Campaign budget sheets beside a calculator

Retention vs Acquisition Campaigns: How to Split Meta Ad Spend

October 29, 2026

Retention campaigns try to get existing customers to buy again. Acquisition campaigns try to win first-time buyers. On Meta, they need different audiences, different creative and different yardsticks, and mixing them is the most common way I see DTC brands fool themselves about performance. The right split is not a universal percentage. It depends on your margins, your repeat rate and how much new-customer growth your current ad spend is actually buying.

I’m Stefano Mazzei, and I run paid acquisition with my team at Cosma for e-commerce and DTC brands, mostly in the U.S. and Canada. This guide covers what I look at when a founder asks whether to push harder on new customers or squeeze more from the ones they already have.

Stefano Mazzei, Performance Marketing Expert

stefanomazzei.co

Make Your Meta Ads Spend Work Harder

Performance marketer Stefano Mazzei shares practical frameworks for balancing acquisition and retention spend, based on work with e-commerce brands.

Explore Meta Ads insights

TL;DR

Acquisition builds the customer base; retention decides how much each customer is worth. You need both, and the economic question is which dollar returns more right now.
On Meta, standard conversion optimization drifts toward people likely to buy anyway, so acquisition campaigns need clean customer exclusions and honest new-customer reporting.
Retention ROAS is usually overstated. Use holdout or lift tests to find the true incremental revenue before you shift budget.
Prioritize acquisition when launching or when your repeat base is thin; lean on retention when acquisition costs climb and your customer base is stable.
Judge the whole system on LTV:CAC and blended CAC, not on platform ROAS alone.

What each campaign type is for
Why the cost argument is not enough
How they differ inside Meta Ads
Measure value, not just ROAS
When to lean on which
A framework for splitting budget as you scale
Perspective: what we see in accounts
Get your acquisition and retention split audited
FAQ
Sources

What each campaign type is for

Customer acquisition is the work of attracting new customers to make a purchase, while retention is a company’s ability to keep generating sales from people who already bought. Retention efforts aim to raise customer lifetime value (LTV) and cut churn; acquisition expands the base.

For an e-commerce brand on Meta, that translates into two very different jobs:

	Acquisition campaigns	Retention campaigns
Audience	Cold prospects, broad or interest/lookalike seeds	Past purchasers, email/SMS subscribers, high-intent site visitors
Core message	Why this brand, why now, proof	New product, replenishment, bundle, loyalty
Creative angle	Problem and category education, social proof, hooks	Familiarity, use cases, what’s new, reasons to reorder
Key metric	New-customer CAC, first-order MER, payback period	Incremental repeat revenue, repeat rate, LTV lift
Main risk	Paying for people who would have bought anyway	Over-crediting ads for purchases that were coming regardless
Typical scale limit	Creative fatigue and rising CPMs	Size of your past-buyer list

Retention is also broader than ads. It is a system: onboarding, post-purchase experience, email and SMS, and the product itself. Loyalty points are only one piece. Meta retention campaigns are one lever inside that system, usually the one that fills gaps your owned channels leave.

Why the cost argument is not enough

Every top-ranking article on this topic repeats that keeping a customer is cheaper than winning a new one, and the oft-quoted multiples vary from five to twenty-five. I treat those numbers as directional at best. They come from different industries and eras, and a lot of them describe subscription software, not a DTC brand selling a $48 product.

What the better evidence says is more useful. In a study of online retail, the average online apparel shopper was not profitable until their fourth purchase, and online grocers needed to keep a customer for 18 months to break even on acquisition. That research dates from 1999, so don’t copy the figures into your model, but the logic holds: if you lose money on the first order, retention is not optional, it is how the acquisition spend gets repaid.

A later industry report on shopper behavior found that in the U.S., 40% of revenue came from returning purchasers who were only 8% of visitors, and that marketers had to acquire five shoppers to match the revenue of one repeat purchaser. The data covers 2011 to 2012, so it is a reminder of the shape of the curve, not a benchmark for your store.

The real takeaway: a small group of repeat buyers can carry a large share of revenue, and your acquisition budget is only as efficient as the repeat behavior behind it. If you want a clear primer on why both matter together, this overview of how to increase customer retention walks through the retention side that acquisition spend depends on.

