{"id":274,"count":0,"description":"","link":"https:\/\/tenjin.com\/zh\/glossary\/predictive-ltv-pltv\/","name":"Predictive LTV (pLTV)","slug":"predictive-ltv-pltv","taxonomy":"glossaries","parent":0,"meta":{"status":["1"],"order":["0"],"glossary_term_description":["<div style=\"border: 1px solid #e5e5e5;padding: 16px;border-radius: 8px;background: #fafafa\">\r\n\r\n<b>Definition:<\/b>\r\n<span style=\"font-weight: 400\">\r\nPredictive LTV, or predicted lifetime value (pLTV) is an estimate of how much revenue a user or acquisition cohort is expected to generate in the future. In mobile marketing, predictive LTV uses early behavioral, revenue, and historical signals to help marketers evaluate campaign quality before the full user lifecycle has matured.\u2028<\/span>\r\n\r\n<\/div>\r\n&nbsp;\r\n<h2><span style=\"font-weight: 400\">What Is Predictive LTV?<\/span><\/h2>\r\n<a href=\"https:\/\/tenjin.com\/blog\/what-is-ltv-prediction\/\"><span style=\"font-weight: 400\">Predictive LTV<\/span><\/a><span style=\"font-weight: 400\"> is a form of <\/span><a href=\"https:\/\/tenjin.com\/glossary\/predictive-analysis\/\"><span style=\"font-weight: 400\">predictive analysis<\/span><\/a><span style=\"font-weight: 400\">. It is a metric used to estimate a user\u2019s value in the future, before the actual value has been generated.\u00a0<\/span><span style=\"font-weight: 400\">It is related to traditional lifetime value (LTV), which looks at past data and measures revenue that a user or cohort has already produced. Unlike traditional LTV, pLTV is a forecast. It uses the data available from early lifecycle signals to infer and estimate how much revenue users of a cohort are projected to generate at a later date.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">pLTV is extremely useful for user acquisition teams, who have to make decisions quickly. Marketers can\u2019t always wait weeks or months for cohorts to mature before deciding whether to add more budget and scale, maintain, or pause a campaign. <\/span><span style=\"font-weight: 400\">For example, a user acquired on Day 1 may eventually generate value through:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">In-app purchases<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Advertising revenue<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Subscription payments<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Renewals<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Other monetization events<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">A good pLTV metric is able to estimate and project value based on certain signals in the algorithm.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">Predictive LTV vs. Historical LTV<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Historical or <\/span><a href=\"https:\/\/tenjin.com\/glossary\/lifetime-value-ltv\/\"><span style=\"font-weight: 400\">traditional LTV<\/span><\/a><span style=\"font-weight: 400\"> measures realized value. Predictive LTV estimates future value. <\/span><span style=\"font-weight: 400\">Let\u2019s consider two user cohorts seven days after acquisition.<\/span>\r\n\r\n<b>Campaign A<\/b>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Day 7 realized LTV: $1.50<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Predicted future LTV: $4.20<\/span><\/li>\r\n<\/ul>\r\n<b>Campaign B<\/b>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Day 7 realized LTV: $1.70<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Predicted future LTV: $2.40<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">If we examine realized value, then Campaign B looks stronger than A. However, looking at the pLTV projections, Campaign A is the winner. The result is based off of the early signals from Campaign A. They could have had more sessions, better retention, higher engagement, stronger purchase intent, or a combination of all of these. <\/span><span style=\"font-weight: 400\">pLTV is a metric that helps identify early signal patterns and communicate their strengths before revenue is fully there.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">How Does Predictive LTV Work?<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Predictive LTV models typically combine early user behavior with historical patterns from mature cohorts. Some of the possible signals include:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Early revenue<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Sessions<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Retention<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Ad impressions<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Purchase activity<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Engagement depth<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Acquisition source<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Geography<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Platform<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">App behavior<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">The job of the predictive model is to learn relationships and correlations between early signals and later projections. The historical data for an app could show that users who completed 5 sessions, viewed 10 rewarded ads, and made a purchase are more likely to generate more LTV. So, when a new cohort of the same app starts to display similar behavior, the model can estimate a higher forecast of LTV.