ROI Analysis for Marketing Campaigns: Separating Real Profit from Platform Vanity Metrics
A myths-to-action framework for US marketing leaders who need defensible ROI math, not another ROAS dashboard.

Image via 123RF
In This Article
ROI Is Not ROAS
Break-Even ROAS From Margin
Incrementality Beats Last-Click
Five Things Most US Marketing Teams Get Wrong About Campaign ROI
Most ROI conversations stall because the people in the room are not measuring the same number. Agree on the definition first: marketing ROI is incremental profit attributable to campaigns above and beyond what would have happened anyway, divided by total campaign cost. Anything else is a ratio that flatters the channel.
- Myth: a 4x ROAS means the campaign is profitable. Fact: margin decides, not the multiple.
- Myth: platform-reported conversions are accurate. Fact: modeled, deduplicated, and consent-suppressed events routinely overstate lift.
- Myth: last-click is the truth. Fact: it over-credits branded search and under-credits upper-funnel work.
- Myth: more traffic equals more ROI. Fact: traffic that does not convert drags the ratio down.
- Myth: ROI is a monthly vanity report. Fact: it is a decision cadence tied to cash flow.
If your reporting deck leads with platform ROAS and never mentions gross margin, returns, or fulfillment cost, it is a performance summary-not an ROI analysis.
How Marketing ROI Differs From ROAS, MER, and CAC Payback
ROAS is revenue divided by ad spend and stops there, ignoring COGS, shipping, discounting, returns, and overhead. MER (marketing efficiency ratio) is total revenue divided by total marketing spend; it is channel-agnostic and robust to attribution error because it reconciles against your Shopify or WooCommerce revenue line. CAC payback measures how many months of gross profit it takes to recover acquisition cost-critical for subscription and B2B models where cash timing matters as much as margin.
ROI vs ROAS vs MER vs CAC Payback: A Metric Comparison
| Metric | Definition | What It Ignores | Best Use | Failure Mode |
|---|---|---|---|---|
| ROI | (Incremental profit - marketing cost) ÷ marketing cost | Attribution error if incrementally unverified | Board-level profitability decisions | Garbage-in incremental assumptions |
| ROAS | Revenue ÷ ad spend | COGS, returns, fulfillment, overhead | In-channel bidding and budget pacing | Looks healthy while losing money |
| MER | Total revenue ÷ total marketing spend | Channel-level granularity | Company-level efficiency guardrail | Masks one channel subsidising another |
| CAC payback | Months to recover CAC from gross profit | Long-term retention value | Cash-flow planning, subscription/B2B | Short payback that ignores churn |
| LTV:CAC | Lifetime value ÷ acquisition cost | Timing of cash | Durability of the growth model | Roughly 3:1 is a rule of thumb, not a law |
How to Calculate Break-Even ROAS From Gross Margin
Break-even ROAS is the multiple at which ad spend consumes all gross profit-the floor, not the target. Derive it from margin before you approve another budget increase.
Break-even ROAS = 1 ÷ gross margin
40% margin → 1 ÷ 0.40 = 2.5x
25% margin → 1 ÷ 0.25 = 4.0x
15% margin → 1 ÷ 0.15 = 6.7x (before returns & fulfillment)
A US homeware brand selling at roughly 40% gross margin earns ZAR 400 profit on a ZAR 1,000 order (figures illustrative). At 2.5x ROAS the campaign breaks even; every rand of ad spend beyond that is subsidy. Net out returns before you celebrate.
Why Platform-Reported Conversions Overstate Campaign Profitability
Under ongoing signal loss-app tracking restrictions, ad blockers, and consent suppression-platforms fill gaps with conversion modeling. Two platforms can legitimately claim the same purchase. Server-side tagging via GTM plus the Meta Conversions API closes much of the gap, but reconciliation against GA4 remains mandatory.
Treat the warehouse (orders, refunds, ad spend) as the single source of truth. Platform dashboards are inputs, not the ledger.
How Consent Choices and Signal Loss Distort Client-Side Campaign Data
Under CCPA/CPRA, opt-outs remove client-side events entirely. Without Consent Mode and server-side capture, your ROI denominator (spend) is complete while your numerator (measured revenue) is not-mathematically guaranteeing understated ROI on consent-heavy segments.
Why Last-Click Attribution Over-Credits Bottom-Funnel Campaigns
Last-click gives 100% of credit to the final touch-usually branded search or a retargeting ad that would have converted anyway-while upper-funnel Meta or TikTok prospecting that created demand gets nothing. Data-driven attribution distributes credit probabilistically, but it still only models what was tracked; it is a better starting point, not a verdict.
Run branded and non-branded search as separate reporting lines. Blended branded search reliably inflates paid ROI on brands with strong organic demand.
How Incrementality Testing Reveals the Lift Attribution Misses
The only defensible answer to “would this have happened anyway?” is an experiment. Geo holdouts-pausing a channel in matched US DMAs while leaving others live-and on/off tests over 2-4 weeks expose true lift. Compare holdout revenue against control to calculate incremental profit, then substitute it into the ROI formula. Many brands find retargeting lift far smaller than platform reports suggest, while email and SMS via Klaviyo outperform modeled credit.
How to Adjust ROI for Returns, Discounts, and Fulfillment Costs
Apparel and footwear routinely see return rates in the 20-30% range (an estimate, not a benchmark for every category). Failing to net returns out of revenue turns a losing campaign into a “winner.” Adjust the numerator three ways: subtract refunded revenue, deduct discount cost, and allocate fulfillment per order.
Adj. revenue = gross rev - refunds - discounts
Adj. contribution = adj. revenue × margin - fulfillment
Campaign ROI = (adj. contribution - spend) ÷ spend
Example: ZAR 50,000 rev, 20% returns, 40% margin,
ZAR 4,000 fulfillment, ZAR 12,000 spend
→ (40,000 × 0.40) - 4,000 = 12,000
→ (12,000 - 12,000) ÷ 12,000 = 0% ROI (break-even)
Without the returns adjustment this example looks like a 2.4x ROAS winner. With it, the campaign is unprofitable.
A Repeatable ROI Analysis Cadence for Marketing Campaigns
Prebo Digital houses this in a Strategy → Build → Test → Scale → Report loop. Weekly: reconcile GA4 revenue against Shopify, Stripe, and platform reports; flag variance over 10%. Monthly: publish MER and CAC payback beside channel ROI. Quarterly: run one structured incrementality test per major channel and reset break-even ROAS targets from updated margin data. Keep the pipeline in services review with your data team so ETL jobs do not silently break parity.
Which ROI Framework Fits Your Business Stage
| Stage | Primary Metric | Test Cadence | Priority |
|---|---|---|---|
| Early DTC (under ZAR 1M/mo) | Break-even ROAS + MER | Monthly spot checks | Tracking accuracy + CRO first |
| Scaling brand (ZAR 1-10M/mo) | Channel ROI + CAC payback | Quarterly geo holdouts | Incrementality + returns adjustment |
| B2B/subscription | LTV:CAC (~3:1 rule of thumb) + payback | On/off channel tests | Warehouse reconciliation, HubSpot pipeline parity |
The common thread: no framework survives bad data. Whether you run Shopify, WooCommerce, or a B2B pipeline, the ROI number is only as good as the server-side events and warehouse joins beneath it. Teams that want the build handled end-to-end can review Prebo Digital’s approach.

