Mobile App Retention Guide: Measure, Benchmark, Boost

Installs create an opportunity. Mobile app retention shows whether users reach value, return to the product, and develop an ongoing relationship with it. It reflects the quality of acquisition, onboarding, product experience, performance, feature discovery, and post-install lifecycle engagement.

This guide explains how to measure a mobile app retention rate consistently, use app retention rate benchmarks in context, diagnose mobile app churn rate patterns, and improve results through targeted product experiments and automated lifecycle Journeys.

Mobile app retention takeaway
→ Day 1 reveals whether users activated and experienced initial value.
→ Day 7 reveals whether users found repeat value.
→ Day 30 reveals whether the app sustains a meaningful relationship or reactivates users effectively.
→ Category, platform, audience, and retention methodology affect what “good” looks like.
→ The best retention tactics post-install address the earliest observable drop-off first.

What is mobile app retention?

Mobile app retention is the percentage of users in a defined cohort who remain active during a later measurement period. A useful retention metric requires a documented cohort entry event and an active-user definition that represents meaningful product use rather than a generic app open.

A meaningful active event depends on the product. It may include:

• Completing a workout
• Placing an order
• Viewing account information
• Sending a message
• Completing a booking
• Using a core feature

Retention and engagement should be reviewed together, although they answer different questions.

Retention affects customer lifetime value, acquisition efficiency, monetization, and revenue forecasting. For a more tactical framework, see Mobile App Retention Strategies: Tactics for Day 1, Day 7, and Day 30.

How to calculate app retention

A cohort-based mobile app retention rate measures the share of an original user group that qualifies as active on a later day. Keeping the cohort fixed makes it easier to compare product releases, onboarding changes, acquisition sources, and lifecycle messaging programs. This approach keeps the original population consistent across each retention checkpoint.

Day-N retention rate = (users from the original cohort active on Day N ÷ total users in the original cohort) × 100

For example:

• 1,000 users install an app or become active on Day 0.
• 280 users qualify as active on Day 1.
• 80 users qualify as active on Day 30.

The Day 1 retention rate is 28%. The Day 30 retention rate is 8%.

Document these rules before publishing a retention dashboard:

• Cohort entry event: Install, signup, first session, or another defined event.
• Active-user definition: The session or meaningful event that qualifies a user as active.
• Time zone: The reporting time zone for cohort entry and retention windows.
• Measurement window: The exact day or time range used for each checkpoint.
• Included platforms: iOS, Android, web, or a documented combination.
• Identity rules: How users with multiple devices are counted.

The OneSignal Retention metrics documentation uses a fixed cohort. Users active on the selected start date become the cohort, and each later point shows the share of that same cohort active on a given day.

What do Day 1, Day 7, and Day 30 retention tell you?

Day 1, Day 7, and Day 30 retention are diagnostic checkpoints. They help teams locate whether the problem sits in activation, repeat value, or sustained value. These milestones are not universal targets because expected usage varies by category and business model. Each checkpoint helps teams focus investigation on the earliest observable retention weakness.

A strong Day 1 result does not guarantee Day 30 success. Users need a clear reason to return after onboarding.

Expected activity also differs by app type:

• A banking app may expect recurring account checks rather than daily sessions.
• A travel app may have infrequent but high-intent usage.
• A fitness, social, or productivity app may depend on frequent repeat behavior.

What is app cohort analysis?

App cohort analysis groups users by a shared starting point and tracks each group over time. Install date, signup week, acquisition source, country, platform, campaign, and onboarding path are common cohort dimensions that reveal changes hidden by aggregate retention reporting. It shows how retention changes across comparable groups of users.

A high-retaining cohort from an earlier period can make overall retention appear stable while newer cohorts are declining. Cohort analysis separates those patterns.

Use this workflow:

1. Create cohorts by install or signup week.
2. Compare Day 1, Day 7, and Day 30 retention for each cohort.
3. Segment results by platform, acquisition source, campaign, intent, and onboarding path.
4. Annotate product releases, bugs, pricing changes, campaign launches, and experiments.
5. Investigate the first material change in the retention curve.

Cohorts also distinguish product issues from acquisition-quality issues. A paid campaign may drive many installs from users with lower intent than an organic cohort. Retention can decline even when the underlying product experience remains unchanged.