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">Every predictive analytics approach is a bit different, each MMP has their own algorithm. Every system uses their own models, signals, and forecast windows, so it's very important that marketers who use pLTV understand what the metric represents.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">What Is Tenjin's Predicted LTV?<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Tenjin includes Predicted Lifetime Value, or pLTV, as part of its revenue dashboard. The current <\/span><b>N-Day All pLTV (Ad Mediation + IAP)<\/b><span style=\"font-weight: 400\"> metric combines ad mediation revenue and in-app purchase signals into one machine-learning forecast. According to <\/span><a href=\"https:\/\/tenjin.com\/docs\/predicted-lifetime-value-pltv\/\"><span style=\"font-weight: 400\">documentation<\/span><\/a><span style=\"font-weight: 400\">, Tenjin\u2019s pLTV model:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Uses cohort sizes from Day 1 through Day 30<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Combines ad mediation and IAP signals<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Uses matured cohort and historical data<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Processes early behavioral and monetization signals<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Has approximately 90% average accuracy<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Can use up to 78 input signals<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">To add the metric to your revenue report, simply go to the add metrics section in the dashboard and search for pLTV. Because it is available directly from the dashboard, marketers are able to immediately evaluate their pLTV metric according to cohorts alongside their spend and LTV.\u00a0<\/span>\r\n\r\n<h2><span style=\"font-weight: 400\">Predictive LTV and Subscription Apps<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Predictive LTV is highly relevant when it comes to subscription apps because the subscription lifetime cycle takes longer to develop compared to many hypercasual and gaming apps. A newly acquired subscription cohort may only show some installs, trial starts, and some first payments. A month later, the same cohort could generate renewals, cancellations, reactivations, and reached a matured subscriber LTV.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">It is a similar pattern and timing issue that is observed with <\/span><a href=\"https:\/\/tenjin.com\/blog\/predicted-ltv-for-hybrid-monetization\/\"><span style=\"font-weight: 400\">hybrid monetization models<\/span><\/a><span style=\"font-weight: 400\">. And when under pressure, marketers may need to make a budget decision before the full subscriber lifecycle is visible.<\/span>\r\n\r\n<span style=\"font-weight: 400\">The early signals for subscription LTV take into account:\u00a0<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Trial start rate<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Trial-to-paid conversion<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Plan selection<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">First renewal behavior<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Early engagement<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Cancellation timing<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">Early signals differ for subscriber LTV by focusing on the value generated by paying subscribers. It could show that mature annual subscribers trend $80 on average. However, the predictive model may estimate a newly acquired cohort has the likelihood of generating $70.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">Similar to traditional or historical LTV, the subscriber LTV describes the value whereas the pLTV describes a projected subscriber value. Both metrics are useful for deciding on campaign economics.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">pLTV vs. pROAS<\/span><\/h2>\r\n<span style=\"font-weight: 400\">If pLTV is a projected estimate of user or cohort value, then pROAS would be a projected estimate of ROAS, or the revenue earned relative to ad spend. In order to calculate pROAS we use the simple formula:\u00a0<\/span>\r\n<pre>Predicted ROAS = Predicted Revenue or LTV \u00f7 Spend \u00d7 100\r\n\r\n\r\n<\/pre>\r\n<span style=\"font-weight: 400\">For example:<\/span>\r\n<ul>\r\n \t<li><span style=\"font-weight: 400\">Campaign spend = $10,000<\/span><\/li>\r\n \t<li><span style=\"font-weight: 400\">Revenue = $6,000<\/span><\/li>\r\n \t<li><span style=\"font-weight: 400\">pLTV = $14,000\u00a0\u00a0<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">In this example, the current realized ROAS is 60% and the pROAS is 140%. The pLTV projection suggests the campaign could earn beyond break-even as the cohort matures. This adds important context compared with early revenue alone. Rather than stopping the campaign based on its current ROAS, the team could continue running the campaign and monitor whether performance continues according to the projection.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">This shows that predictive LTV can be more useful than early revenue signals alone. Early signals are noisy: a small number of whales could make a new cohort seem stronger than it actually is. Likewise, a cohort with slow but steady purchases could look weak at the start, and end up with more long-term potential. The same principle applies to hybrid monetization and subscription apps: the first payment does usually reveal renewal quality.<\/span><span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\"> Good predictive models take into account a wider range of signals and add a trained metric to the decision context.