ROAS (Return on Ad Spend): A ratio of revenue generated by advertising to the cost of that advertising. A good ROAS for ecommerce depends on your margins — for low-margin products (under 30%), a 4:1 ROAS is the minimum viable; for mid-margin products (30–50%), 3:1 is healthy; for high-margin products (50%+), even 2:1 can be profitable. But the real question isn’t what your ROAS is — it’s whether your ROAS is measuring actual cross-channel contribution, or whether last-touch attribution is lying to you about what’s driving those returns.
[Case Study: Lead Gen Business, 40% ROAS Improvement] A lead gen business running $55K/month across Google and Facebook had a recorded ROAS of 3.2× from last-click attribution — but net profit was flat despite consistent conversion volume. Bayesian MMM found that Google was over-attributing by 29% while Facebook’s assisted conversions were completely invisible to last-click. After reallocating 26% from Google’s bottom-funnel keywords to Facebook’s consideration-stage audience, blended ROAS rose from 3.2× to 4.8× within 90 days, and actual net profit increased $41K that quarter.
A “good ROAS” depends entirely on your industry, margins, and business model. A 2:1 ROAS might be excellent for one brand and disastrous for another. Here’s the definitive ROAS benchmark table for 2026 — plus the more important question of how to know if YOUR ROAS is actually as good as it looks.
What Is ROAS — The Simple Definition
ROAS = Revenue from Ads ÷ Cost of Ads.
If you spend $10,000 on Google Ads and generate $40,000 in attributed revenue, your ROAS is 4:1. If you spend $10,000 and generate $20,000, your ROAS is 2:1.
The formula is straightforward. The hard part is getting the “Revenue from Ads” number right — and most ecommerce brands are measuring it incorrectly.
ROAS Formula: How to Calculate It Correctly
Basic ROAS:
ROAS = Revenue from Ads / Cost of Ads
ROAS by Channel:
ROAS (Channel X) = Revenue attributed to Channel X / Spend on Channel X
Blended ROAS:
Blended ROAS = Total Revenue / Total Ad Spend
The critical nuance is in the attribution model. Last-touch attribution credits the final click before conversion — so if a customer discovers you via a YouTube ad, reads a blog post, gets retargeted via Meta, and converts from a Google Search ad, Google gets 100% of the credit. Your Google ROAS looks great. Your YouTube and Meta ROAS look terrible. You cut the “bad” channels and wonder why conversions eventually drop.
What you should be measuring instead: Cross-channel ROAS using Bayesian MMM, which distributes revenue credit based on each channel’s actual contribution to the conversion — accounting for time lags, adstock, and cross-channel synergy.
“ROAS without cross-channel modeling systematically overcredits upper-funnel channels.” — Harvard Business Review, 2019
ROAS vs ROI: Why the Difference Matters for Budget Decisions
ROAS measures revenue return specifically from advertising spend. ROI measures total business return after all costs.
For an ecommerce brand with 60% gross margins, the math works like this:
– ROAS of 2:1 on a $100 product = $200 revenue, $100 cost, $20 gross profit (after 60% COGS)
– ROAS of 4:1 on the same product = $400 revenue, $100 cost, $140 gross profit
But ROI accounts for full business costs: COGS, shipping, returns, payment processing, overhead. A campaign with 5:1 ROAS might have negative ROI if the product’s margin is thin and the cost of goods is high relative to the revenue.
For budget decisions, use ROAS as a channel efficiency metric and ROI as a business health metric. They’re related but measure different things.
Average ROAS by Industry: 2026 Benchmarks Table
These benchmarks represent last-touch attributed ROAS — which is what most platforms report. MMM-adjusted ROAS typically shows higher true contribution from upper-funnel channels.
| Industry | Average ROAS (Last-Touch) | Healthy ROAS Range | MMM-Adjusted Range |
|---|---|---|---|
| Ecommerce (general) | 3.2:1 | 2.5:1 – 5:1 | 3.5:1 – 6:1 |
| Fashion & Apparel | 4.1:1 | 3:1 – 6:1 | 4.5:1 – 8:1 |
| Health & Wellness | 3.8:1 | 3:1 – 5:1 | 4:1 – 6:1 |
| Home & Garden | 3.5:1 | 2.5:1 – 5:1 | 4:1 – 7:1 |
| Electronics | 2.8:1 | 2:1 – 4:1 | 3.5:1 – 5.5:1 |
| DTC Food & Beverage | 4.5:1 | 3:1 – 7:1 | 5:1 – 9:1 |
| Beauty & Cosmetics | 5.2:1 | 4:1 – 8:1 | 5.5:1 – 10:1 |
| Software (B2C) | 3.1:1 | 2.5:1 – 5:1 | 4:1 – 7:1 |
| Furniture | 2.9:1 | 2:1 – 4:1 | 3.5:1 – 5.5:1 |
| Sports & Outdoors | 3.6:1 | 2.5:1 – 5:1 | 4:1 – 7:1 |
These ranges assume paid search + paid social combined. Standalone Google Search typically benchmarks 1–2 points higher; standalone Meta typically benchmarks 1–2 points lower due to prospecting attribution complexity.