For cohort setup requirements, including active-user definitions and reporting windows, review Mobile App Retention Strategies: Tactics for Day 1, Day 7, and Day 30.

What is the difference between mobile app retention rate and churn rate?

Retention and mobile app churn rate are complementary measures when they use the same cohort and timeframe. Retention counts users who remain active, while churn estimates the share that no longer meets the app’s definition of expected activity. The two metrics require matching cohort definitions and measurement windows.

Mobile app churn rate = 100% − retention rate

If Day 30 retention is 8%, the matching churn rate is 92%.

Use clear definitions across the organization:

• Retention: Users active at a selected checkpoint.
• Inactivity: Users who have not completed a meaningful action within an app-specific period.
• Churn: Users who are no longer expected to return or deliver value under the business’s definition.
• Uninstall rate: Users who remove the app.

An uninstall is useful evidence, but it does not capture all churned users. Some users stop using an app without uninstalling it. Others return after a long interval.

Set inactivity thresholds according to the app’s normal usage rhythm. A high-frequency app may identify risk after several inactive days, while a financial or travel app may require a longer window.

Classic vs. rolling retention: Which should you use?

Classic retention counts users active on an exact day, such as Day 7. Rolling retention counts users active on or after a selected day. Teams should choose one methodology for each report and label it beside every benchmark, dashboard, and experiment result. The selected method should match the app’s expected usage cadence.

Rolling retention produces a higher result for the same cohort because users have more opportunities to qualify. Classic retention works well when repeated behavior on a schedule matters. Rolling retention can better reflect value for apps used periodically by design.

Mobile app retention rate benchmarks

App retention rate benchmarks provide directional context rather than universal goals. A meaningful comparison matches the app category, platform, audience, business model, active-user definition, and retention methodology used in the team’s own reporting. They help teams identify performance questions while accounting for different product and user contexts.

Retention can vary by:

• App category and business model
• iOS versus Android
• Geography and acquisition source
• Organic versus paid cohorts
• Subscription, transactional, ad-supported, or marketplace use
• Exact-day versus rolling retention
• The event used to define an active user

What do app retention rate benchmarks actually mean?

A good mobile app retention rate depends on how users expect to use the product. Daily-use apps require frequent return behavior, while finance, travel, and utility apps may retain users through periodic but intentional activity. Benchmark interpretation must account for the value users expect from each type of app.

Consider the expected usage pattern:

• Daily-use products often depend on frequent repeat behavior.
• Financial apps may retain users through periodic account activity.
• Travel apps can have long intervals between high-value sessions.
• Subscription and free users may have different retention curves.

Avoid blending cohorts with different commercial relationships without labeling them. A useful benchmark comparison uses like-for-like cohorts with the same platform, market, acquisition source, user intent, and measurement method.

Why users leave: How to diagnose mobile app churn

Mobile app churn occurs when the product stops delivering expected value for a user. Lifecycle messaging can support retention, although it cannot replace clear onboarding, reliable performance, relevant content, or an acquisition promise that matches the product experience. Teams should diagnose where expected value first breaks down in the user journey.

Common churn drivers include:

• Users do not reach a first meaningful action.
• Onboarding asks for too much before delivering value.
• Crashes, errors, or slow load times interrupt the experience.
• Users do not discover features that create repeat value.
• Notifications are irrelevant, overly frequent, or poorly timed.
• The app does not adapt to preferences, behavior, or lifecycle stage.
• Acquisition messaging creates expectations the product does not meet.

Pair quantitative data with qualitative evidence. Review support contacts, user feedback, app-store review themes, and session recordings where available.

The critical retention windows: First session, first 72 hours, and first 30 days

The first session, first 72 hours, and first 30 days present different retention risks. Teams should define expected value in each window, observe user behavior, and intervene only when a message can help users complete their next useful action. Each window requires its own measurement, messaging, and product experience decisions.

First session: How do you get users to a meaningful first action?

The first meaningful action is the earliest behavior that demonstrates core product value. The first session should guide users toward that action with low friction, useful context, and one clear next step. Early onboarding should focus on helping users experience that value before requesting unnecessary information or permissions.

