There Is No ROAS Benchmark. I Went Looking For One.

By Brennan Lunin, Founder at BL AdWorks

I set out to update this post with real 2026 ROAS benchmarks by platform. I wanted a table: Google Search does this, Meta does that, here is what good looks like in your industry.

I could not build it honestly, and the reason turns out to be more useful than the table would have been.

What I found instead

Search “average ROAS by platform” and you get a dozen confident pages. Here is what three of them say Google Ads averages:

• One says 4.2x
• One says 3.7:1
• One says 3.52x

Same platform. Same year. Three numbers, no overlap in methodology, no primary source under any of them. Ecommerce customer acquisition cost is worse: I found “$68 to $84 average” on one page and “$274, rising to $318” on another. Those are not different estimates of the same thing. They are not even the same order of magnitude.

None of these pages cite a study. Several cite each other. It is a closed loop of content quoting content, and if you follow it down you never reach a floor.

Why the number does not exist

ROAS is revenue divided by ad spend. That is the whole formula. Which means a “good” ROAS depends entirely on what a dollar of your revenue is worth to you, and nobody publishing a benchmark table knows that.

Consider two stores, both running at exactly 3.0x.

The first sells a physical product with a 30% gross margin. At 3.0x, every $1 of ad spend brings back $3 of revenue, of which $0.90 is gross profit. Subtract the $1 you spent. That store is losing ten cents on every dollar it advertises. It is scaling itself out of business, and the dashboard says 3.0x.

The second sells software with an 85% margin. Same 3.0x, and $2.55 of every $3 is gross profit. That store is printing money.

Identical ROAS. Opposite businesses. Any benchmark table that does not ask for your margin is not describing your business, and it cannot, because it does not know it.

The number you actually need

Your breakeven ROAS is one divided by your gross margin.

• 30% margin → you break even at 3.33x
• 50% margin → you break even at 2.0x
• 70% margin → you break even at 1.43x
• 85% margin → you break even at 1.18x

Anything above that line is profit before overhead. Anything below it is a subsidy you are paying to your customers.

That is the only benchmark that means anything, and you can calculate it in about fifteen seconds with a number you already have. It does not need a study. It needs your P&L.

Then there is a second number, which is what you can *afford* to do. Breakeven ROAS assumes the customer buys once and disappears. If a real customer buys three times over two years, you can go below breakeven on the first order on purpose, because you are buying a relationship rather than a transaction. That is a strategic decision, and it is the correct one for a lot of stores. But it is only correct if the second and third purchases actually happen, which is a retention question, not an ads question.

What is actually measurable

Some things about ad platforms *are* published, by people who show their work.

WordStream publishes Google Ads benchmarks from real accounts, and they say so: their 2026 numbers come from roughly 13,000 search campaigns across 23 industries, running April 2025 through March 2026. That data gives an average cost per click of $5.42, an average conversion rate of 8.18%, and an average click-through rate of 6.64%. They also found cost per click rose for 87% of industries.

Notice what is in that list and what is not. Cost per click, click-through rate, conversion rate: all published, all sourced, all real. ROAS is not there. Not because they forgot, but because ROAS depends on the revenue side, and the revenue side is yours.

That is the tell. The people with the actual data do not publish a ROAS benchmark. The people with no data publish it confidently.

So how do you know if your ads are working

Three questions, in order.

Is my tracking honest? If the conversion tag is misfiring or double-counting, every number below this is fiction and you will optimize with confidence toward nothing. Check this first, always.

Am I above my breakeven ROAS? The one you calculated above, from your own margin. Not a number from a blog.

Is my ROAS going up or down against my own last 90 days? This is the benchmark that actually matters, and it is the one nobody can sell you: you, last quarter. Your margin, your AOV, your attribution window, your customers. It is the only comparison where the two sides of the equation are measuring the same thing.

What I am not going to do

I am not going to give you a table. I looked for the data to build one honestly, and it does not exist. Publishing one anyway would make this post rank better and would make you dumber, and I would rather have the second thing.

If you want a number to aim at, aim at one divided by your gross margin, and then beat your own last quarter.

One more thing

The stores that obsess over ROAS are usually the ones with nothing else running. When the only way to make a sale is to buy a stranger, the ad account carries the whole business, and a bad month in the ad account is a bad month, full stop.

The stores that are relaxed about ROAS usually have a list. They can afford a mediocre acquisition quarter because a meaningful share of revenue comes from people who already bought. If your ROAS anxiety is high, the fix might not be in the ad account at all.

Sources: WordStream 2026 Google Ads Benchmarks (~13,000 campaigns, 23 industries, April 2025 to March 2026).