Customer acquisition is winning someone's first order. Customer retention is getting the next orders from people who already bought. On paid traffic the two share one budget: what your buyers are worth after 90 days sets the CPA you can pay, and the ads you buy decide which buyers you'll have to keep.
So skip the "which matters more" debate and find the side that is losing money right now. Compare a customer's 90-day value with what you paid for them, split by ad source.
If both teams could see only one number, I'd make it that 90-day value per customer.
Most articles on this topic land on the same advice: retention is cheaper, so move budget to it. The research behind it comes from businesses that don't buy each new customer in an ad auction. The best-known figure was measured in financial services. If you sell a supplement, a gadget or a course through Meta, Google or native ads, you can't keep a buyer you never won, and buying customers who never come back gets more expensive every month you scale.
The two sides usually break at the same seam: the ad report knows the first order, the email tool knows the reorders, and nothing joins them. So the media buyer optimises for a cheap first sale while the retention team works with whoever shows up. When I wrote our about page I put it this way: "Every quarter I watched revenue leak between tools."
Customer acquisition vs retention: the difference
Both work on the same customer, at different moments and with different money.
The same comparison in more detail, for a funnel that buys its traffic:
| On a paid-traffic funnel | Customer acquisition | Customer retention |
|---|---|---|
| What it buys | The first order | The second order and the ones after it |
| Main cost | Ad spend, creative, affiliate commissions | Discounts, sending fees, agent time |
| Number to watch | CPA and what the first order leaves | Repeat purchase rate by cohort, churn for subscriptions |
| Channels | Meta, Google, TikTok, native, affiliates | Email, SMS, phone, subscriptions, a reorder link |
| When it pays | On the first order, or not at all | Over weeks or months |
On a funnel that sells on the first click, your customer acquisition cost is mostly the CPA, plus creative and any affiliate commission on top.
The columns overlap more than the table makes it look. A retargeting ad shown to past buyers is paid traffic doing retention work. An order bump on the first checkout is extra revenue from a buyer you already won, earned before they've even left the page.
When I look at a funnel I read the two columns together. A CPA tells me very little until I know what that buyer does next.
Is retention really five times cheaper than acquisition?
You've seen the line: winning a new customer costs five times more than keeping one. Before you move budget on it, look at where it comes from.
- The profit figure traces to Fred Reichheld at Bain & Company. His article "Prescription for cutting costs" says: "In financial services, for example, a 5% increase in customer retention produces more than a 25% increase in profit." The example was banking customers.
- The cost ratio is usually quoted from a 2014 Harvard Business Review article by Amy Gallo: a new customer is "anywhere from five to 25 times more expensive" than keeping one. The article itself says the number depends on which study you believe.
Neither was measured on a funnel that pays for every click. You don't need them, though, because your own ratio takes two numbers to work out. Same example product as in our customer retention strategies post: a $59 bottle bought at a $55 CPA, and a reorder email with a 10% discount (illustrative numbers):
| Cost of one order from | Cost | Where it comes from |
|---|---|---|
| A new buyer | $55 | The CPA |
| An existing buyer | $5.90 | 10% off a $59 order |
| Ratio | About 9 to 1 | $55 ÷ $5.90 |
So on this product a reorder looks nine times cheaper, before sending costs. The catch is that some of those buyers would have reordered at full price anyway. Their discount bought nothing, so the real cost of each extra reorder is higher than $5.90. Test it with a holdout: send the offer to most of a cohort, keep a slice back, and compare the two reorder rates.
Even when the ratio holds, it doesn't tell you where the next dollar goes. Retention has a ceiling, because it only works on the buyers acquisition already brought in.
How retention sets your acquisition budget
This is where the two meet in numbers. Each order of the example product leaves $40 after product, shipping and fees (its contribution margin). If nobody ever reorders, $40 is the most you can pay for a buyer and break even.
Add reorders and the ceiling moves:
| Buyers who reorder within 90 days | Contribution per customer | Break-even CPA | CPA for $10 profit per customer |
|---|---|---|---|
| 0% | $40 | $40 | $30 |
| 20% | $40 + 20% × $40 = $48 | $48 | $38 |
| 40% | $40 + 40% × $40 = $56 | $56 | $46 |
In an ad auction a higher ceiling buys reach. At $56 you can bid on audiences and placements that a $40 ceiling priced you out of, which is why retention is what lets acquisition scale. A competitor whose buyers come back can simply outbid you for the same people.
