What is LTV (lifetime value) in app and ad marketing?

Short answer. LTV, lifetime value, is the average revenue a user brings in over the whole time they use your app or buy from your brand. It sets the ceiling on what you can pay to acquire a user: if LTV is below your CPI or cost per purchase, growth loses money.

Definition

AppsFlyer defines LTV as "an estimate of the average revenue a customer will generate over the time that they use a given product or service." It is the metric that tells marketers "how much money they can afford to spend on user acquisition while remaining profitable."

AppsFlyer separates LTV from two near neighbours. CLV usually refers to one customer, while "LTV looks at the average value across your entire customer base." ARPU (average revenue per user) covers a fixed period rather than the whole relationship.

Formula

AppsFlyer gives two working formulas:

LTV = average purchase size x number of purchases x retention period

LTV = total revenue generated since install date (or during a defined period) / total number of users who installed on that date (or during that period)

The second, cohort version is the one app teams use day to day, read at fixed points such as D7, D30 and D90.

Worked example (illustration)

These are made-up round numbers. 10,000 users install a game in March. By day 30 they have generated $15,000 in purchases and ad revenue. D30 LTV = $15,000 / 10,000 = $1.50. If the March cohort's CPI was $2.00, it has not paid back by day 30 and needs later revenue to break even. If D90 LTV reaches $2.40, the cohort pays back somewhere between day 30 and day 90.

What a good or bad reading tells you

LTV only means something next to acquisition cost. AppsFlyer frames it as the LTV:CAC ratio ("LTV / CAC") and says that "if your result is below one, you have a problem." A rising LTV for the same CPI means better users or a better product. A falling LTV at stable CPI often means a channel or creative is bringing in users who do not stay.

We give no benchmark LTV. It depends entirely on your monetisation model, price and retention.

How it is used in app, game and brand advertising

Games use predicted LTV to plan spend before cohorts mature; AppsFlyer notes "LTV predictions can help you make forward-looking decisions around ad spend." Subscription apps typically watch trial-to-paid and renewal. Consumer brands read LTV on repeat purchase and use it to decide how much first-order loss is acceptable. Creative affects LTV too: an ad that promises something the product does not deliver buys installs that churn.

Common mistakes

1. Using one blended LTV for all channels and creatives. Users from different sources behave differently.
2. Treating a forecast as a measurement. Predicted LTV is a model; check it against matured cohorts.
3. Comparing LTV to CPI without a time window, so payback is never actually tested.
4. Ignoring that attribution limits (see MMP) can misassign which source a high-value user came from.

Flixr's view

Our view: LTV is where a creative's honesty shows up. A hook is a promise, and ads that over-promise lift installs and sink lifetime value, which is the argument of a hook is a promise. We would rather judge creative on downstream value in the client's own numbers, as in measure creative on purchases, than on the click.