How they differ inside Meta Ads

Optimization drifts toward existing customers

Meta optimizes for conversions. It does not inherently care whether the buyer is new. One analysis of new-customer campaigns notes that standard conversion optimization can favor returning customers, which is why exclusions and audience segmentation matter when the real objective is first-time buyers.

I’ve seen this inflate dashboards many times. An account shows a 3.5 ROAS on a “prospecting” campaign, but when we split by new versus returning customers, a big slice of that revenue came from people already on the email list. The campaign looked great and added little growth.

What we do at the start of any audit:

Upload a fresh customer list (purchasers, ideally the last 180 days or longer) and exclude it from acquisition ad sets or campaign-level audience controls.
Keep a separate retention campaign for past buyers so their spend and results never blend into prospecting.
Report new-customer revenue from your store or analytics (Shopify, your data warehouse), not from Meta’s number alone.
Check the reconciliation monthly. If you want to see how I put the sources together, see how to make blended CAC board ready.

Acquisition needs creative that does the explaining

Cold audiences know nothing. Acquisition creative has to earn attention in the first seconds, name the problem, and show proof: customer language, demonstration, comparison. This is where customer research pays off, because the hooks come from how buyers describe the problem, not from how the brand describes the product.

If an account is stuck before creative even gets a fair test, the issue may be the learning phase. I wrote a DTC audit playbook for exiting Facebook ads learning around the 50-conversion threshold, and a guide on warming up new ad accounts with a conversion-first approach. Lookalike audiences still have a role as seeds for broad delivery; I cover that in Meta lookalike audiences as precision seeds for Advantage+.

Retention creative assumes familiarity

Past buyers don’t need the category explained. They need a reason to come back: a new flavor, a refill reminder, a bundle that fits what they own, a how-to for getting more from the product. Short, simple creative often beats polished brand film here because the audience already trusts you.

Two tactics worth testing:

Remarketing with a tight window. Visitors who showed intent but didn’t buy are not retention, but they sit between the two. My remarketing on Meta playbook covers how to structure it under current privacy constraints.
Post-purchase ads. The moment right after checkout is a high-trust window to offer a second item. I documented results from practitioner tests of post-purchase ads, where the lift per order is modest but arrives with almost no extra acquisition cost.

Measure value, not just ROAS

Retention ROAS is easy to inflate

People who already bought from you have a high baseline chance of buying again. If you show them ads and they purchase, the platform takes credit, but many would have purchased anyway. Some operators in the space, including the CEO of a well-known DTC agency, have made this point and argue for incrementality testing to find true lift.

The practical fix is a holdout: suppress ads for a slice of a segment, such as 10 to 15%, and compare repurchase rates with the group that saw the ads. The gap is your real return.

Holdout and ad-exposed groups compared by repurchase rate

I’ve written the mechanics out in two places: how to run a geo holdout test on Meta when you can’t cleanly split audiences, and how to run Facebook lift tests to prove incrementality faster. For brands on Shopify Plus, there’s also a 60-day holdout test using Shopify Audiences on Meta.

Use LTV:CAC, but check what’s inside it

Researchers at the Darden School of Business describe customer lifetime value as one of the more important metrics for informing prospecting decisions. I agree, with a warning: a 3:1 LTV:CAC ratio can look healthy while hiding slow payback or LTV assumptions that rest on a few outlier customers. I break down the creative and funnel fixes in when 3:1 LTV:CAC lies.

The measures I want on one page:

New-customer CAC (not blended with returning buyers)
First-order contribution margin after COGS and shipping
30, 60 and 90-day repeat rate by acquisition cohort
Payback period in days
Incremental retention revenue from the latest holdout

When to lean on which

HubSpot’s guidance is a sound starting framework: prioritize acquisition when starting out, launching a new product, entering an untapped market or chasing rapid growth, and emphasize retention when acquisition costs are high, the customer base is stable or the economy is in a downturn. A PYMNTS report makes the same point from the small-business side: newer businesses lean on acquisition because they have no repeat base yet.