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">What Makes a Good Predictive LTV Model?<\/span><\/h2>\r\n<span style=\"font-weight: 400\">A useful LTV prediction model should have:<\/span>\r\n\r\n<strong>Relevant Inputs\r\n<\/strong><span style=\"font-weight: 400\">The model needs signals that actually relate to future value.<\/span>\r\n\r\n<span style=\"font-weight: 400\"><strong>Enough Mature Historical Data<\/strong>\r\n<\/span><span style=\"font-weight: 400\">The model needs examples of what happened to similar users in the past.<\/span>\r\n\r\n<span style=\"font-weight: 400\"><strong>Regular Recalibration<\/strong>\r\n<\/span><span style=\"font-weight: 400\">User behavior, monetization, creative strategy, and acquisition channels change over time.<\/span>\r\n\r\n<span style=\"font-weight: 400\"><strong>Clear Forecast Definitions<\/strong>\r\n<\/span><span style=\"font-weight: 400;font-size: 16px\">Marketers should know what revenue streams and time horizon are being predicted.<\/span>\r\n\r\n<span style=\"font-weight: 400\"><strong>Validation Against Actuals<\/strong>\r\n<\/span><span style=\"font-weight: 400\">Predictions should be compared with realized cohort performance as the data matures.<\/span>\r\n\r\n<span style=\"font-weight: 400\">A model that cannot be validated is difficult to trust for budget decisions.<\/span>\r\n<h2><span style=\"font-weight: 400\">How Can an MMP Use Predictive LTV for Campaign Optimization?<\/span><\/h2>\r\n<span style=\"font-weight: 400\">An MMP already has acquisition context such as:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Network<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Campaign<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Creative<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Country<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Platform<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Spend<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">When predictive revenue is available in the same reporting environment, marketers can compare forecast value with acquisition cost much earlier.<\/span>\r\n\r\n<span style=\"font-weight: 400\">This supports decisions such as:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Scale a campaign showing strong predicted value<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Maintain spend while waiting for more data<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Reduce spend on a cohort with weak predicted economics<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Compare creative quality beyond CPI<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Prioritize channels expected to deliver higher long-term value<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">Tenjin's pLTV is designed for this use case: bringing a forecast into the same UA workflow where marketers already evaluate spend and actual LTV.<\/span>\r\n<h2><span style=\"font-weight: 400\">Predictive LTV and Hybrid Monetization<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Predictive LTV becomes especially valuable when users monetize in more than one way. However, there\u2019s a huge lack of those out there supporting apps that have a monetization mix. Many platforms continue to treat app monetization as if it exists as a single stream of IAA or IAP or subscription, never a hybrid or mix. This can turn into a measurement issue if you\u2019re using more than one revenue stream.<\/span>\r\n\r\n<span style=\"font-weight: 400\">For example, if a hybrid monetization game generated both IAA and IAP revenue and the model only predicts one stream, the total value could be understated. That\u2019s why Tenjin's pLTV combines ad mediation and IAP signals and it reflects a broader measurement principle: user value should be evaluated across the monetization streams that contribute to the business.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">Read more about this on our blog: <\/span><a href=\"https:\/\/tenjin.com\/blog\/predicted-ltv-for-hybrid-monetization\/\"><span style=\"font-weight: 400\">Predicted LTV for Hybrid Monetization: 4 Challenges Solved<\/span><\/a><span style=\"font-weight: 400\">.<\/span>\r\n<h2><span style=\"font-weight: 400\">Best Practices for Using Predictive LTV<\/span><\/h2>\r\n<b>Treat the prediction as an estimate.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">It supports decisions but does not replace realized performance.<\/span>\r\n\r\n<b>Know what revenue is included.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Different pLTV products can forecast different monetization streams.<\/span>\r\n\r\n<b>Compare predictions with actuals.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Validate the model as cohorts mature.<\/span>\r\n\r\n<b>Use pLTV alongside spend.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Predicted value becomes more useful when translated into expected ROAS or profitability.<\/span>\r\n\r\n<b>Segment by campaign and channel.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Blended predictions can hide meaningful acquisition differences.<\/span>\r\n\r\n<b>Avoid comparing cohorts at different definitions.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Make sure forecast windows and revenue inputs are consistent.<\/span>\r\n\r\n<b>Use enough data.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Very small cohorts can produce unstable signals.