How Should DTC Brands Set ROAS Targets by Channel and Buyer?

Your account runs one 3x ROAS target across every campaign, and every month the prospecting campaigns miss it while retargeting beats it. So budget drifts toward people who were already going to buy, and your count of new customers goes flat.

Set your targets by working back from what each type of customer is worth, then adjust for how much of each channel's reported revenue is real. You end up with a lower target for campaigns that find new customers, a higher one for campaigns that reach existing customers and brand searchers, and a written rule for each.

Running one blended number is like paying your salespeople the same commission whether they open a new account or take a reorder from a customer who calls in anyway. Platform details were checked on 28 October 2026.

Why is one ROAS target wrong for most DTC brands?

One ROAS target is wrong because it rewards the campaigns that reach people who were already going to buy, and starves the campaigns that bring in new customers.

The gap between those two is larger than you might expect. In Common Thread Collective's Q1 2026 benchmark of 299 DTC brands, Meta prospecting campaigns returned a median 1.83x in-platform, against 5.88x for retargeting and retention, both on 7-day click. Put a blended 3x target across both and your prospecting looks like a failure while retargeting looks like the hero.

Google itself warns about the cost. Its target ROAS guidance says that setting a target that's too high may limit the amount of traffic your ads may get. Hold a prospecting campaign to a retargeting-level target and it spends less, finds fewer new customers and stops feeding your customer file.

Our position is simple. A prospecting campaign below the blended target is often the most profitable spend in your account. It only looks weak because the platform report counts the first order and ignores the second.

Split your campaigns into two lists, new customers and everything else, before you judge either one.

What should the target be for new customers?

The target for new customers should come from what a new customer contributes over a payback window, which sets the most you can pay to acquire one, and we recommend a window of 60 to 90 days.

Start from breakeven on the first order, set out in what is a good ROAS for a DTC brand. Then add the margin from repeat orders that arrive inside your window.

A short window keeps you honest, because most reorders come early. In BS&Co's agency dataset of 156,110 customers across several DTC brands, about three in four of the customers who bought again placed their second order within 90 days. Common Thread's MER guide also builds a 60-day version of its acquisition metric. A 90-day window catches those early reorders without betting your budget on a year of loyalty that may never arrive.

Measure your result on first-time customers only. Common Thread defines that metric, aMER, as first-time customer revenue divided by total ad spend. Shopify's analytics reports new and returning customers separately for any period, and its CAC formula divides all acquisition costs in a period by the new customers who bought in it.

Ask your analyst for those three numbers from the last six months before you set anything.

A worksheet for setting the new-customer target

This worksheet turns margin and repeat data into a target your media buyer can enter. The figures are illustrative, not from a CB/I client.

The last line is where your target is most likely to go wrong, because platforms over-claim or under-claim depending on the channel. Common Thread's worked example shows that if you need a 2.0 aMER, in a channel where only half of reported conversions are incremental, you'd need a 4x ROAS target in-platform. Divide your business target by your incrementality factor, taken from your own test.

If you don't have a test yet, treat the factor as unknown and run one first, as set out in when a DTC brand needs incrementality testing.

How should targets differ for returning customers and brand search?

Returning-customer and brand campaigns should carry a higher target than prospecting, because a large share of the sales they report would have happened without the ad.

Retargeting and retention

Retargeting reaches people who already know you. Common Thread's incrementality factors put Meta's non-acquisition campaigns at 0.60, which means a reported 5.88x is closer to 3.5x in real terms. Hold these campaigns to at least breakeven divided by their incrementality factor, and cap their share of your budget so they can't crowd out your prospecting.

Brand search

The evidence on brand search depends on whether competitors bid on your name. In eBay's field experiment, later published in Econometrica, brand-keyword ads had no measurable short-term benefits. A larger Bing study across thousands of brands found brand ads added only 1% to 4% in clicks when no competitor was present. Where competitors bid and the brand didn't, the competitors took 18% to 42% of its clicks.

So brand search is cheap insurance where rivals bid on your name, and mostly paid-for organic traffic where they don't. Give it its own campaign and its own high target or a fixed budget, so its easy revenue doesn't flatter your prospecting numbers.

How do Google and Meta let you split targets by customer type?

Both platforms can separate new and existing customers, and both need the brand's own customer list to do it well.

Google Ads

Google's new customer acquisition goal can bid higher for new customers than existing ones, and Google recommends this New Customer Value setting for all advertisers with purchase conversion goals. A stricter New Customer Only setting bids only for new customers. Without your own list, Google detects existing customers from up to the last 540 days of campaign activity and purchases, which it calls less comprehensive. Upload your Shopify customer file through Customer Match, since Google warns that a list under 1,000 members can hold the goal back.

One catch matters when you read the reports. Google adds the value you assign to a new customer to reported conversion value, so reported ROAS rises when the goal is switched on, even if revenue doesn't. Ask to see the New customer lifetime value and Original conv. value columns before you compare periods.

Meta

If you used Meta's existing-customer budget cap, it's gone for new Advantage+ campaigns. Its developer documentation says that to replicate it, you create two ad sets per campaign that separate existing customers from new ones, using a custom audience of existing customers. That split is what lets you give each ad set its own target.

Questions to ask your media buyer or agency

These questions show quickly whether your targets are built on the business or on the platform report.

1. What is our breakeven ROAS, and how did you calculate the margin behind it?
2. Which campaigns are judged on new-customer revenue, and what is their target?
3. Have you uploaded our customer list to Google and Meta, and when was it last refreshed?
4. Is the new customer acquisition goal on in Google Ads, and are you showing me ROAS with or without the added new-customer value?
5. What attribution window sits behind each ROAS number you report?
6. When did we last test incrementality on our largest channel, and what factor did it give?
7. What share of spend goes to retargeting and brand search, and what caps that share?

A good answer to the sixth question is a date and a number, and a red flag is "the platform says it's working." Customers who first find you in ChatGPT or Google AI Mode often arrive without any ad click, which is one more reason for judging the whole program on store data rather than on platform claims. Ask these before you renew a contract, not after.

Frequently asked questions

What is a good new-customer ROAS?

There is no universal number. It is the first order value divided by the most you can pay for a new customer, which depends on margin and how many customers reorder within 90 days.

Should we exclude existing customers from prospecting campaigns?

Exclude them or separate them, so prospecting is judged on new customers only. In Meta that now means separate ad sets, and in Google the new customer acquisition goal with an uploaded customer list.

How often should ROAS targets change?

Review them monthly against new-customer revenue and contribution margin, and reset them when margin, repeat rate or an incrementality test result changes. Google advises lowering a target ROAS gradually when you need more volume.