Lifetime Value (LTV)

LTV is the central economic measure of a mobile-app business. It answers: "what is the total revenue a single user generates over their relationship with this app?" Every UAGlossaryPaid UA (Paid User Acquisition)Mobile app user acquisition via paid advertising channels — Meta, TikTok, Google, AppLovin, ironSource, and other ad networks. decision sits inside the LTV / CPIGlossaryCost Per Install (CPI)The average amount an advertiser pays for a single install attributed to a campaign — total spend divided by attributed installs. ratio. If your LTV is $30 and your blended CPI is $8, you can afford to buy almost any user on any channel — you'll recoup 3-4× your spend. If your LTV is $4 and your CPI is $8, you're burning cash on every install. All paid acquisition strategy fits inside that single ratio.

That $0.02 median D30 LTV is a derived IAPGlossaryIn-App Purchase (IAP)Any purchase made inside an app after installation — consumables, non-consumables, or subscriptions. Processed by Apple or Google, who take a 15-30% commission.-only proxy — computed by multiplying each app's measured ARPDAUGlossaryARPDAU (Average Revenue Per Daily Active User)Total revenue in a period divided by the average daily active users in that period — the per-day-per-active-user revenue rate. by the trapezoidal integration of its D1/D7/D30 retentionGlossaryRetention (D1 / D7 / D30)The percentage of users from a cohort who return to the app N days after install. D1 = day-1 retention, D7 = 7-day, D30 = 30-day. The single most important non-reven... curve. Real lifetime LTV is several times higher because (a) revenue continues past D30 on the asymptotic retention tail and (b) ad revenueGlossaryAd RevenueThe revenue an app earns from showing ads — fundamentally impressions × eCPM, after fill rate and the mediation layer decide which requests get filled and at what pr... is not in this dataset. Treat it as the shape of the catalog — a steep right-skew where most apps generate pennies of D30 IAP value per install, and only a thin tail clears a dollar.

LTV is almost always modelled, not observed. True lifetime revenue takes years to materialise; UA decisions need to happen this week. The three dominant modelling approaches:

1. 01Retention-curve extrapolation: fit a curve (typically a power-law decay) to your first 30-60 days of cohort revenue and project it out 12-24 months.
2. 02ARPUGlossaryARPU (Average Revenue Per User)Total revenue in a period divided by the user base for that period — the blended revenue-per-user metric, including non-paying users. × tenure: estimate average paying-user ARPU and multiply by an expected tenure (months of active use). Simple, useful for early-stage modelling.
3. 03Parametric churnGlossaryChurnThe rate at which users stop using an app (general churn) or cancel a subscription (subscription churn) over a time window — retention's mathematical complement. models (Pareto-NBD, BG-NBD): probabilistic models that estimate the probability a user is still active given their purchase frequency. Standard in subscriptionGlossarySubscriptionA recurring in-app purchase that charges the user on a fixed cadence (weekly, monthly, annual) until they cancel. businesses.

Subscription LTV has a closed-form approximation: LTV ≈ ARPPU_monthly ÷ monthly_churn_rate. A $10 / month subscription at 10% monthly churn implies an LTV of roughly $100 per paying user; at 5% churn, $200; at 20% churn, $50. This is why "reduce churn by 1 percentage point" is one of the highest-ROIGlossaryROI (Return on Investment)Net profit divided by total investment — the bottom-line profitability measure, distinct from ROAS because ROI nets out platform commission, cost of goods, and overh... moves a subscription app can make — it compounds across every future cohort.

Blended LTV hides everything that matters. A single LTV number averaged across the whole user base typically obscures a 4-10× spread between cohort types. Users acquired from a Tier-1 country on rewarded videoGlossaryRewarded VideoA video ad format that users opt-into voluntarily in exchange for an in-app reward (currency, lives, extra content). The highest-eCPM mobile ad format because the us... may have 4× the LTV of users acquired from an emerging market on cheap bannerGlossaryBanner AdA small persistent ad shown at the top or bottom of the app screen — the lowest-attention and lowest-eCPM mobile ad format, but high in impression volume. inventory — and yet a blended bidding strategy charges them all to the same UA budget. Mature UA programmes segment LTV by acquisition channel × country × creativeGlossaryAd CreativeThe visual and audio content of an ad — video, image, copy, sound. The single most important variable in modern mobile UA, lifting IPM and lowering effective CPI. cluster, and bid into each segment separately.

Common pitfall: confusing gross revenue LTV with net contribution LTV. Gross LTV is the revenue you collect from a user. Net contribution LTV is what's left after Apple/Google's 30% commission, hosting/serving costs, content licensing, and customer-service overhead. For UA decisions, what matters is whether net contribution exceeds CPI — not gross revenue. A $10 LTV against an $8 CPI looks fine on gross but is unprofitable on net.

LTV maturity ladder for an app: at launch, you have no observed LTV and must rely on industry benchmarks. By month 3, you can fit a basic retention curve and project. By month 12, you have your first observed full-year cohorts to anchor the model. By year 2-3, your LTV model can be highly accurate — but at that point your bigger problem is usually keeping the model fresh against changing acquisition channels and product changes, not initial modelling.

The histogram exposes the LTV power law in action. Roughly 80% of measurable apps generate under $0.10 of D30 IAP LTV per install; perhaps 5% clear $0.25; a handful pass $1. That's why blended LTV reporting is so misleading — the top 5% drives most of the dollar volume while the long tail anchors any unweighted median you compute.

Game apps lead category medians (no surprise — IAP is the game-native model), but the top-decile gap is what's interesting. Games' top-10% D30 LTV is roughly 7× their median, while Productivity's is only ~5×. Inside Games, the high-end tail is where whaleGlossaryWhale (High-Spending Player)A player who spends substantially more than the average paying user — typically the top 2-5% of paying players by lifetime spend, who generate ~80% of F2P revenue.-economics show up (gambling-mechanic games, mid-core RPGs, dating-as-game). For non-game apps, the top-decile is mostly subscription-driven — predictable revenue per active user but a smaller multiplier.