<\/span>\r\n\r\n<hr \/>\r\n\r\n<h2><span style=\"font-weight: 400\">Related Terms<\/span><\/h2>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/lifetime-value-ltv\/\"><span style=\"font-weight: 400\">Lifetime Value<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/predictive-analysis\/\"><span style=\"font-weight: 400\">Predictive Analytics<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/return-on-ad-spend-roas\/\"><span style=\"font-weight: 400\">Return on Ad Spend<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/subscriber-ltv\"><span style=\"font-weight: 400\">Subscriber Lifetime Value<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/subscription-event\/\"><span style=\"font-weight: 400\">Subscription Events<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/ad-revenue-attribution\/\"><span style=\"font-weight: 400\">Ad Revenue Attribution<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/in-app-purchases-iap\/\"><span style=\"font-weight: 400\">In-App Purchase<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/hybrid-monetization\/\"><span style=\"font-weight: 400\">Hybrid Monetization<\/span><\/a><\/li>\r\n<\/ul>\r\n\r\n<hr \/>\r\n\r\n<h2><span style=\"font-weight: 400\">Frequently Asked Questions<\/span><\/h2>\r\n<h4><span style=\"font-weight: 400\">What is predictive LTV?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">Predictive LTV is an estimate of how much revenue a user or cohort is expected to generate in the future based on early and historical data.<\/span>\r\n<h4><span style=\"font-weight: 400\">What is the difference between LTV and predictive LTV?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">LTV can describe realized value that has already accumulated, while predictive LTV forecasts value that has not yet occurred.<\/span>\r\n<h4><span style=\"font-weight: 400\">How is predictive LTV used in mobile marketing?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">Marketers use it to evaluate campaign quality and expected ROAS earlier, before cohorts have fully matured.<\/span>\r\n<h4><span style=\"font-weight: 400\">What does Tenjin's pLTV include?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">Tenjin's current N-Day All pLTV combines ad mediation and in-app purchase signals into a machine-learning forecast for mobile UA analysis.<\/span>\r\n<h4><span style=\"font-weight: 400\">Is predictive LTV accurate?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">Accuracy depends on the model, inputs, app, and cohort. Tenjin's current documentation reports approximately 90% average accuracy for its pLTV metric, but predictions should still be validated against actual performance as cohorts mature.<\/span>"]},"acf":[],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.5 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Predictive LTV (pLTV) | Tenjin<\/title>\n<meta name=\"description\" content=\"&quot;Learn what predictive LTV is, how LTV prediction works, and how mobile marketers use early revenue signals to forecast ROAS and optimize campaigns faster.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/tenjin.com\/zh\/glossary\/predictive-ltv-pltv\/\" \/>\n<meta property=\"og:locale\" content=\"zh_CN\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Predictive LTV (pLTV) | Tenjin\" \/>\n<meta property=\"og:description\" content=\"&quot;Learn what predictive LTV is, how LTV prediction works, and how mobile marketers use early revenue signals to forecast ROAS and optimize campaigns faster.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/tenjin.com\/zh\/glossary\/predictive-ltv-pltv\/\" \/>\n<meta property=\"og:site_name\" content=\"Tenjin\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:site\" content=\"@TenjinMMP\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"CollectionPage\",\"@id\":\"https:\\\/\\\/tenjin.com\\\/glossary\\\/predictive-ltv-pltv\\\/\",\"url\":\"https:\\\/\\\/tenjin.com\\\/glossary\\\/predictive-ltv-pltv\\\/\",\"name\":\"Predictive LTV (pLTV) | Tenjin\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/tenjin.com\\\/#website\"},\"description\":\"\\\"Learn what predictive LTV is, how LTV prediction works, and how mobile marketers use early revenue signals to forecast ROAS and optimize campaigns faster.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/tenjin.com\\\/glossary\\\/predictive-ltv-pltv\\\/#breadcrumb\"},\"inLanguage\":\"zh-Hans\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/tenjin.com\\\/glossary\\\/predictive-ltv-pltv\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/tenjin.com\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Predictive LTV (pLTV)\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/tenjin.com\\\/#website\",\"url\":\"https:\\\/\\\/tenjin.com\\\/\",\"name\":\"Tenjin\",\"description\":\"Growth Made Simple\",\"publisher\":{\"@id\":\"https:\\\/\\\/tenjin.com\\\/#organization\"},\"alternateName\":\"Tenjin - Mobile Measurement Partner\",\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/tenjin.com\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"zh-Hans\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/tenjin.com\\\/#organization\",\"name\":\"Tenjin\",\"url\":\"https:\\\/\\\/tenjin.com\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"zh-Hans\",\"@id\":\"https:\\\/\\\/tenjin.com\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/tenjin.com\\\/wp-content\\\/uploads\\\/2026\\\/04\\\/images.webp\",\"contentUrl\":\"https:\\\/\\\/tenjin.com\\\/wp-content\\\/uploads\\\/2026\\\/04\\\/images.webp\",\"width\":429,\"height\":117,\"caption\":\"Tenjin\"},\"image\":{\"@id\":\"https:\\\/\\\/tenjin.com\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/TenjinMMP\",\"https:\\\/\\\/www.youtube.com\\\/@TenjinMMP\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/tenjin\"]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Predictive LTV (pLTV) | Tenjin","description":"\"Learn what predictive LTV is, how LTV prediction works, and how mobile marketers use early revenue signals to forecast ROAS and optimize campaigns faster.