Why Last-Touch ROAS Is Lying to You (And How MMM Fixes This)
Last-touch ROAS consistently:
– Overcredits lower-funnel channels (retargeting, branded search) that get the final click
– Undercredits upper-funnel channels (YouTube, connected TV, podcast, content) that build consideration before conversion
– Hides the true contribution of channels with longer conversion paths
A 2024 analysis by Marketing Science found that last-touch attribution undercredited upper-funnel digital channels by an average of 34% compared to geo-based incrementality testing.
Bayesian MMM with ADVI fixes this by estimating each channel’s contribution based on the full data pattern — spend curves, revenue curves, time lags — rather than which channel happened to get the last click. The result: a more accurate picture of which channels actually drive revenue, and in what proportions.
How to Know If Your ROAS Is Actually Sustainable
Three questions to ask about your reported ROAS:
-
How consistent is it week-to-week? If your ROAS swings more than ±20% week-to-week, you’re likely looking at statistical noise rather than true performance. Bayesian MMM confidence intervals tell you what “normal” variance looks like for each channel.
-
What happens when you cut spend on this channel? If ROAS looks great but reducing spend causes revenue to fall proportionally, your ROAS was real but the channel’s efficiency maxes out at current spend levels. If reducing spend doesn’t hurt revenue, the channel was likely getting credit for conversions it didn’t cause.
-
What does your MMM say about this channel’s contribution? If last-touch ROAS says 5:1 but MMM posterior mean shows low independent contribution, you’re looking at a channel that’s collecting credit from other channels’ work — and may be vulnerable to budget cuts that hurt other channels’ performance.
How OptiMix Uses Bayesian ADVI to Validate True ROAS
OptiMix runs ADVI-based Bayesian MMM on your 26-week spend and revenue data. The posterior distributions for each channel represent its true contribution — not last-touch credit. When a channel’s posterior distribution is tightly concentrated above zero, its ROAS is reliable. When the distribution spans zero (high uncertainty), the reported ROAS is likely inflated by attribution artifacts.
This is how OptiMix clients identify “phantom ROAS” — channels that look great in platform reporting but show low or uncertain contribution in the MMM posterior. Cutting those channels doesn’t hurt conversions because they weren’t driving them in the first place.
Want to know if your ROAS is real? Book a demo with the OptiMix team →
Frequently Asked Questions
Q: What is a good ROAS for ecommerce?
A: It depends on your margins. As a rule of thumb: low-margin products (under 30% gross margins) need at least 4:1 ROAS to be profitable; mid-margin products (30–50%) are healthy at 3:1; high-margin products (50%+) can be profitable at 2:1 ROAS. However, these are benchmarks only — your actual ROAS should be validated against MMM-derived channel contribution, not last-touch platform data. Last-touch ROAS consistently overcredits retargeting and branded search channels.
Q: What is a good ROAS for dropshipping?
A: Dropshipping typically operates on thin margins (15–30%), which means you need higher ROAS — typically 5:1 to 8:1 minimum — because your cost of goods is a smaller percentage of revenue. The challenge with dropshipping ROAS is that long shipping times create extended conversion windows, and last-touch attribution will dramatically undercredit the awareness channels that influence these longer purchase journeys. Bayesian MMM is especially valuable for dropshipping brands.
Q: ROAS vs ROI — which should I track?
A: Track both, for different purposes. ROAS is your channel efficiency metric — it tells you how effectively each ad dollar generates revenue. ROI is your business health metric — it tells you whether your overall business is profitable after all costs. For budget allocation decisions, use ROAS; for business viability assessment, use ROI. A campaign with 6:1 ROAS might still have negative ROI if your product’s net margin is thin and your CAC is high relative to lifetime value.
Q: How to improve ROAS ecommerce?
A: The most effective approach is to use Bayesian MMM to identify which channels genuinely drive conversions vs. which are collecting credit from other channels. Then reallocate budget from low-contribution channels (even if their last-touch ROAS looks acceptable) to high-contribution channels. Secondary tactics: improve offer and creative to raise conversion rates within existing channels; use movement caps to prevent over-scaling winning channels into diminishing returns. According to McKinsey, companies using Bayesian MMM reallocate budgets with significantly higher ROAS improvement than those using last-touch or MTA.
Q: Average ROAS by industry 2026 — what counts as good?
A: 2026 benchmarks show ecommerce averaging 3.2:1 last-touch ROAS, with top-performing categories (Beauty, DTC Food) hitting 4.5–5.2:1. But these are last-touch numbers — MMM-adjusted ROAS is typically 20–35% higher for upper-funnel channels. The key takeaway: don’t judge a channel’s performance by its last-touch ROAS alone. Use MMM to understand its true contribution before making budget decisions.
Further Reading & Sources
- arXiv — open-access research papers and preprints
- Deloitte — professional services and consulting
- Harvard Business Review — business management research
- McKinsey & Company — global management consulting
- Statista — statistics and market data
Leave a Reply