Improve the first-session experience by:

• Removing unnecessary signup gates.
• Reducing setup steps.
• Using progressive disclosure for advanced features.
• Guiding users toward one clear next action.
• Delaying nonessential permission requests until users understand the benefit.

Measure:

• Onboarding completion
• Time to first value
• First-session abandonment
• First key-event completion
• Day 1 retention

Onboarding For Apps That Have Nothing to Say Yet explains how early behavioral signals can support more relevant onboarding paths.

First 72 hours: How do you build a reason to return?

The first 72 hours should reinforce observed intent and help users continue a valuable action. Users arrive with different goals, permissions, preferences, and levels of progress, so generic reminders often provide limited value. Teams should use observed behavior to determine the most useful next action for each user.

Capture real-time signals such as:

• Platform
• Acquisition source
• Notification permission status
• Key screens viewed
• Incomplete setup
• First key action
• Session recency

Build behavior-based Journeys around those signals:

• Users who viewed a key screen but did not complete an action receive a deep link back to that screen.
• Users who completed activation receive relevant feature education.
• Users who have not returned after a defined window receive one helpful re-entry message.

Explain the benefit of notifications before requesting operating-system permission. Use frequency limits, suppression rules, and exit conditions to reduce notification fatigue.

First 30 days: How do you sustain value and detect fading habits?

Day 30 retention depends on reinforcing the product’s natural recurring value. Teams should track changes in expected behavior and tailor messages to user context instead of sending a preset number of reminders. This period shows whether users continue finding value after the initial activation experience.

Monitor:

• Session recency
• Repeat key events
• Feature adoption
• Purchase or subscription behavior
• Content consumption
• Notification opt-outs
• Email unsubscribes

An at-risk audience includes users whose expected behavior is decaying compared with their earlier pattern. Build separate lifecycle paths for:

• Newly activated users
• Engaged repeat users
• Users who adopted one feature but not another
• Users with declining engagement
• Users returning after inactivity

How to Re-Engage Inactive App Users Depending on Your Audience covers reactivation measurement, frequency controls, suppression rules, exit conditions, and downstream retention analysis.

How do you diagnose why users are leaving?

Retention diagnosis starts by finding the earliest point where users fail to receive expected value. Instrument the journey, segment the affected cohort, form one testable hypothesis, and evaluate downstream behavior after each product or messaging change. This process helps teams isolate the first material cause of retention decline.

Use this diagnostic framework:

1. Define the expected journey from acquisition to repeat value.
2. Instrument meaningful steps as observable events.
3. Find the earliest substantial drop-off in the funnel or cohort curve.
4. Segment the issue by cohort, source, platform, permissions, and user intent.
5. Form one testable hypothesis.
6. Run a controlled product, onboarding, or messaging experiment.
7. Measure later behavior, not only clicks or opens.

Review these signals together:

• Onboarding completion
• Time to first meaningful action
• First-session abandonment
• Day 1, Day 7, and Day 30 cohort retention
• Repeat key-event rate
• Feature adoption
• Session frequency and recency
• Notification delivery, engagement, and opt-outs
• Uninstalls, where available
• Support contacts and feedback themes
• Retention by campaign and acquisition channel

How to Increase App Retention and Avoid Mobile Churn provides a lifecycle process for identifying where expected value breaks down and connecting campaign outcomes to product behavior.

How to improve app retention

The best app retention strategies are sequenced. Fix the earliest product-value gap first, then reinforce return behavior with relevant lifecycle messaging. Push notifications can support retention when they are permission-based, timely, and useful to the recipient. Teams should measure whether each tactic improves meaningful actions and later retention.

1. How do you improve onboarding and time to value?

Onboarding should help users reach a meaningful action quickly. Every screen, field, tutorial, and permission prompt should support that action or move to a later point when the user has a clear reason to engage with it. The onboarding flow should remove friction that delays a user’s first experience of product value.

Test:

• Onboarding copy
• Screen order
• Signup friction
• Deep-link destinations
• Tutorial steps
• Permission-prompt timing
• Progressive onboarding paths

Measure:

• First-session completion
• Time to first value
• First meaningful action
• Day 1 retention
• Early opt-out rates

Onboarding For Apps That Have Nothing to Say Yet outlines how early user signals can branch onboarding flows.