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/tenjin.com\/zh\/glossary\/predictive-ltv-pltv\/","og_locale":"zh_CN","og_type":"article","og_title":"Predictive LTV (pLTV) | Tenjin","og_description":"\"Learn what predictive LTV is, how LTV prediction works, and how mobile marketers use early revenue signals to forecast ROAS and optimize campaigns faster.","og_url":"https:\/\/tenjin.com\/zh\/glossary\/predictive-ltv-pltv\/","og_site_name":"Tenjin","twitter_card":"summary_large_image","twitter_site":"@TenjinMMP","schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"CollectionPage","@id":"https:\/\/tenjin.com\/glossary\/predictive-ltv-pltv\/","url":"https:\/\/tenjin.com\/glossary\/predictive-ltv-pltv\/","name":"Predictive LTV (pLTV) | Tenjin","isPartOf":{"@id":"https:\/\/tenjin.com\/#website"},"description":"\"Learn what predictive LTV is, how LTV prediction works, and how mobile marketers use early revenue signals to forecast ROAS and optimize campaigns faster.","breadcrumb":{"@id":"https:\/\/tenjin.com\/glossary\/predictive-ltv-pltv\/#breadcrumb"},"inLanguage":"zh-Hans"},{"@type":"BreadcrumbList","@id":"https:\/\/tenjin.com\/glossary\/predictive-ltv-pltv\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/tenjin.com\/"},{"@type":"ListItem","position":2,"name":"Predictive LTV (pLTV)"}]},{"@type":"WebSite","@id":"https:\/\/tenjin.com\/#website","url":"https:\/\/tenjin.com\/","name":"Tenjin","description":"\u8ba9\u589e\u957f\u53d8\u7b80\u5355","publisher":{"@id":"https:\/\/tenjin.com\/#organization"},"alternateName":"Tenjin - Mobile Measurement Partner","potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/tenjin.com\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"zh-Hans"},{"@type":"Organization","@id":"https:\/\/tenjin.com\/#organization","name":"Tenjin","url":"https:\/\/tenjin.com\/","logo":{"@type":"ImageObject","inLanguage":"zh-Hans","@id":"https:\/\/tenjin.com\/#\/schema\/logo\/image\/","url":"https:\/\/tenjin.com\/wp-content\/uploads\/2026\/04\/images.webp","contentUrl":"https:\/\/tenjin.com\/wp-content\/uploads\/2026\/04\/images.webp","width":429,"height":117,"caption":"Tenjin"},"image":{"@id":"https:\/\/tenjin.com\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/TenjinMMP","https:\/\/www.youtube.com\/@TenjinMMP","https:\/\/www.linkedin.com\/company\/tenjin"]}]}},"glossary_term_description":"<div style=\"border: 1px solid #e5e5e5;padding: 16px;border-radius: 8px;background: #fafafa\">\r\n\r\n<b>Definition:<\/b>\r\n<span style=\"font-weight: 400\">\r\nPredictive LTV, or predicted lifetime value (pLTV) is an estimate of how much revenue a user or acquisition cohort is expected to generate in the future. In mobile marketing, predictive LTV uses early behavioral, revenue, and historical signals to help marketers evaluate campaign quality before the full user lifecycle has matured.\u2028<\/span>\r\n\r\n<\/div>\r\n&nbsp;\r\n<h2><span style=\"font-weight: 400\">What Is Predictive LTV?<\/span><\/h2>\r\n<a href=\"https:\/\/tenjin.com\/blog\/what-is-ltv-prediction\/\"><span style=\"font-weight: 400\">Predictive LTV<\/span><\/a><span style=\"font-weight: 400\"> is a form of <\/span><a href=\"https:\/\/tenjin.com\/glossary\/predictive-analysis\/\"><span style=\"font-weight: 400\">predictive analysis<\/span><\/a><span style=\"font-weight: 400\">. It is a metric used to estimate a user\u2019s value in the future, before the actual value has been generated.\u00a0<\/span><span style=\"font-weight: 400\">It is related to traditional lifetime value (LTV), which looks at past data and measures revenue that a user or cohort has already produced. Unlike traditional LTV, pLTV is a forecast. It uses the data available from early lifecycle signals to infer and estimate how much revenue users of a cohort are projected to generate at a later date.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">pLTV is extremely useful for user acquisition teams, who have to make decisions quickly. Marketers can\u2019t always wait weeks or months for cohorts to mature before deciding whether to add more budget and scale, maintain, or pause a campaign. <\/span><span style=\"font-weight: 400\">For example, a user acquired on Day 1 may eventually generate value through:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">In-app purchases<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Advertising revenue<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Subscription payments<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Renewals<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Other monetization events<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">A good pLTV metric is able to estimate and project value based on certain signals in the algorithm.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">Predictive LTV vs. Historical LTV<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Historical or <\/span><a href=\"https:\/\/tenjin.com\/glossary\/lifetime-value-ltv\/\"><span style=\"font-weight: 400\">traditional LTV<\/span><\/a><span style=\"font-weight: 400\"> measures realized value. Predictive LTV estimates future value. <\/span><span style=\"font-weight: 400\">Let\u2019s consider two user cohorts seven days after acquisition.