Repeat Purchase Rate and LTV: How Much You Can Really Pay Per Customer

Why first-order ROAS understates what you can afford

Most Shopify ad accounts are optimized to a first-purchase ROAS or CPA target, which only measures profitability on the very first transaction. If a meaningful share of customers buy again, the true value of an acquisition is first-order profit plus the expected profit from future repeat orders — and ignoring that second part means turning away profitable customers because their first order alone didn't clear your CPA bar. The fix isn't abandoning first-order discipline, it's calculating a repeat-adjusted acceptable CPA ceiling so you know how much room you actually have.

The starting inputs you need, all of which should be available from Shopify order and customer data: average order value (AOV), gross margin percentage, repeat purchase rate within a defined window (for example, the percentage of first-time customers who place a second order within 12 months), and average number of total orders per customer over a defined lifetime window. These numbers should come from your own Shopify analytics or order exports, not industry averages, since repeat rate varies enormously by category — consumables and skincare often see high repeat rates while one-time durable goods purchases see very low ones.

Worked example: building a simple LTV

Assume AOV of $50, gross margin of 55% (so $27.50 gross profit per order before ad spend), a first-to-second-order repeat rate of 30% within 12 months, and that repeat customers who do order again average 2.2 total orders in that 12-month window. Expected orders per acquired customer = 1 (the first order) + 0.30 x 1.2 additional orders (1.2 being the average additional orders among repeaters, i.e., 2.2 total minus the first) = 1 + 0.36 = 1.36 orders per customer over 12 months. Expected 12-month gross profit per acquired customer = 1.36 x $27.50 = $37.40.

Turning LTV into an acceptable CPA

If the store's policy is to spend up to 50% of 12-month expected gross profit on acquisition — a common conservative target that leaves margin for fixed costs, tools, and cash flow buffer — the acceptable CPA ceiling becomes 0.50 x $37.40 = $18.70. Compare that to a first-order-only view: 50% of the single-order gross profit of $27.50 is $13.75. The repeat-adjusted ceiling of $18.70 is about 36% higher than the first-order-only ceiling, meaning a campaign converting at $16 CPA — which would look marginal or even unprofitable under a first-order-only lens — is actually comfortably profitable once repeat purchases are counted.

The risk of over-crediting future repeat orders

The repeat-adjusted ceiling only works if the repeat rate assumption is real and measured, not aspirational. A common mistake is using a repeat rate from a small, recent, unrepresentative cohort — for example, measuring repeat rate only from customers acquired in the last 60 days, before most of them have even had time to place a second order. Repeat rate should be measured on cohorts old enough that the full measurement window (12 months in the example) has actually elapsed, otherwise the rate is systematically understated and any CPA ceiling based on a properly mature cohort would look more generous than an immature one.

Segmenting by acquisition channel and product

Repeat rate often differs by which product someone first bought and by which channel acquired them — a customer who found you through a low-intent broad Meta ad may repeat at a lower rate than one who came through a high-intent Google Shopping search for your exact product. Where the data volume supports it, calculating separate repeat-adjusted CPA ceilings per channel, or at minimum per major product category, avoids either overpaying for channels that only produce one-time buyers or underpaying for channels that quietly produce your best long-term customers.

Worked example: two channels, two ceilings

Assume Google Shopping customers show a 40% repeat rate and 1.35 average additional orders among repeaters, while a broad Meta prospecting channel shows an 18% repeat rate and 1.1 average additional orders among repeaters, with the same $50 AOV and 55% margin for both. Google Shopping: expected orders = 1 + 0.40 x 1.1 = 1.44, gross profit = 1.44 x $27.50 = $39.60, 50% ceiling = $19.80. Meta prospecting: expected orders = 1 + 0.18 x 1.1 = 1.198, gross profit = 1.198 x $27.50 = $32.95, 50% ceiling = $16.47. The Google Shopping channel can sustainably support a CPA roughly 20% higher than the Meta prospecting channel purely because of repeat behavior, even before considering any first-order ROAS difference between the two.

How this interacts with cash flow, not just profit

A repeat-adjusted CPA ceiling describes what's profitable over a 12-month horizon, but ad spend is due immediately while repeat-order profit arrives later, sometimes much later. A store with tight cash flow may need to bid closer to the first-order-only ceiling even if the repeat-adjusted math justifies more, simply because it can't front the cash gap between paying for the acquisition today and collecting the repeat-order profit months from now. Fast-growing stores in particular should treat the repeat-adjusted ceiling as a ceiling on strategic willingness to pay, not a target CPA to run every campaign at.

Shortening the measurement window for faster feedback

A full 12-month repeat-rate measurement is the most complete but also the slowest to update — if repeat behavior shifts (a new subscription option, a product line change), a 12-month window means you won't see the shift reflected in your CPA ceiling for a year. Many Shopify merchants also track a shorter secondary window, such as 90-day repeat rate, as an earlier leading indicator; it will always be lower than the 12-month figure since it's a subset of the same behavior, but a sudden drop in the 90-day figure quarter over quarter is a useful early warning before the full 12-month number would show it.

Repeat rate as a lever, not just a measurement

Because the CPA ceiling is directly proportional to expected repeat orders, improving repeat rate has the same effect on how much you can spend on acquisition as improving first-order margin does. A post-purchase email flow, a loyalty or subscription option, or simply better product-fit targeting (acquiring customers more likely to repeat in the first place) can raise the repeat rate input and mechanically raise the acceptable CPA ceiling — meaning retention investment and acquisition budget aren't separate line items, they're connected through this calculation.

Worked example: a 5-point repeat rate improvement

Returning to the base example — AOV $50, margin 55%, 1.2 average additional orders among repeaters — raising repeat rate from 30% to 35% changes expected orders per customer from 1.36 to 1 + 0.35 x 1.2 = 1.42, and 12-month gross profit per customer from $37.40 to $39.05, a $1.65 increase. At a 50% acquisition spend policy, that's an extra $0.825 of acceptable CPA per customer — modest per customer, but across 1,000 acquired customers a year that's $825 of additional sustainable acquisition budget generated purely by a retention improvement, with zero change to ad spend efficiency itself.

Building the calculation into a recurring routine

Because repeat rate and margin can drift over time — margin from cost changes, repeat rate from product mix or customer service quality — the LTV-to-CPA-ceiling calculation shouldn't be a one-time exercise. Recalculating quarterly, using a properly matured cohort each time, keeps the acceptable CPA ceiling aligned with current reality rather than an assumption made a year earlier that may no longer hold, especially for stores whose product mix or margin structure changes meaningfully across quarters.

Putting the ceiling to work in campaign structure

In practice, a repeat-adjusted CPA ceiling is most useful as a bidding-strategy input rather than a literal target CPA to type into every campaign — setting Target CPA at the exact ceiling leaves no margin for the natural variance in daily performance. A more workable approach is running Smart Bidding toward a CPA meaningfully below the ceiling (for example, 70-80% of it) as the primary operating target, using the full ceiling as the maximum you'd tolerate during a deliberate scaling push or seasonal acquisition sprint where you're intentionally trading some near-term margin for growth.