2. How do behavior-based cross-channel journeys improve retention?

Cross-channel retention Journeys assign each channel a role based on user context, consent, and urgency. A coordinated plan helps teams avoid duplicate messages while giving users a relevant path back to the app or toward a useful next action. Each channel should support the user’s current stage and intended product action.

OneSignal supports Push, Email, SMS, RCS, in-app messaging, and Live Activities from one platform, as described on Why OneSignal?. OneSignal’s State of Customer Engagement report also found that apps using Journeys had 13.6% higher average 30-day retention.

Build automated Journeys that branch based on:

• User behavior
• Campaign response
• Lifecycle stage
• Permission status
• Product events
• Frequency history
• Conversion and exit events

3. How do you personalize retention messaging with real-time data?

Behavioral personalization uses what users do, prefer, and need to determine message content, timing, channel, and destination. The goal is to reduce friction by guiding users to a useful action instead of sending broad reminders. It uses current signals to match communication with the user’s relevant next step.

Useful segments include:

• Recently activated users
• High-value purchasers
• Users who abandon a key workflow
• Users who repeatedly use one feature
• Users who have become inactive
• Users who completed onboarding but have not adopted another core feature

Use deep links that bring users to the exact feature, item, content, or task mentioned in the message. Personalization should improve relevance and reduce effort rather than insert a name into generic copy.

OneSignal is a customer retention platform built around real-time behavioral signals and intent-driven communication. Why OneSignal? describes its approach to multichannel Journeys and behavioral segmentation.

4. How do persistent in-app experiences support feature discovery?

Persistent in-app experiences can help users discover features and useful content without relying only on notification permission. These surfaces remain available during future sessions, allowing users to explore relevant value when they are ready. They can guide feature discovery while users are actively engaged with the product.

Use in-app content for:

• Personalized recommendations
• Loyalty progress
• Feature announcements
• Educational tips
• Event reminders
• Post-purchase guidance
• Relevant content collections

Measure more than impressions. Track feature adoption, repeat actions, session frequency, conversion, and later cohort retention after exposure.

5. How do you re-engage lapsing users?

Re-engagement campaigns should respond to declining behavior using an inactivity definition that fits the app’s normal usage rhythm. The objective is to restore meaningful product use and later retention, not simply generate an app open. Each re-engagement path should connect the user with a relevant missed value or unfinished task.

Use this lifecycle sequence:

1. Identify an at-risk user through recency or behavior decay.
2. Diagnose the likely missed value or unfinished task.
3. Segment by behavior, value, preferences, and campaign response.
4. Select the message, channel, and deep-link destination.
5. Stop or change the Journey when users return, convert, opt out, or reach a frequency cap.
6. Measure repeat behavior and later retention after reactivation.

How to Re-Engage Inactive App Users Depending on Your Audience explains how to use frequency limits, suppression rules, and downstream retention measurement in reactivation programs.

6. How do you test mobile app retention strategies?

Retention improvements require controlled testing because short-term engagement can hide weak long-term outcomes. Start with aligned cohorts and a baseline, test one major variable, and measure whether users complete meaningful actions and return later. Controlled experiments show whether a change improves retention beyond immediate campaign response.

Test one major variable at a time:

• Onboarding flow
• Message timing
• Channel sequence
• Copy
• Offer
• Deep link
• Audience definition
• Frequency cap

Measure:

• Immediate conversion
• Completion of the intended key event
• Return frequency
• Day 7 or Day 30 retention after exposure
• Unsubscribes and permission changes
• Revenue, subscription renewal, or purchase behavior

How to Increase App Retention and Avoid Mobile Churn recommends establishing a baseline, comparing aligned cohorts, running controlled experiments, and reviewing downstream behavior.

Mobile app retention strategies in action

Retention evidence becomes more useful when it connects a specific intervention to meaningful user behavior. Testing campaign variables, responding to behavioral context, and measuring activity after engagement can provide stronger evidence than clicks alone. Case results should be evaluated through the retention principle and downstream activity they demonstrate.