<\/span>\r\n\r\n<b>Campaign A<\/b>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Day 7 realized LTV: $1.50<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Predicted future LTV: $4.20<\/span><\/li>\r\n<\/ul>\r\n<b>Campaign B<\/b>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Day 7 realized LTV: $1.70<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Predicted future LTV: $2.40<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">If we examine realized value, then Campaign B looks stronger than A. However, looking at the pLTV projections, Campaign A is the winner. The result is based off of the early signals from Campaign A. They could have had more sessions, better retention, higher engagement, stronger purchase intent, or a combination of all of these. <\/span><span style=\"font-weight: 400\">pLTV is a metric that helps identify early signal patterns and communicate their strengths before revenue is fully there.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">How Does Predictive LTV Work?<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Predictive LTV models typically combine early user behavior with historical patterns from mature cohorts. Some of the possible signals include:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Early revenue<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Sessions<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Retention<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Ad impressions<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Purchase activity<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Engagement depth<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Acquisition source<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Geography<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Platform<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">App behavior<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">The job of the predictive model is to learn relationships and correlations between early signals and later projections. The historical data for an app could show that users who completed 5 sessions, viewed 10 rewarded ads, and made a purchase are more likely to generate more LTV. So, when a new cohort of the same app starts to display similar behavior, the model can estimate a higher forecast of LTV.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">Every predictive analytics approach is a bit different, each MMP has their own algorithm. Every system uses their own models, signals, and forecast windows, so it's very important that marketers who use pLTV understand what the metric represents.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">What Is Tenjin's Predicted LTV?<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Tenjin includes Predicted Lifetime Value, or pLTV, as part of its revenue dashboard. The current <\/span><b>N-Day All pLTV (Ad Mediation + IAP)<\/b><span style=\"font-weight: 400\"> metric combines ad mediation revenue and in-app purchase signals into one machine-learning forecast. According to <\/span><a href=\"https:\/\/tenjin.com\/docs\/predicted-lifetime-value-pltv\/\"><span style=\"font-weight: 400\">documentation<\/span><\/a><span style=\"font-weight: 400\">, Tenjin\u2019s pLTV model:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Uses cohort sizes from Day 1 through Day 30<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Combines ad mediation and IAP signals<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Uses matured cohort and historical data<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Processes early behavioral and monetization signals<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Has approximately 90% average accuracy<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Can use up to 78 input signals<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">To add the metric to your revenue report, simply go to the add metrics section in the dashboard and search for pLTV. Because it is available directly from the dashboard, marketers are able to immediately evaluate their pLTV metric according to cohorts alongside their spend and LTV.\u00a0<\/span>\r\n\r\n<h2><span style=\"font-weight: 400\">Predictive LTV and Subscription Apps<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Predictive LTV is highly relevant when it comes to subscription apps because the subscription lifetime cycle takes longer to develop compared to many hypercasual and gaming apps. A newly acquired subscription cohort may only show some installs, trial starts, and some first payments. A month later, the same cohort could generate renewals, cancellations, reactivations, and reached a matured subscriber LTV.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">It is a similar pattern and timing issue that is observed with <\/span><a href=\"https:\/\/tenjin.com\/blog\/predicted-ltv-for-hybrid-monetization\/\"><span style=\"font-weight: 400\">hybrid monetization models<\/span><\/a><span style=\"font-weight: 400\">. And when under pressure, marketers may need to make a budget decision before the full subscriber lifecycle is visible.<\/span>\r\n\r\n<span style=\"font-weight: 400\">The early signals for subscription LTV take into account:\u00a0<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Trial start rate<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Trial-to-paid conversion<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Plan selection<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">First renewal behavior<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Early engagement<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Cancellation timing<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">Early signals differ for subscriber LTV by focusing on the value generated by paying subscribers. It could show that mature annual subscribers trend $80 on average. However, the predictive model may estimate a newly acquired cohort has the likelihood of generating $70.