-time bid decisions (if you only use ROI). The two-number setup exists because bid decisions and scale decisions run on different clocks.
Multiply your ROAS by contribution margin (as a decimal), then discount for your payback window and any repeat-purchase uplift. A 3.0x ROAS at 40% contribution margin is a 1.2x first-order ROI before payback effects — usually below any healthy scale hurdle.
The ROI threshold. Your ROAS floor moves quarterly with CPM inflation, creative fatigue, and seasonality; your ROI threshold should stay stable because it reflects cost of capital and target contribution. If finance keeps loosening the ROI threshold to match a slipping ROAS floor, you've lost the point of having two numbers.
Subscription brands can run a lower ROAS floor because repeat orders recover the margin, but should hold a stricter ROI threshold — scaling the wrong cohort locks in months of unprofitable retention. Non-subscription DTC is the reverse: higher ROAS floor, more forgiving ROI threshold.
MER (marketing efficiency ratio: total revenue ÷ total paid spend) is a portfolio check that overrides both when blended performance breaks. If channel-level ROAS and ROI look fine but MER is dropping, you likely have cannibalisation or attribution inflation — investigate before scaling further.
Yes. Prospecting ROAS floors are typically 30-50% below your blended target because you're paying for future retention value, not first-order profit. But the ROI threshold on that prospecting spend should still clear the blended hurdle once payback is included.
ROAS floor: quarterly, or whenever CPMs shift more than 15%. ROI threshold: annually, alongside the finance planning cycle. Changing the ROI threshold mid-quarter to unblock a scale decision is usually a sign the underlying assumption (margin, payback) was wrong, not that the threshold was too strict.
That's fine — set the platform target to your ROAS floor and use the ROI threshold as an internal scale gate. Approve budget increases only when the channel clears both. The platform doesn't need to know about your ROI hurdle; your finance and growth leads do.
Growth or performance marketing owns the ROAS floor and tunes it against auction reality. Finance owns the ROI threshold and updates it against margin and payback assumptions. Both numbers should sit side-by-side on the weekly pacing dashboard so neither team is optimising in isolation.
Track CAC, channels, and funnel conversion in one place
Metricuno connects ad spend, funnel events, and revenue so you can see CAC by channel, cohort, and campaign — without stitching together five tools.