What is a Good Marketing ROI? Benchmarks by Channel

There is no universal good ROAS. A 3x return is excellent for a low-margin brand and poor for a high-margin SaaS business. Benchmarks only make sense when you adjust them for your margin structure and measurement methodology.

Marketing ROI benchmarks appear constantly in industry reports, but most of them are useless without context. A 4x ROAS measured by last-click attribution is a completely different number from a 4x ROAS measured by Marketing Mix Modelling. Understanding what a good return looks like for your category, margin, and measurement approach is more valuable than any published benchmark.

How to set your own ROAS floor

Before looking at benchmarks, calculate your break-even ROAS. Divide 1 by your gross margin percentage. If your gross margin is 40 percent, your break-even ROAS is 2.5x (1 divided by 0.4). Anything above that figure is profitable at the gross margin level. Anything below means marketing is destroying margin. Your target ROAS should be meaningfully above break-even to allow for operating costs and profit.

Benchmarks by channel

• Paid search (brand terms): 8x to 20x average ROAS on last-click, but much of this revenue would have come anyway. Incremental ROAS often 2x to 4x.
• Paid search (non-brand): 3x to 6x average ROAS, 2x to 4x incremental. Highly variable by category competitiveness.
• Paid social (prospecting): 1.5x to 3x incremental ROAS. Social excels at reach and awareness at costs below TV.
• Paid social (retargeting): 4x to 8x average ROAS, but incrementality is often low. Many of those converters would have come back anyway.
• Display and programmatic: 0.8x to 2x incremental ROAS. Most valuable as part of a full-funnel strategy rather than standalone.
• Email: very high reported ROAS (10x to 40x), but almost entirely non-incremental for established brands. Customers who receive emails were mostly going to buy anyway.
• TV and video (broad reach): 1.5x to 3x incremental over 12 months, with significant brand equity effects that compound over time.
• Affiliate and partnership: 5x to 12x reported ROAS, but heavy last-click bias. Incremental value varies widely by affiliate type.

Email marketing consistently reports the highest ROAS of any channel. It also consistently shows the lowest incrementality. If you removed all your email marketing, most of your email revenue would shift to other channels within 90 days. High average ROAS and high incremental value are not the same thing.

Why measurement methodology changes the benchmark

A 4x ROAS measured by last-click attribution is not comparable to a 4x ROAS measured by Marketing Mix Modelling. Last-click over-attributes to the final touchpoint and includes revenue that would have happened anyway. MMM measures incremental returns only. If your channel reports 5x on last-click and 2.5x on MMM, both numbers are accurate, but only the MMM figure tells you what the channel actually caused.

How to use benchmarks without misleading yourself

Use benchmarks as a sanity check, not a target. If your paid search non-brand ROAS is 12x on last-click, that is suspiciously high and worth investigating for attribution inflation. If it is 1x, you are likely losing money. The benchmark range gives you a corridor of plausibility. Your actual target should be derived from your margin structure and the specific conditions of your market.

The brand investment exception

Brand-building channels often show poor short-term ROAS and excellent long-term returns. TV, sponsorship, and broad video campaigns raise brand awareness and purchase intent, which then converts through lower-funnel channels weeks or months later. If you evaluate these channels on a 30-day ROAS window, they will always look inefficient. Evaluate them on a 12 to 24 month horizon, capturing the downstream effect on search volume, direct traffic, and conversion rates.