Here is how I translate that for a Meta-driven DTC brand:

Situation	Lean toward	Why
Under ~1,000 customers, new product-market fit signals	Acquisition	There is almost no past-buyer audience to work with
CPMs and CAC rising, repeat rate healthy	Retention and offer testing	Past buyers convert at lower cost; protect margin
Strong repeat rate, thin top of funnel	Acquisition	You are leaving cheap growth on the table
Single-purchase product (mattress, appliance)	Acquisition plus referral and cross-sell	Little natural repurchase
Consumable, replenishment product	Both, with retention timed to the usage cycle	Reorder windows are predictable

Note that retention seldom “wins” at the scale level. It tops out at the size of your customer list. Once you have mined that list, growth needs new customers again.

A framework for splitting budget as you scale

Harvard Business School research frames the question well: acquisition supports market entry, network effects and scale, while retention supports lifetime value, long-term profitability and word of mouth. The framework recommends comparing returns on both rather than using a universal split, using Return on Customer Investment to allocate resources so the two work in concert.

In practice, my team at Cosma starts with a simple allocation process:

Set a floor for acquisition. Whatever your retention numbers say, protect enough prospecting budget to keep new customers flowing, or the repeat base shrinks later.
Cap retention at the size of the opportunity. Estimate how many past buyers are due to reorder in the next 30 days. Spending beyond what that pool can absorb pushes frequency up and returns down.
Test incrementality quarterly. Run a holdout on retention and a lift or geo test on acquisition. Move budget toward whichever has the higher incremental return per dollar.
Feed what you learn back. Klaviyo’s guidance notes that good acquisition identifies prospects likely to need the product and buy again, then uses what it learns to improve retention. We build creative angles from the reasons repeat customers give, then push those angles to cold audiences.
Re-run the math after every big change in pricing, shipping thresholds, AOV or channel mix.

There is a creative angle too. Acquisition and retention aren’t competing departments; one builds the foundation and the other determines the value created after the sale. Klaviyo’s Amanda Kwasniewicz argues that brands relying on acquisition alone keep replacing customers instead of building relationships. I agree, and I would add that it also keeps your CAC permanently exposed to auction prices.

Perspective: what we see in accounts

We rarely see a brand fail because it chose retention over acquisition or the other way around. We see brands fail because they cannot tell the two apart in their reporting. A retention campaign that looks like a 6x ROAS and a prospecting campaign quietly serving existing customers both produce the same illusion: Meta is working, but growth is flat.

Our approach at Cosma ties media buying to creative testing and customer research. We want to know which hooks make a stranger buy the first time, which messages bring a customer back, and what the incremental data says about each. Then we put the budget where the proven return is, and we re-test it because it moves. Practitioner-led tactics, grounded in account results, beat theoretical splits like “80/20” every time.

— Stefano Mazzei

Get your acquisition and retention split audited

If you’re not sure whether your Meta account is buying new customers or relabeling existing ones, I can help. My team at Cosma looks at your new-customer CAC, your holdout results and your creative pipeline, then builds a plan that scales acquisition without starving the repeat revenue behind it. You can see how we’ve done this on our case studies page or book a call with Cosma.

You can also look at examples of the ad creative we produce in the Cosma portfolio, or visit my site for more practitioner notes.

FAQ

Is retention really cheaper than acquisition?

Usually, yes per order, because past buyers already know you. But the famous multiples (5x, 7x, 25x) come from varying contexts and shouldn’t be used as your own numbers. Calculate your own cost to win a repeat order against your new-customer CAC, and validate it with a holdout.

Should I run retention and acquisition in the same Meta campaign?

I’d keep them separate. Mixing them lets Meta favor the cheaper conversions from existing customers, which makes prospecting look better than it is. Separate campaigns with customer-list exclusions give you clean reporting and cleaner decisions.

What is a good split between acquisition and retention budgets?

There isn’t a universal ratio. Academic work from Harvard Business School points to comparing returns on both activities instead of using a fixed split. Start with a prospecting floor, cap retention to what your past-buyer pool can absorb, and shift budget based on incrementality results.

When should a new DTC brand focus on acquisition?

At launch, with a new product, or when entering a new market, because there is no repeat base yet. As your customer list grows and repeat behavior shows up, add retention campaigns and measure whether they produce incremental revenue.

How do I know if my retention ads work?

Run a holdout. Suppress ads for a small share of your past-buyer segment and compare repurchase rates. If the groups look alike, the ads are collecting credit, not creating sales.