Meta's New-Customer Setting Is Now Open to All Advertisers. Audit Your Exclusion List Before You Trust It.

If you run Meta Sales campaigns, open one ad set this week and look at the Customer Lifecycle Strategy setting. If you plan to switch any campaign to “acquire new customers,” check the exclusion list Meta fills in for you before you spend against it. Then judge the result on blended revenue and new customer count, not on in-platform ROAS. That is the whole recommendation. The rest is why.

What changed

At Advertising Week New York on October 6, 2026, Meta announced updates that trade coverage reports as making Customer Lifecycle Strategy available to all advertisers. The setting sits at the top of a Sales ad set and gives you a choice: reach new and existing customers (the default), or acquire new customers only. When you pick new only, Meta recommends excluding people it can identify as existing customers.

A few details are worth separating from the headlines. Meta’s own post was not available for this article, so what follows comes from trade reports and from hands-on testing by independent media buyer Jon Loomer, and should be read as reported rather than confirmed by Meta.

• The feature is not brand new. Screenshots circulated in the spring, and Loomer documented it in April and again in June. October’s news is the broader rollout.
• It is not an AI targeting system. Loomer tested it and found no hidden data source. His read: it is Meta surfacing a second place to exclude your custom audiences. Reported testing also found that audience labels and predefined audience segments did not change what got excluded.
• Meta reportedly defines existing customers through the Existing Customers segment in your Advertising Settings, which is the 2026 successor to the old existing customer budget cap in Advantage+ Shopping.
• The older “Exclude these custom audiences” field under Audience Controls has been relabeled “Legal and operational custom audience exclusions.” Its tooltip tells you to use Customer Lifecycle Strategy for existing customers.

The problem: the list you did not build

Loomer’s follow-up testing is the part operators should care about. After he selected the new customer option, Meta added custom audiences to the existing customer exclusion list on its own. In his account the list was incomplete and included audiences that did not match his actual customers. The engaged audience field was left blank. His screenshots are dated August 30, 2026, so this is one account’s result, not a guarantee about yours.

Why it matters: the setting only protects you as well as the list is built. If your best past-buyer list is missing, you pay to reach repeat customers and call it acquisition. If the list is padded with the wrong audiences, you cut off prospects you wanted. Either way the report says “new customers” and the spend does something else.

What to expect from the numbers

Meta’s own guidance, as relayed through a vendor write-up of the early-access version, is that excluding past buyers can make costs look higher, that CPA will likely rise, and that in-platform ROAS will drop. That is not a bug. Repeat buyers convert cheaply and often. Remove them and the average gets worse while the work gets more honest.

Here is an illustrative example, with made-up numbers to show the mechanism. Say you spend $50,000 a month on a Sales campaign.

• Today: 1,000 purchases at a $50 CPA. Suppose 350 of those are past buyers who would likely have bought anyway.
• After excluding past buyers, assume the same $50,000 buys 700 purchases, all new. CPA becomes about $71 ($50,000 divided by 700).
• In-platform ROAS looks about 30% worse (700 purchases against 1,000 at the same order value).
• But new customer purchases went from 650 to 700. Repeat revenue did not vanish, it just stopped being credited to your ad spend. If your email and SMS programs catch those repeat buyers anyway, you lost very little.

The real question is whether the 350 repeat buyers would have come back without Meta’s help. For a consumable with a 30 to 60 day reorder cycle, many would. For a one-time purchase category, few repeat at all and the setting may change almost nothing. Your own data decides this, not a rule of thumb.

The audit to run this week

1. Open an ad set and read the list. Switch to “acquire new customers” in a draft or duplicate, not a live campaign. Write down every audience Meta auto-selected.
2. Compare it to your real customer sources. You need purchasers from your pixel and Conversions API, plus your uploaded customer list from Shopify and Klaviyo. If any of those are missing, add them manually.
3. Check the engaged audience field. If you want to exclude people who engaged but never bought, you have to fill it in. Do not assume it is populated.
4. Verify your Existing Customers segment in Advertising Settings. The exclusion is only as accurate as that definition.
5. Do not mix legal and customer exclusions. Keep suppression lists you must honor in the legal and operational field, and past buyers in Customer Lifecycle Strategy, so the next person reading the ad set can tell them apart.
6. Set the success metric before you launch. Compare new customer count, blended MER (total revenue divided by total ad spend) and cost per new customer. Do not compare ROAS to last month’s.

When to leave it alone

Do not flip this on across the account right before peak. Every structural edit to a live campaign can reset learning, and Q4 CPMs are already the most expensive of the year. If you want to test, split one budget: run a duplicate ad set with new-only on a defined slice, hold the rest on default, and read results over at least two weeks of conversion data. If you are a brand with a strong repeat business, the default may be the right setting for most of your spend, with new-only reserved for a dedicated prospecting campaign.

Other items from the same week

Meta also said it is testing an agentic assistant that can plan campaigns, adjust targeting and change budgets, with explicit approval required before anything is published. Meta says assistant users typically see 5% more conversions at the same cost per conversion, but it did not publish the method or comparison group, so do not plan budgets around that number. Wider availability was described only as “soon.” TikTok announced Buy Direct checkout and GMV Max changes the same week, with no stated dates or markets, so there is nothing to act on yet.

Meta’s setting changes, TikTok’s checkout tests and your own Shopify and marketplace data rarely line up on their own. If you want an operator team to build the exclusion lists, set the right success metrics and read the blended numbers before Q4 budgets scale, book a call.