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">Similar to traditional or historical LTV, the subscriber LTV describes the value whereas the pLTV describes a projected subscriber value. Both metrics are useful for deciding on campaign economics.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">pLTV vs. pROAS<\/span><\/h2>\r\n<span style=\"font-weight: 400\">If pLTV is a projected estimate of user or cohort value, then pROAS would be a projected estimate of ROAS, or the revenue earned relative to ad spend. In order to calculate pROAS we use the simple formula:\u00a0<\/span>\r\n<pre>Predicted ROAS = Predicted Revenue or LTV \u00f7 Spend \u00d7 100\r\n\r\n\r\n<\/pre>\r\n<span style=\"font-weight: 400\">For example:<\/span>\r\n<ul>\r\n \t<li><span style=\"font-weight: 400\">Campaign spend = $10,000<\/span><\/li>\r\n \t<li><span style=\"font-weight: 400\">Revenue = $6,000<\/span><\/li>\r\n \t<li><span style=\"font-weight: 400\">pLTV = $14,000\u00a0\u00a0<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">In this example, the current realized ROAS is 60% and the pROAS is 140%. The pLTV projection suggests the campaign could earn beyond break-even as the cohort matures. This adds important context compared with early revenue alone. Rather than stopping the campaign based on its current ROAS, the team could continue running the campaign and monitor whether performance continues according to the projection.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">This shows that predictive LTV can be more useful than early revenue signals alone. Early signals are noisy: a small number of whales could make a new cohort seem stronger than it actually is. Likewise, a cohort with slow but steady purchases could look weak at the start, and end up with more long-term potential. The same principle applies to hybrid monetization and subscription apps: the first payment does usually reveal renewal quality.<\/span><span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\"> Good predictive models take into account a wider range of signals and add a trained metric to the decision context.\u00a0<\/span>\r\n<h2><span style=\"font-weight: 400\">What Makes a Good Predictive LTV Model?<\/span><\/h2>\r\n<span style=\"font-weight: 400\">A useful LTV prediction model should have:<\/span>\r\n\r\n<strong>Relevant Inputs\r\n<\/strong><span style=\"font-weight: 400\">The model needs signals that actually relate to future value.<\/span>\r\n\r\n<span style=\"font-weight: 400\"><strong>Enough Mature Historical Data<\/strong>\r\n<\/span><span style=\"font-weight: 400\">The model needs examples of what happened to similar users in the past.<\/span>\r\n\r\n<span style=\"font-weight: 400\"><strong>Regular Recalibration<\/strong>\r\n<\/span><span style=\"font-weight: 400\">User behavior, monetization, creative strategy, and acquisition channels change over time.<\/span>\r\n\r\n<span style=\"font-weight: 400\"><strong>Clear Forecast Definitions<\/strong>\r\n<\/span><span style=\"font-weight: 400;font-size: 16px\">Marketers should know what revenue streams and time horizon are being predicted.<\/span>\r\n\r\n<span style=\"font-weight: 400\"><strong>Validation Against Actuals<\/strong>\r\n<\/span><span style=\"font-weight: 400\">Predictions should be compared with realized cohort performance as the data matures.<\/span>\r\n\r\n<span style=\"font-weight: 400\">A model that cannot be validated is difficult to trust for budget decisions.<\/span>\r\n<h2><span style=\"font-weight: 400\">How Can an MMP Use Predictive LTV for Campaign Optimization?<\/span><\/h2>\r\n<span style=\"font-weight: 400\">An MMP already has acquisition context such as:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Network<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Campaign<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Creative<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Country<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Platform<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Spend<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">When predictive revenue is available in the same reporting environment, marketers can compare forecast value with acquisition cost much earlier.<\/span>\r\n\r\n<span style=\"font-weight: 400\">This supports decisions such as:<\/span>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Scale a campaign showing strong predicted value<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Maintain spend while waiting for more data<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Reduce spend on a cohort with weak predicted economics<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Compare creative quality beyond CPI<\/span><\/li>\r\n \t<li style=\"font-weight: 400\"><span style=\"font-weight: 400\">Prioritize channels expected to deliver higher long-term value<\/span><\/li>\r\n<\/ul>\r\n<span style=\"font-weight: 400\">Tenjin's pLTV is designed for this use case: bringing a forecast into the same UA workflow where marketers already evaluate spend and actual LTV.