According to Why OneSignal?:

• Sprocket increased overall 30-day activity by 66% after using OneSignal A/B testing to improve campaign performance.
• Beach Bum Games increased paid user reactivation by 140% within 10 days after implementing OneSignal.

These examples show practical retention principles:

• Test campaign assumptions instead of relying on one approach.
• Use behavioral context to create relevant reactivation paths.
• Measure downstream activity after users return.
• Treat results as case-specific rather than guaranteed outcomes.

Measuring what matters: Key mobile app retention metrics

Mobile app retention is an outcome metric. Activation, session behavior, feature adoption, message engagement, opt-outs, and revenue help explain why retention changes. Review these measures as a connected system rather than as isolated dashboards. Together, they help teams identify the behavior and experience changes behind retention movement.

How do you measure retention rate by cohort and time period?

Cohort-based reporting is the central retention view because it shows whether newer users retain better or worse than earlier groups. It also helps teams connect product releases, acquisition changes, and lifecycle experiments to changes in user behavior. This view makes recent changes visible across comparable groups of users.

Report Day 1, Day 7, Day 14, and Day 30 where the product’s usage pattern supports those checkpoints. Newer cohorts can reveal whether recent onboarding changes, releases, or campaigns improved the experience.

The OneSignal Retention metrics documentation explains how the retention chart keeps a cohort fixed and reports the percentage active on each subsequent day.

How should you read session frequency and session length?

Session frequency and session length show how users interact with an app after activation. Healthy patterns depend on the intended usage model, so teams should compare trends and audience segments instead of assuming longer sessions indicate stronger engagement. Review these measures alongside meaningful product actions and the expected use frequency.

Falling session frequency combined with declining key-event completion can act as an early warning signal. A utility app may provide value through short sessions, while a content or fitness app may require a different pattern.

What is the DAU/MAU stickiness ratio?

The DAU/MAU stickiness ratio measures how often monthly active users return daily. It provides useful context for usage frequency, although it should support cohort retention reporting rather than replace it. Teams can use it to track changes in daily return behavior among monthly users.

DAU/MAU stickiness = daily active users ÷ monthly active users

Compare stickiness over time and by audience segment. OneSignal’s State of Customer Engagement Report, referenced in Onboarding For Apps That Have Nothing to Say Yet, reports that apps sending onboarding messages have a 56.25% higher DAU/MAU ratio.

How should you use uninstall rate and notification opt-outs?

Uninstall timing can identify likely friction points, especially when a spike follows onboarding, a product release, or a campaign. Uninstall data remains incomplete because users can churn without removing the app. Review uninstall patterns with opt-outs and product behavior to understand likely communication or experience issues.

Review these early signs of communication fatigue:

• Notification opt-outs
• Email unsubscribes
• SMS opt-outs
• Permission changes
• Declining engagement after increased message frequency

Assess opt-outs alongside targeting rules, message frequency, user feedback, and post-message product behavior.

How does retention relate to customer lifetime value?

Customer lifetime value is the projected or realized revenue a user contributes during their relationship with an app. Retention can increase the number of users who reach repeat purchases, renewals, referrals, ad activity, or recurring product use. Teams should measure whether retained users complete the commercial actions that create revenue.

Teams should confirm the relationship between retention and revenue in their own data. An increase in retained users does not automatically increase customer lifetime value if users do not reach the relevant commercial actions.

How do you connect retention programs to product and revenue outcomes?

Message delivery, clicks, and app opens are intermediate indicators. Retention attribution should connect campaigns to the product or commercial outcome the campaign was designed to influence. Teams should define measurable downstream behavior before launching each retention program and compare results with an appropriate baseline.

Define the intended outcome before launch:

• Feature adoption
• First purchase
• Subscription renewal
• Recurring action
• Repeat session
• Replenishment
• Content consumption

Compare exposed users against an aligned baseline or holdout cohort where possible. A mobile app analytics tool and customer retention platform becomes more useful when it connects user behavior, Journey performance, and downstream outcomes in one workflow.

OneSignal vs. Braze retention: What should teams compare?

Teams evaluating OneSignal vs. Braze retention capabilities should compare how each platform supports the full retention workflow. Channel availability matters, although real-time data, orchestration, measurement, governance, and operational fit determine whether teams can run effective retention programs. The evaluation should focus on the team’s specific retention use cases and operating requirements.

