LTV vs ROAS
Lifetime value looks forward at what a user is worth; ROAS looks back at what a campaign has earned so far. Read together, they tell you when spend pays back.
LTV is the total revenue (or margin) a user is expected to bring over their life in the app; ROAS is attributed revenue so far divided by ad spend. Cohort ROAS at day n is simply revenue per install by day n divided by CPI, so as a cohort ages its ROAS climbs towards LTV ÷ CPI.
What is the difference between LTV and ROAS?
LTV is a value per user over their whole lifetime; ROAS is a ratio of revenue to ad spend over a window that is usually much shorter. LTV tells you what you can afford to pay for a user, and ROAS tells you how much of that a campaign has earned back so far.
| LTV | ROAS | |
|---|---|---|
| Measures | Revenue or margin per user over a lifetime (or long horizon) | Attributed revenue ÷ Ad spend |
| Unit | Money per user | Percentage or multiple |
| Time | Forward-looking, long horizon | Backward-looking, D0 to D30 or a period |
| Includes cost? | No | Yes, divides by spend |
| Used for | Setting bid and CAC limits | Judging and pacing campaigns |
Cohort ROAS: revenue from one day's installs, by day n, divided by that day's spend. D7 ROAS compares the first week's revenue with what those installs cost.
How are LTV and ROAS connected?
For an install cohort, ROAS by day n = revenue per install by day n ÷ CPI. As the cohort ages, revenue per install approaches LTV, so ROAS approaches LTV ÷ CPI. Early ROAS is a down payment on the LTV that the campaign's users will eventually produce.
- Break-even: the day cohort ROAS, measured on gross margin, reaches 100% is the payback day.
- Maximum CPI: LTV ÷ target ROAS. Paying more than that cannot meet the target, however the early ROAS looks.
- Early read: the ratio between D7 and later ROAS in your past cohorts lets you predict where a new cohort is heading from its first week.
The calculator takes revenue per install at three ages and shows ROAS at each, plus how much of the expected LTV the campaign has already earned.
Example inputs: change them to your own numbers. The examples only make the calculator show a worked answer; they are not benchmarks or targets. Runs in your browser; nothing you type is sent anywhere.
- D7 ROAS
- 40%
- D30 ROAS
- 90%
- Lifetime ROAS (LTV ÷ CPI)
- 180%
- Share of LTV earned by day 30
- 50%
When to use which
- Before buying: LTV, ideally margin-based cohort LTV, to set the most you can pay per install or customer.
- While buying: D0 and D7 ROAS by campaign, compared with the D7 ROAS that past cohorts needed to reach break-even.
- After a quarter: realised D90 or D365 LTV by channel, to check whether early ROAS predicted the outcome.
Common mistakes
- ROAS on revenue, LTV on margin. Mixing bases makes a profitable campaign look unprofitable or the reverse. Use gross margin for both when judging profit.
- Comparing ROAS at different ages. A D30 ROAS is not comparable with a D7 ROAS. Compare cohorts at the same age.
- Platform ROAS vs MMP ROAS. Ad platforms and attribution tools credit revenue differently, so the same campaign shows different ROAS. Attribution Debugger explains why ROAS differs between tools even when install counts match.
- LTV without a horizon. "LTV of $40" means little until you say D90, D365 or predicted.
The glossary entries for return on ad spend and lifetime value give short definitions of both.
Measure these by channel
Questions
Can ROAS be above 100% while LTV is below CAC?
Yes, if ROAS is measured on revenue and CAC is fully loaded with non-media costs, or if margin is low. Compare on the same cost and revenue basis.
What is D7 ROAS?
Revenue in the first seven days from one day's installs, divided by the ad spend that bought them. It is an early read on whether a cohort will pay back.
How do you predict LTV from early ROAS?
Use past cohorts: if D7 revenue has typically been about a third of D90 revenue, a new cohort's D7 revenue times three gives a rough D90 estimate. Recheck the multiplier as the product changes.
Is LTV to CAC the same as ROAS?
Not quite. LTV to CAC compares lifetime value with full acquisition cost, while ROAS compares revenue so far with ad spend only.
What is the difference between a metric and a KPI?
A metric is any measured quantity; a KPI is a metric a team has chosen as a target for judging performance. Most metrics here become KPIs only for the teams whose work moves them.
Do I need an MMP to measure these metrics?
Not for all of them. Engagement and monetisation metrics come from product analytics and billing data, while metrics split by ad channel, such as CPI, ROAS or retention by source, need installs attributed to campaigns.
More comparisons
- CPI vs CAC vs CPAThree cost metrics that are often mixed up. Each prices a different step of the funnel, from the install to the paying customer.
- ARPU vs ARPPUTwo revenue-per-user metrics that differ only in who you divide by, and that can move in opposite directions.
- DAU/MAU vs retention rateTwo engagement measures that sound alike and answer different questions: how often active users come back, and how many new users stay at all.