Both get quoted in board decks. Only one should be steering how much you spend to acquire a customer, and it's usually not the one that gets the most airtime, which leads to acquisition budgets that are either too conservative or too aggressive relative to what the business can actually support.
What each number actually tells you
AOV (Average Order Value) is a single-transaction number. It tells you how much a customer spends per purchase, right now, this order. It's useful for pricing, bundling, and upsell decisions, and it's a number you know with certainty because it's already happened, no projection or assumption involved.
LTV (Lifetime Value) is a customer-level number projected or measured over time. It tells you what a customer is actually worth across their full relationship with the brand, accounting for repeat purchases, returns, and churn. Unlike AOV, LTV always involves some degree of projection or extrapolation, even when it's based on real historical cohort data, because you're making a claim about future behavior based on past patterns.
The confusion happens when brands use AOV as a proxy for value when setting acquisition budgets. A high AOV with a low repeat rate can be worth less over time than a moderate AOV with strong repeat behavior, and spending against AOV alone misses that entirely. Two brands with identical AOV can have wildly different acquisition economics if one has a customer base that reorders three times a year and the other has one that almost never comes back.
Why this matters for CAC decisions
If you're setting a target CAC based on a single order's contribution margin, you're implicitly assuming zero value from repeat purchases. For categories with genuine repeat behavior (which most fashion and accessories brands have, even if it's slower than consumables), that undervalues the customer and leaves growth on the table by being too conservative on acquisition spend. A brand that could profitably spend more to acquire a customer, because that customer is worth more over time than the first order suggests, ends up underinvesting in growth out of unnecessary caution.
The reverse mistake is more dangerous: using an optimistic LTV projection to justify an aggressive CAC before the repeat behavior is actually proven. This is the mechanism behind a lot of D2C brands that scaled fast, looked healthy on paper, and ran into cash flow trouble, spending against a value that hadn't materialized yet. The gap between projected LTV and realized LTV is where a lot of D2C cash flow problems actually originate, not in the acquisition spend itself but in the assumption baked into how that spend was justified.
A practical framework
Use AOV to set first-order contribution margin and a hard floor CAC. This is the number you know is real, today, no projection required. A CAC set at or below the first-order contribution margin means the business is never structurally dependent on repeat purchases to break even on acquisition, which is a genuinely safer starting position even if it caps growth speed somewhat.
Use a conservative, cohort-based LTV (60 to 90 day, not a lifetime projection) to set how much above that floor you're willing to spend, and only once you have actual repeat-purchase data from cohorts, not an assumption borrowed from an industry benchmark or a more mature competitor's public numbers. A 60 to 90 day window is long enough to capture a meaningful first repeat purchase for most D2C categories without requiring years of data history the brand may not have yet.
Never use a 12-month-plus LTV projection to justify current acquisition spend unless the brand has multiple years of stable repeat-purchase data. Most D2C brands don't, and the projection becomes a story that makes the current spend feel safer than it is, rather than a genuinely reliable input into the decision.
How this framework plays out at different stages
Early stage, pre-product-market-fit signal. AOV-based CAC ceiling is the only defensible approach here, since there's no reliable repeat-purchase data yet to build an LTV number from. This naturally constrains acquisition spend to what the business can support on first-order economics alone, which is the right constraint at this stage even though it feels limiting.
Growth stage, with 6 to 12 months of order history. A cohort-based 60 to 90 day LTV becomes usable, and it typically allows for a somewhat more aggressive CAC ceiling than the AOV-only floor, since there's now real data showing at least some repeat behavior. This is usually the stage where brands can meaningfully accelerate acquisition spend without taking on unreasonable risk.
Mature stage, with multiple years of stable cohort data. Longer LTV windows become defensible, and acquisition strategy can incorporate a genuinely longer-term view of customer value. This is also the stage where the earlier framework's caution about 12-month-plus projections relaxes somewhat, since the data supporting those projections is now actually there.
A common trap: treating blended LTV as if it applies to every acquisition channel
Even with solid cohort data, a single blended LTV number can mislead if applied uniformly across every acquisition channel. A customer acquired through a referral or organic search often has meaningfully different repeat behavior than one acquired through a cold paid social ad, since the acquisition context itself carries information about purchase intent and brand affinity. Where the data supports it, segmenting LTV by acquisition channel, not just applying one blended number everywhere, produces a more accurate picture of what each channel can actually support in CAC.
FAQ
How long of a cohort window is reliable for LTV? 60 to 90 days is a reasonable starting point for most D2C categories, long enough to capture a first repeat purchase without waiting so long the data becomes stale relative to current business conditions like pricing, competition, or product mix changes.
Should AOV or LTV determine pricing strategy? AOV is more directly actionable for pricing and bundling, since it reflects real, current transaction behavior. LTV is more relevant for deciding how aggressively to invest in retention versus acquisition, since it captures the value of the ongoing relationship rather than a single transaction.
What's a healthy relationship between CAC and LTV? A common starting benchmark is CAC recovered within the first order or two, with LTV providing the margin for aggressive growth beyond that floor, though exact ratios vary meaningfully by category, margin structure, and how capital-constrained the business is at a given stage.
How often should LTV assumptions be revisited? At least quarterly for a growing brand, since repeat-purchase patterns can shift with product mix changes, pricing changes, or shifts in the acquisition channel mix. An LTV assumption calculated a year ago on a different product lineup and customer mix may no longer reflect current reality.
Does a high return rate affect how LTV should be calculated? Yes, significantly, particularly in fashion where returns can be substantial. LTV calculations that don't net out returns, and RTO for COD-heavy markets, systematically overstate customer value, which compounds the risk of setting an acquisition budget too aggressively.
Want to know your actual acquisition budget ceiling based on real repeat-purchase data? Run the numbers with our calculators.