The cost is cash flow. Say you raise the CPA to $50 on the strength of that 40% reorder rate. The first order leaves you $10 short per customer, and you only get it back once a quarter of the cohort has reordered ($10 ÷ $40 = 25%). Until then the money sits in your ad account. That gap is your payback period, and it's worth knowing in days before you raise bids.
I'd raise a CPA target on reorders a cohort has already made. Forecast reorders have a way of arriving later and smaller than the spreadsheet said.
How acquisition decides who you have to keep
The ad and the page choose the customer. The customer then decides how hard retention will be.
- The promise. An ad that oversells brings refunds in the second week. No reorder email fixes a buyer who feels misled.
- The discount. Someone who came in for 50% off has learned your price is 50% off, and many will wait for the next one.
- The source. Buyers from different ad sources come back at different rates, and the first-order report can't show it. The retention strategies post works through two sources where the dearer one wins after 90 days.
- The offer structure. A subscription or a three-bottle pack on the first checkout books part of the "second order" up front.
That's why I'd rather pay a few dollars more for buyers who come back than chase the lowest CPA on the dashboard. The cheapest buyers are often the ones the discount found, and many of them wait for the next discount before they order again.
The first order is where both happen
An upsell or an order bump raises what the first order leaves, so both shorten the payback period without a second ad or a second email.
Say the example checkout adds a $19 bump that leaves $12 after its own cost, and 30% of buyers take it (illustrative). That's 30% × $12 = $3.60 more per buyer, and the first-order break-even CPA goes from $40 to $43.60 before a single reorder. It also tells you something about retention: a buyer who takes the bump has already said yes twice.
To act on any of this you need the revenue per ad source next to the spend. In ElasticFunnels the checkout carries order bumps, and one-click upsells run after it on Stripe, NMI and PayPal. The analytics dashboard's Top UTM sources card shows sessions and revenue per source, and an audience can combine the traffic source with lifetime value. Filter the dashboard to that audience and "revenue from Meta buyers only" takes one filter.
If you want to see which of your sources brings buyers worth their CPA, the analytics page shows what the dashboard and audiences cover. Every feature is on every plan.
Acquisition or retention: which to fix first
Start from the 90-day value per customer against the CPA, per source. The answer usually falls into one of three branches.
Two more patterns show up often enough to name:
| What your numbers show | Fix first | First move |
|---|---|---|
| Loses money even at 90 days, few reorders | Acquisition | The offer, the price, the page |
| First order breaks even, repeat rate flat | Retention | Reorder timing and failed payments |
| Profitable at 90 days, spend held back by the CPA target | Acquisition | Raise the target toward the 90-day value |
| CPA climbing as you scale | Both | Lift first-order value with a bump or upsell |
| Refunds bunched in the first two weeks | Acquisition | What the ad and page promise |
You'll see fixed budget splits recommended, 70/30 or 80/20. I wouldn't use one. A split that ignores your reorder rate is a guess, and your own cohorts will tell you more within a quarter.
Customer acquisition and retention in Q4
Black Friday pulls both levers at once. First orders get cheaper to win, because buyers are looking, and thinner, because the discount comes out of the margin.
- Set the Black Friday CPA ceiling from last year's Black Friday buyers. If their 90-day reorder rate was lower than an ordinary month's, the ceiling is lower too, however good the weekend's ROAS looks.
- Plan the January reorder before the weekend. A buyer who got 40% off in November needs a reason to pay more in January, and that message is easier to write before the rush than during it.
- Keep the Black Friday cohort separate in your reports, so its lower repeat rate doesn't drag down the number you use to set CPA targets for the rest of the year.
The Black Friday marketing strategy post covers the discount math and the ads themselves.
- Hand the media buyer a CPA target computed from the last full cohort's 90-day value, and recompute it every quarter. A target set once at launch is out of date by the second month.
- Count retargeting ads to past buyers as retention spend. Mixed into acquisition, they make your CPA look better than it is.
- A discount that wins the first order is a retention cost you pay later, in the price that buyer expects next time. Price it that way.
- When a source looks bad on first-order ROAS, I'd wait for its 90-day number before killing it, unless it's far off. A month of reorders can turn the ranking of two sources around.
Acquisition and retention get argued about as if a company had to pick a side. For a seller who buys traffic, they're the two halves of one sum: what a customer is worth, minus what you paid to get them. Work both halves out for last quarter's buyers, per source, and most of the argument settles itself. Then raise the bids where the sum says you can.
