Meta's Customer Lifecycle Strategy: stop paying for customers you already have.

Most prospecting campaigns spend part of their budget on people who already bought. Meta's delivery hunts for the cheapest conversion it can find, and a past customer who was about to reorder is about as cheap as conversions get. The ad takes the credit. You'd have had the order anyway.

Meta now has a setting for this that every account can use. It does less than most of the coverage suggests, and how you judge it matters more than the toggle itself.

01 What is Meta's Customer Lifecycle Strategy?

Customer Lifecycle Strategy is a setting on Sales ad sets that tells Meta who the ad set is for. On the default, Meta can serve anyone, past buyers included. Mark the ad set as acquisition and Meta stops delivering to the audiences you exclude, starting with your existing customers. Its job is to keep acquisition budget away from people who already buy from you.

Meta's own announcement, published on October 6, 2026, is short. It says the setting is now available to all advertisers, and that marketers who flag a campaign as acquisition get recommended exclusions instead of building them by hand. Meta's developer documentation has the mechanics: the field sits on the ad set, works only with the Sales objective, and lists three modes.

The exclusions are ordinary custom audiences: customer lists, website purchasers, app events, offline events. Meta is also testing a simpler version that excludes a whole group of existing customers through one audience label. The acquisition modes are blocked on campaigns in restricted special ad categories, which Meta's documentation lists as housing, employment, credit, financial products and services, education, and online gambling and gaming.

A lot of what's been written about it overstates what it does:

02 Why acquisition budgets leak to existing customers

Meta optimizes for the cheapest conversion inside your targeting. People who already know you convert more easily, so broad delivery drifts toward them unless you stop it. Reported conversions look strong, but plenty of those orders would have arrived through email, search or a direct visit anyway, and the ad set gets paid for them.

Incrementality data points the same way. Haus, which runs incrementality tests for brands, analyzed 640 Meta experiments run on its platform since early 2024. Retargeting took about 15% of spend, yet new customers accounted for 70% of the measured impact in Advantage+ campaigns and 65% in manual ones. Measured, another incrementality vendor, reported that 64% of Meta-driven incremental conversions across more than 10,000 campaigns in 2025 came from new-to-brand customers.

Both are vendors, and neither defines new in quite the same way. The direction is still hard to argue with. Most of what Meta adds, it adds by finding people who hadn't bought yet.

The symptom in an account is a prospecting campaign whose platform ROAS sits well above the account average while first-time customers in Shopify barely move. It's one of the first things we check in an agency audit.

03 How to set up Customer Lifecycle Strategy

Build a clean existing-customer audience first, decide how far back existing goes, pick the mode, then check what Meta actually excluded. Run it against a broad ad set before you switch every campaign over. The setting can only exclude people it can match, so the customer data behind it decides how well it works.

1. Upload your full customer list with every identifier you hold: email, phone, name, city. Meta's documentation says extra identifiers raise the match rate. Add a website purchasers audience from your pixel and Conversions API events, and keep the list syncing automatically through your store or email platform. Meta says customer list changes usually take up to 24 hours to apply.
2. Decide what existing means for your brand. For a product people buy once, lifetime buyers is the safe default. For consumables, someone whose last order was 18 months ago has probably lapsed, and winning them back is close to acquisition. Meta hasn't published a lookback window for this setting, so the window is whatever you build into the audience.
3. Pick the mode. Exclude existing customers for most prospecting. Exclude engaged people as well only if another ad set retargets site visitors and video viewers, or you'll cut off the warmest non-buyers you have.
4. Read the exclusion list before launch. Since mid-October some advertisers report that Meta adds an automatic exclusion built from conversion data when you choose acquisition. Open the ad set and see what's there rather than assuming either way. Meta is switching settings on for advertisers a lot this autumn, product browsing included.
5. Test it before you roll it out. Duplicate a prospecting ad set, leave one copy on broad and set the other to acquisition, with the same budget and creative, for two to four weeks. Meta's A/B test tool splits the audience cleanly if you'd rather not run them side by side.

04 Your ROAS will drop. Judge it on new-customer CAC.

Excluding past buyers takes the cheapest conversions out of the ad set, so reported ROAS usually falls. That's expected, and it isn't the verdict. Judge the change on new-customer CAC: ad spend divided by first-time customers, counted in your store rather than in Ads Manager.

Here's an example with a $120 average order. The numbers are illustrative, so plug in your own test results:

Reported ROAS fell 24%. The acquisition ad set bought 25 more new customers, and each one cost 14% less. The broad version had claimed 100 returning buyers, and most of them keep ordering anyway, from your emails or by typing in your URL. That's why most of the dashboard's drop isn't a real loss. Confirm it with blended numbers: total new customers and MER for the test window, against the weeks before.

Count first-time customers in your store, not in Meta. Meta can only recognize past buyers it can match to an account, so a customer it can't match looks new to it. Your store's first-time versus returning split, or a match of order emails against your customer list, gives the honest count. For proof beyond a before-and-after comparison, an incrementality test with a holdout settles it.

05 When Customer Lifecycle Strategy won't help

Skip it, or keep it narrow, when your campaigns sit in a restricted category, when your customer base is too small to change delivery, when repeat orders are the business, or when an ad set already lacks conversions to learn from.

• Special ad categories. Restricted campaigns can't use the acquisition modes at all.
• A small customer list. If 1,500 people have ever bought from you, excluding them barely changes who sees the ads. The problem is more likely the offer or the funnel, and our seven growth leaks are a better place to start.
• Subscription and consumable brands. Repeat orders carry the margin. Exclude customers from acquisition ad sets, but give retention its own budget instead of letting it ride on prospecting.
• Low purchase volume. Narrower delivery means fewer conversions to learn from. If an ad set already struggles to exit learning, fix volume before you narrow it.

06 Where to start

Pull the last 90 days of orders your prospecting campaigns claimed and match them against your customer list, or use your store's first-time versus returning split. If a large share were returning customers, test the acquisition setting now.

Returning buyers flood in on their own over Black Friday, and paying prospecting rates for them that week is the most expensive version of this leak. The calculator shows what a new customer can cost at your margin, and if you'd like a second pair of eyes on the split, send us the numbers. How many of the customers your prospecting campaign acquired last month had already bought from you?

Sources: Meta for Business, Advertising Week New York 2026 announcement (October 6, 2026); Meta Marketing API reference for the ad set goal field and the Custom Audiences guide; Haus, The Meta Report: lessons from 640 Haus incrementality experiments (July 28, 2025), based on tests run by its customers; Measured, incrementality analysis of Meta platform performance (2026), based on its clients' 2025 campaigns. The example table uses illustrative numbers, not benchmarks.