Catalog ad (DPA) ROAS: what is good and how to measure it

You run catalog ads (also called dynamic product ads, or DPA) on Meta. You look at ROAS every day. Then you ask the question every store owner asks: is this number good?

This page will not give you a benchmark. Meta does not publish a ROAS benchmark for catalog ads that we could read. Numbers you see online come from other stores, with other margins. Your own margin tells you far more. Here is how to use it.

The short answer

There is no single good ROAS. A number that makes one store rich can lose another store money. The difference is margin.

So start with your break even ROAS. That is the ROAS where your ads pay for themselves and no more. Above it, your ads make money. Below it, they lose money. Once you know that line, every ROAS number means something.

What ROAS means

ROAS is return on ad spend. You take the revenue your ads brought in and divide it by what you spent on them.

For example, say you spend $1,000 on ads in a month. Meta says those ads led to $3,000 in sales. Your ROAS is $3,000 divided by $1,000, which is 3x. You got $3 back for each $1 you spent.

That is just arithmetic. These are made up round numbers to show the sum, not a result from any store.

ROAS counts revenue, not profit. A 3x ROAS does not mean you made $2,000. You still paid for the goods, the shipping and the fees. That is why margin matters so much.

How to find your break even ROAS

Break even ROAS is 1 divided by your gross margin. Gross margin is the share of each sale you keep after the cost of the goods.

For example, if you keep 40% of each sale, your break even ROAS is 1 divided by 0.4. That is 2.5x. At 2.5x, your ads pay for themselves. At 3x, they make money. At 2x, they lose money.

Here is how to work out your own:

1. Pick a typical product, or use your store's average.
2. Take the price a shopper pays.
3. Subtract what the product costs you.
4. Divide what is left by the price. That is your gross margin.
5. Divide 1 by that margin. That is your break even ROAS.
6. Write it down and read every ROAS number against it.

A store with thin margins needs a high ROAS just to break even. A store with fat margins can run at a lower ROAS and still profit. That is why two stores can see the same number and feel very different about it.

Why catalog ad ROAS looks different

Catalog ads can do two jobs. Meta says you can use them to retarget shoppers, or to find new customers. You can read more in Advantage+ catalog ads explained.

These two jobs give very different numbers.

• Retargeting shows products to people who already looked at them. They are close to buying, so ROAS often looks high.
• Prospecting shows products to people who have not been to your store. They need more time, so ROAS often looks lower.

Do not compare the two as if they were the same thing. Read each one against your break even line on its own.

Attribution also changes the number. Meta credits a sale to an ad inside a set window after a click or a view. A longer window counts more sales, so ROAS reads higher. A shorter one reads lower. Keep the same window each time you compare, or you are comparing two rulers.

How to read ROAS during learning

When you start a new ad or ad set, or make a big edit, Meta enters a learning phase. The Delivery column reads "Learning". Meta says that during learning, "ad sets are less stable and usually have a higher CPA".

So do not judge ROAS while an ad set is learning. The numbers swing, and cost per sale tends to run high. Meta's own advice is to "Wait to edit your ad set until it's out of the learning phase" and to "Avoid unnecessary edits".

Wait until learning ends. Then read ROAS over a few weeks, not a few days.

How to test a creative change fairly

Say you want to know if better images lift ROAS. You need a fair test, or the number will fool you.

• Keep the budget the same. A big budget change can put an ad set back into learning.
• Compare two periods of the same length. Use the weeks before the change and the weeks after.
• Do not edit the ad. Meta lists "Any change to ad creative" as a significant edit. So a new image, text, frame or overlay on the ad restarts learning.
• Change one thing. If you change the images and the audience at once, you will not know which one worked.

This is where the feed helps. Convinci changes the images in your catalog feed, not the ad. Meta does not list catalog item updates as a significant edit. Meta also does not promise they never matter. But your ad sets stay as they are, and the ad is not edited. Read more in does changing catalog images reset learning.

If you want to learn why plain cards tend to underperform, read why dynamic product ads look bad.

The Convinci 50% ROAS lift guarantee

Convinci turns your plain catalog photos into branded ad cards. They show your store name, price, sale badge and reviews. Our offer is plain: a 50% lift in ROAS in 30 days or your money back.

Here is how it works:

• If your ROAS is not up 50% in 30 days, we refund what you paid Convinci.
• We do not refund your ad spend.
• You press Refund me in your account. No proof needed.
• You keep your images.

You can see your own products as cards first. The preview on the home page is free, with no account and no card. Plans are on our pricing section. Free covers 10 product photos. Pro is $99 a month for up to 1,000 photos.

If your ROAS is low and clicks do not turn into sales, work through our catalog ads not converting checklist.