ROI Decision Frameworks
ROI thresholds are how finance-literate operators turn a vague "is this working?" into a yes/no decision on channels, tools, tests, and scale-or-cut calls.
ROI Decision Frameworks
A set of ROI thresholds operators use to make yes/no calls on channel spend, tool stack, test priority, and scale-or-cut decisions.
An ROI decision framework is the layer that sits between your reporting and your budget. It turns return-on-investment from a number you look at into a number you act on — by pre-committing to thresholds (a payback window, a minimum LTV:CAC ratio, a contribution-margin floor) that decide what gets funded, paused, or killed.
The framework matters most in four recurring conversations: how much to put behind each acquisition channel, which tools earn their seat in the stack, which experiments are worth running first, and when a campaign is mature enough to scale versus when it's quietly bleeding margin. The thresholds are the bridge between marketing intent and CFO arithmetic.
Most online stores in the €1M-€15M band don't lack ROI numbers — they lack rules. GA4 tells you a channel returned 2.1x last month; nobody on the team can say whether that's a green light, a yellow light, or a quiet disaster once you back out shipping, returns, and the agency retainer.
A decision framework fixes that by writing the rule down before the result is in. The threshold is set against contribution margin, not revenue, and against payback period, not point-in-time ROAS. Once it's written, every channel, tool, and test gets graded against the same line.
Setting the threshold: what counts as "working"
A usable ROI threshold has three numbers attached to it: a minimum return multiple, a payback window, and a confidence level. For most apparel and beauty stores running paid social and search, the working defaults are roughly 3:1 contribution-margin ROI, payback inside 90 days, and at least 30 days of data behind the call.
Those numbers aren't universal — a high-AOV electronics store with a 25% gross margin needs a tighter threshold than a beauty SKU with 70% margin and repeat purchase. The discipline isn't picking the "right" number; it's picking yours and applying it consistently. See ROI Threshold Setting for the full calibration walkthrough.
Channel allocation: where the next euro goes
The first job of the framework is to rank channels against the threshold and against each other. A channel clearing 4:1 with a 45-day payback deserves more budget than one stuck at 2.5:1 with a 120-day payback, even if both are technically "profitable." The marginal euro question matters more than the average.
This is where LTV:CAC ratio enters the conversation. A channel under the short-term ROI threshold can still be a buy if its cohorts repeat — beauty subscriptions are the canonical example. Pair the framework with ROI by Channel Benchmarks to see what each channel typically returns at your AOV tier, and adjust before you redistribute spend.
Payback period beats point-in-time ROAS
A 4:1 ROAS that takes nine months to pay back is a cash-flow problem, not a marketing win. For stores under €15M revenue, payback period is the constraint that breaks first — your supplier terms, ad-platform billing, and VAT cycle don't care about your annualised return. Anchor the framework on payback before you celebrate the multiple.
Tool stack, test priority, and scale-or-cut
The same threshold logic applies to the tools you pay for monthly. A heatmap tool at €200/month needs to drive at least one funnel fix per quarter worth more than €600 in recovered revenue to clear a 1:1 break-even — and most stacks fail that test once you actually do the maths. Tool Stack ROI walks through the audit; the Profitability Dashboard makes it a standing view rather than a once-a-year panic.
For test prioritisation, the framework reframes "what should we test next?" as "which test has the highest expected ROI given hit-rate and traffic?" — typically pricing-page, checkout, and PDP wins, in that order. For scale-or-cut, the rule is mechanical: two consecutive 30-day windows above threshold means scale, two below means cut, anything in between holds spend flat.
Decision outcome by ROI threshold choice (illustrative Shopify apparel store, 6 channels)
ROI decision framework FAQ
A reasonable starting point is 3:1 contribution-margin ROI with a 90-day payback window. Adjust upward if your gross margin is below 40% or your repeat-purchase rate is low, downward if you have strong cohort economics and a beauty- or supplements-style subscription tail.
No. ROAS measures revenue against ad spend; ROI measures contribution margin against fully-loaded cost. A 4:1 ROAS on a 30% margin product is a 1.2:1 ROI before you've paid for shipping, returns, or the agency fee. Always run the framework on margin, not revenue.
LTV:CAC ratio is a long-horizon view of cohort economics; the ROI framework is a shorter-horizon, decision-grade view. You use LTV:CAC to validate the business model and the ROI threshold to make the weekly channel and tool calls. The two should be consistent — if your LTV:CAC is 3:1, your ROI threshold can't sustainably be 4:1.
Once per quarter, or whenever a structural input changes — a price increase, a shipping-cost shift, a new payment provider, a CAC step-change after an iOS update. The thresholds are the framework's slowest-moving layer; channels and tools get re-graded monthly against them.
That's a capacity question, not an ROI one. Treat the channel as the benchmark and use the framework to compare other channels against its return profile, but don't force spend into a channel that's already saturated. The next euro goes to the next-best channel above threshold.
Build a longer measurement window into the framework for those channels — typically 90-180 days versus 30 for paid. Use post-purchase surveys or a holdout test to estimate the true contribution, then grade the channel against the same ROI threshold as paid, just with a longer payback allowance.
Yes — and the maths is usually harsher than for channels. A tool clears the threshold when the documented revenue lift (from tests it enabled, leaks it caught, hours it saved at a real hourly rate) exceeds its annual cost by the threshold multiple. Most stacks have two or three tools that fail this test quietly.
Two consecutive 30-day windows above threshold means increase spend by 20-30%; two consecutive below means cut by 50% or pause. One-and-one holds spend flat and triggers a diagnostic. The mechanical rule matters more than the exact percentages — it removes the "let's give it one more month" trap.
Each test gets an expected-ROI score: (expected lift × affected revenue) / (build cost + traffic opportunity cost). Tests with expected ROI above your threshold go into the queue; ones below get parked. This is why checkout and PDP tests usually win — they touch high-intent traffic with low build cost.
The dashboard is the operational surface of the framework — the place where contribution margin, payback, and ROI by channel are visible on the same screen. Without it, the framework lives in a spreadsheet that gets opened during budget season and ignored the rest of the year.
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.