ROAS Benchmarks
Typical Return on Ad Spend ranges by vertical and channel — apparel 2-3x, beauty 3-4x, supplements 2-3x, high-AOV electronics 3-5x — with guidance on diagnosing under-performance.
ROAS Benchmarks
Industry-specific Return on Ad Spend ranges used to judge whether a channel, campaign, or creative is performing in line with peers.
ROAS benchmarks describe the revenue-per-euro-of-ad-spend ranges that comparable online stores typically achieve on paid channels. They differ sharply by vertical, average order value, channel, and funnel stage — a 2.5x prospecting ROAS that flags trouble for a supplements brand can be a healthy result for an apparel store with thin margins on first-order discounts.
Benchmarks are not targets in themselves; they are a reference frame. A store sitting two bands below its peer median almost always has a fixable problem — creative fatigue, audience mismatch, a leaking checkout, or attribution gaps — rather than a structural ceiling. Read the ranges below as starting points, then layer in your contribution-margin reality.
ROAS is the single most-cited number in paid acquisition, and also the most-misread. The headline figure hides where the revenue came from, what margin sat behind it, and whether the same buyers would have arrived organically anyway. Benchmarks help you separate "my channel is broken" from "this is just what the category looks like."
The ranges that follow assume blended paid social and search activity for stores between €1M and €15M in revenue. They reflect last-click or platform-reported ROAS — the number you see in Meta Ads Manager or Google Ads — not contribution-margin ROAS, which is typically 30-50% lower once COGS, shipping, and returns are deducted.
Typical platform-reported ROAS by vertical and funnel stage
| Vertical | Prospecting ROAS | Retargeting ROAS | Blended ROAS |
|---|---|---|---|
| Apparel & fashion | 1.5x – 2.2x | 4.0x – 6.0x | 2.0x – 3.0x |
| Beauty & skincare | 2.0x – 2.8x | 5.0x – 8.0x | 3.0x – 4.0x |
| Supplements & wellness | 1.4x – 2.0x | 4.0x – 6.0x | 2.0x – 3.0x |
| Electronics (high AOV) | 2.0x – 3.5x | 5.0x – 9.0x | 3.0x – 5.0x |
| Home & furniture | 1.8x – 2.5x | 4.5x – 7.0x | 2.5x – 3.5x |
| Food & beverage (DTC) | 1.2x – 1.8x | 3.5x – 5.0x | 1.8x – 2.5x |
Two patterns repeat across the table. First, retargeting ROAS is always 2-3x prospecting — if it isn't, your retargeting audiences are too cold or your prospecting is over-claiming. Second, verticals with high repeat rates (beauty, supplements) tolerate lower prospecting ROAS because the second and third orders rescue the unit economics. Apparel and food, with weaker repeat behaviour, need closer-to-breakeven first orders.
Median blended ROAS by paid channel
How to read your ROAS against the band
Start by isolating branded search. It almost always reports a ROAS of 6-10x and inflates the blended figure by 0.5-1.5 points. If you strip it out and your non-brand paid ROAS is still inside the benchmark band, the channels are healthy and you should focus elsewhere — site speed, AOV, or retention.
If non-brand ROAS sits a full band below the table, the issue is almost always upstream of the ad account. The three usual suspects: creative that hasn't refreshed in 6+ weeks, an audience that has saturated (frequency above 3.5 on Meta), or a landing page converting at less than 60% of the site average. A historical funnel audit usually surfaces which one in under an hour.
Platform ROAS is not margin ROAS
A 3.0x platform ROAS on a beauty SKU with 65% gross margin and 8% return rate nets roughly 1.8x on contribution. On apparel with 55% margin and 25% returns, the same 3.0x nets closer to 1.2x — barely breakeven once CAC and shipping are paid. Always benchmark after you've translated to the metric that actually pays salaries.
When a low ROAS is actually fine
Three scenarios justify a deliberately sub-benchmark ROAS. First, a launch period where you're buying first-order data to train the algorithm — expect 1.2-1.6x for 4-8 weeks. Second, a subscription or replenishment product where LTV-to-CAC is the real metric and a 1.5x first-order ROAS still produces a 4x payback by month nine. Third, a category where you have a structural margin advantage (own manufacturing, vertical integration) and can outbid competitors profitably.
What rarely justifies a low ROAS is "brand building". If a campaign is genuinely brand-led, measure it on assisted conversions, branded search lift, and direct traffic — not ROAS. Mixing brand goals into a performance line item is how reporting becomes unreadable and budgets get cut by the CFO who can't see the upside.
ROAS benchmark questions
A blended ROAS of 3.0x-4.0x is healthy for most Shopify stores in apparel, beauty, and home categories. Stores with high AOV (€150+) or strong repeat purchase economics can run profitably at 2.5x; thin-margin or heavily-discounted catalogues often need 4.0x or more to clear contribution-margin breakeven.
Three forces are compressing Meta ROAS across the board: CPM inflation of 15-25% year-over-year, iOS attribution undercounting of roughly 20-30% of true conversions, and audience saturation in mature accounts. A 20% drop in reported ROAS often masks flat real performance — check Shopify's own attribution and post-purchase surveys before cutting spend.
ROAS divides revenue from a specific channel by spend in that channel. MER (Marketing Efficiency Ratio) divides total store revenue by total marketing spend across all channels. MER is harder to game with attribution choices and is increasingly the metric Performance Managers steer by — a target MER of 3.5-4.5x is typical for healthy DTC stores.
Contribution margin, always — but ROAS is the daily steering signal. Set a ROAS floor that, given your gross margin and return rate, produces positive contribution. For most apparel stores that floor sits around 2.0x; for beauty closer to 1.8x; for low-margin food categories often 2.5x or higher.
TikTok typically reports 1.5x-2.5x platform ROAS for prospecting, lower than Meta on a like-for-like basis. The view-through and brand-lift contribution is real but undercounted by TikTok's own pixel. Most stores running TikTok seriously see a 0.3-0.5x lift in blended ROAS or MER within 90 days, even when the TikTok ad account itself looks underwhelming.
Plan for 2-4 weeks of learning-phase volatility on Meta and 3-6 weeks on Google Shopping. Campaigns that haven't reached benchmark by week 8 usually have a structural issue — creative, offer, or landing page — rather than an algorithm one. Killing campaigns inside week one is the most common cause of permanently weak account performance.
Some of it is theatre — retargeting often claims credit for users who would have bought anyway. Industry incrementality tests typically show 30-50% of reported retargeting revenue is genuinely incremental. That still makes it the most profitable line in most accounts; just don't budget as if 100% would disappear without it.
Use category benchmarks halved for the first 30 days, then category benchmarks at 75% for days 30-60, then full benchmark by day 90. For an apparel launch, that's roughly 1.0x → 1.5x → 2.0x prospecting. If you're not on the trajectory by day 45, the creative or offer is the problem, not the algorithm.
Yes, though less than in 2022. Meta's Advantage+ and Conversions API recover most of the signal loss for well-instrumented stores, but expect a persistent 10-20% undercount versus server-side or post-purchase-survey truth. Compare platform ROAS to MER monthly to keep the gap honest.
Email and SMS through Klaviyo or similar typically report 20-40x ROAS because the "spend" is just the platform fee. Treat it as a different metric class — measure revenue per recipient (€0.10-€0.30 is healthy for ecom flows) rather than ROAS, which makes the channel look unrealistically dominant in blended reports.
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.