User Acquisition

pLTV (Predicted Lifetime Value)

Also known asPredicted LTVpLTV ModelingEarly LTV Prediction

A class of machine-learning models that predict a user's eventual lifetime value within hours or days of install, based on early-funnel signal — not observed multi-year revenue.

trending

Key takeaways

  1. 01pLTV models score individual users in their first 24-72 hours based on early signals: onboarding, session depth, first-purchase, event sequences.
  2. 02Major ad networks (Meta, TikTok) now consume pLTV optimization events directly — high-pLTV users get bid up automatically.
  3. 03Typical signal lift: pLTV-targeted UA delivers 20-40% higher net revenue per dollar versus install-volume optimization, in apps where it works.
  4. 04Build threshold: 3-6 months of cohort data minimum, a feature pipeline that scores in near-real-time, and a feedback loop that retrains as user behaviour drifts.

pLTV (predicted lifetime value) is a class of machine-learning models that estimate how much a user will eventually spend — not from observed multi-year revenue, but from early-window signal. The model scores a user within hours or days of install based on what they did in onboarding, how deep their first session went, whether they completed the tutorial, whether they made a first purchase, what event sequences they generated. The output is a forecasted LTV per user, used for bid optimization.

Why pLTV matters now: a decade ago, UA optimization was install-count or cost-per-install. The mature version was ROAS — target return on ad spend at D7 or D30. pLTV is the next evolution: instead of waiting for D30 observed revenue, you can bid into estimated D365 revenue starting from D1. Meta and TikTok's app campaigns now consume pLTV signals as their optimization event directly — when you send their network a "this user scored 85th-percentile pLTV" event, the network will bid up similar lookalike users.

What signals drive pLTV models in practice

Build threshold: pLTV is not a quick win. You need (a) at least 3-6 months of historical cohort data to fit a model with predictive power, (b) a feature pipeline that can score users in near-real-time (a few hours of latency is usually acceptable; days is not), (c) an event-export pipeline to fire pLTV scores back into Meta / TikTok / AppLovin as conversion events, (d) a feedback loop that retrains the model regularly as user behaviour drifts and the product changes.

Vendors abstract this for mid-market. Apps that don't want to build pLTV in-house can use vendors like AppsFlyer Predict, Adjust Predictive Analytics, Tenjin pLTV, Liftoff Vungle's audience tools. The trade-off is the vendor's model is less customized to your app's specific monetization shape — but they handle the data pipeline, model training, and network event-export plumbing for you.

When pLTV doesn't work: apps with monetization that ramps slowly (LTV materializes mostly in months 6-12), apps with very low daily activity (insufficient early signal to score on), and apps with monetization shaped by external events (seasonal F2P games where in-game events drive revenue). pLTV is most powerful when first-week behaviour strongly predicts year-1 revenue — common in subscription, less common in deep F2P meta loops.

pLTV vendors vs in-house build

ApproachSetup timeCustomizationBest for
AppsFlyer PredictDays to integrateLimited — vendor modelMid-market apps already on AppsFlyer
Adjust Predictive AnalyticsDays to integrateLimitedAdjust customers
Tenjin pLTVDaysLimitedIndie + smaller publishers
Liftoff audience toolsDaysLimitedApps spending on Liftoff DSP
In-house build3-6 monthsFull — tailored to your monetizationLarge publishers ($500K+/mo UA) with ML capacity

Vendor pLTV works for most apps that don't have unusual monetization shapes. In-house pLTV is justified when (a) you have 3-6 months of cohort data and ML engineering capacity, (b) your monetization shape is atypical, (c) UA spend is high enough that 5-10% improvement justifies the engineering investment.

Quick answers

What is pLTV (predicted LTV) and how is it different from LTV?

**LTV** measures actual lifetime revenue from a cohort — but takes months or years to materialize. **pLTV** predicts that lifetime revenue within hours or days of install, based on early-funnel signals: onboarding completion, session depth, first-purchase events, event sequences. UA bids need to happen this week, not 12 months from now, so pLTV is what UA managers actually optimize toward in 2026.

What signals do pLTV models use?

The most predictive signals are: (a) first-purchase events in the first session — IAP / subscription start is the single strongest pLTV indicator, (b) onboarding depth and completion, (c) session-1 engagement (time-in-app, screens, content interaction), (d) event sequences (did the user do X then Y), (e) macro-features baked alongside: device, country, acquisition channel, creative cluster.

How do Meta and TikTok use pLTV?

When your SDK or postback pipeline sends Meta / TikTok a pLTV score for a user (typically as a custom optimization event), the network treats that user as a high-quality conversion and bids up lookalike users in subsequent campaigns. The mechanism is the same as sending any value-based conversion event — Meta and TikTok's bidding algorithms optimize for users similar to the ones with high reported event values. The pLTV use case is just sending the predicted (rather than observed) LTV.

How much data do I need to build a pLTV model?

Practical floor: 3-6 months of historical cohort data with both early-funnel events and downstream revenue observations. Below that, you don't have enough cohort-by-cohort observation to fit a model with predictive power. Beyond data volume, you also need: a real-time feature pipeline that can score new users within hours of install, an event-export pipeline to fire pLTV scores into ad networks, and a regular retraining loop as user behaviour drifts.

Back to glossary