<\/span>\r\n<h2><span style=\"font-weight: 400\">Predictive LTV and Hybrid Monetization<\/span><\/h2>\r\n<span style=\"font-weight: 400\">Predictive LTV becomes especially valuable when users monetize in more than one way. However, there\u2019s a huge lack of those out there supporting apps that have a monetization mix. Many platforms continue to treat app monetization as if it exists as a single stream of IAA or IAP or subscription, never a hybrid or mix. This can turn into a measurement issue if you\u2019re using more than one revenue stream.<\/span>\r\n\r\n<span style=\"font-weight: 400\">For example, if a hybrid monetization game generated both IAA and IAP revenue and the model only predicts one stream, the total value could be understated. That\u2019s why Tenjin's pLTV combines ad mediation and IAP signals and it reflects a broader measurement principle: user value should be evaluated across the monetization streams that contribute to the business.\u00a0<\/span>\r\n\r\n<span style=\"font-weight: 400\">Read more about this on our blog: <\/span><a href=\"https:\/\/tenjin.com\/blog\/predicted-ltv-for-hybrid-monetization\/\"><span style=\"font-weight: 400\">Predicted LTV for Hybrid Monetization: 4 Challenges Solved<\/span><\/a><span style=\"font-weight: 400\">.<\/span>\r\n<h2><span style=\"font-weight: 400\">Best Practices for Using Predictive LTV<\/span><\/h2>\r\n<b>Treat the prediction as an estimate.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">It supports decisions but does not replace realized performance.<\/span>\r\n\r\n<b>Know what revenue is included.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Different pLTV products can forecast different monetization streams.<\/span>\r\n\r\n<b>Compare predictions with actuals.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Validate the model as cohorts mature.<\/span>\r\n\r\n<b>Use pLTV alongside spend.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Predicted value becomes more useful when translated into expected ROAS or profitability.<\/span>\r\n\r\n<b>Segment by campaign and channel.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Blended predictions can hide meaningful acquisition differences.<\/span>\r\n\r\n<b>Avoid comparing cohorts at different definitions.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Make sure forecast windows and revenue inputs are consistent.<\/span>\r\n\r\n<b>Use enough data.<\/b> <span style=\"font-weight: 400\">\r\n<\/span><span style=\"font-weight: 400\">Very small cohorts can produce unstable signals.<\/span>\r\n\r\n<hr \/>\r\n\r\n<h2><span style=\"font-weight: 400\">Related Terms<\/span><\/h2>\r\n<ul>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/lifetime-value-ltv\/\"><span style=\"font-weight: 400\">Lifetime Value<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/predictive-analysis\/\"><span style=\"font-weight: 400\">Predictive Analytics<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/return-on-ad-spend-roas\/\"><span style=\"font-weight: 400\">Return on Ad Spend<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/subscriber-ltv\"><span style=\"font-weight: 400\">Subscriber Lifetime Value<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/subscription-event\/\"><span style=\"font-weight: 400\">Subscription Events<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/ad-revenue-attribution\/\"><span style=\"font-weight: 400\">Ad Revenue Attribution<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/in-app-purchases-iap\/\"><span style=\"font-weight: 400\">In-App Purchase<\/span><\/a><\/li>\r\n \t<li style=\"font-weight: 400\"><a href=\"https:\/\/tenjin.com\/glossary\/hybrid-monetization\/\"><span style=\"font-weight: 400\">Hybrid Monetization<\/span><\/a><\/li>\r\n<\/ul>\r\n\r\n<hr \/>\r\n\r\n<h2><span style=\"font-weight: 400\">Frequently Asked Questions<\/span><\/h2>\r\n<h4><span style=\"font-weight: 400\">What is predictive LTV?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">Predictive LTV is an estimate of how much revenue a user or cohort is expected to generate in the future based on early and historical data.<\/span>\r\n<h4><span style=\"font-weight: 400\">What is the difference between LTV and predictive LTV?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">LTV can describe realized value that has already accumulated, while predictive LTV forecasts value that has not yet occurred.<\/span>\r\n<h4><span style=\"font-weight: 400\">How is predictive LTV used in mobile marketing?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">Marketers use it to evaluate campaign quality and expected ROAS earlier, before cohorts have fully matured.<\/span>\r\n<h4><span style=\"font-weight: 400\">What does Tenjin's pLTV include?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">Tenjin's current N-Day All pLTV combines ad mediation and in-app purchase signals into a machine-learning forecast for mobile UA analysis.<\/span>\r\n<h4><span style=\"font-weight: 400\">Is predictive LTV accurate?<\/span><\/h4>\r\n<span style=\"font-weight: 400\">Accuracy depends on the model, inputs, app, and cohort. Tenjin's current documentation reports approximately 90% average accuracy for its pLTV metric, but predictions should still be validated against actual performance as cohorts mature.<\/span>","_links":{"self":[{"href":"https:\/\/tenjin.com\/zh\/wp-json\/wp\/v2\/glossaries\/274","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/tenjin.com\/zh\/wp-json\/wp\/v2\/glossaries"}],"about":[{"href":"https:\/\/tenjin.com\/zh\/wp-json\/wp\/v2\/taxonomies\/glossaries"}],"wp:post_type":[{"href":"https:\/\/tenjin.com\/zh\/wp-json\/wp\/v2\/docs?glossaries=274"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}