Use this evaluation checklist:

• Can the platform capture and act on real-time behavioral signals?
• Can teams define audiences by lifecycle stage, activity, preferences, and key events?
• Can teams coordinate push, in-app, email, SMS, and other channels without duplicating messages?
• Can teams build automated Journeys with branching, frequency caps, suppression rules, and exit conditions?
• Can messages deep-link users to the exact next action?
• Can teams test and measure downstream conversion, opt-outs, reactivation, and cohort retention?
• Can product, growth, and lifecycle teams operate efficiently with appropriate engineering support?
• Does pricing and implementation fit the team’s scale and resources?

Why OneSignal? describes OneSignal’s approach to real-time, intent-driven multichannel engagement. Validate platform fit through required integrations, data model needs, channel coverage, analytics requirements, governance controls, and a retention use case specific to the business.

How to choose a push notification service for real-time user engagement

A push notification service should support a broader retention system rather than operate as a standalone broadcast tool. The goal is to respond to user intent with relevant messages while protecting users from unnecessary or repetitive communication. Teams should evaluate whether the service supports timely, behavior-based engagement across the retention lifecycle.

Evaluate these criteria:

• Real-time event triggers and behavioral segmentation
• Reliable delivery and platform support
• In-app messaging and multichannel coordination
• Journey orchestration, branching, frequency controls, and exit conditions
• Personalization and deep-linking capabilities
• Experimentation and outcome measurement
• Consent, preferences, and opt-out management
• Ease of implementation for marketing, product, and engineering teams
• Transparent pricing and the ability to scale

The best push notification service for real-time user engagement helps teams act on intent while reducing notification fatigue. Why OneSignal? outlines OneSignal’s multichannel approach across Push, Email, SMS, RCS, in-app messaging, and Live Activities.

Mobile app user retention FAQs

These mobile app retention FAQs answer common questions about measurement, benchmarks, churn, cohort analysis, messaging, and re-engagement. Each answer uses the same retention principles outlined in this guide: define meaningful activity, compare aligned cohorts, identify the earliest drop-off, and measure downstream user behavior after each intervention.

What is mobile app retention?

Mobile app retention is the percentage of users in a defined cohort who remain active at a later point. Teams commonly measure Day 1, Day 7, and Day 30 retention using an active-user definition that represents meaningful product use. The cohort entry event and active-user event should remain consistent across reporting periods.

How do you calculate app retention?

Calculate Day-N retention by dividing active users from the original cohort on Day N by the total original cohort, then multiplying by 100. Document the cohort entry event, active-user event, time zone, and methodology before comparing results. This documentation ensures that teams compare retention results calculated through the same rules.

Day-N retention = active users from the original cohort on Day N ÷ total users in the original cohort × 100

What is a good mobile app retention rate?

A good mobile app retention rate exceeds a comparable benchmark for the app’s category, platform, business model, and retention definition. OneSignal’s Mobile App Benchmarks of 2024 reports overall averages of 28.29% on Day 1, 17.86% on Day 7, and 7.88% on Day 30. Category-level comparisons provide more useful context than an all-industry average.

What is the difference between retention rate and mobile app churn rate?

Churn is the inverse of retention when both metrics use the same cohort and time period. Teams should define inactivity and churn according to the app’s natural usage cycle because a long gap may be normal for some categories. Matching the cohort and timeframe keeps the two measures comparable.

What are the best app retention strategies?

The best retention tactics post-install reduce time to value, improve onboarding, segment users by behavior, coordinate relevant messages across channels, support feature discovery in-app, re-engage lapsing users with context, and measure downstream retention through controlled experiments. Teams should address the earliest observable product-value gap before adding lifecycle messaging.

For a full lifecycle framework, see How to Increase App Retention and Avoid Mobile Churn.

How does cohort analysis improve app retention?

App cohort analysis shows whether recent groups of users retain better or worse than earlier groups. It can expose the effects of acquisition changes, product releases, onboarding adjustments, and lifecycle programs that aggregate retention can conceal. Teams can then investigate the first material change in the retention curve.

How do push notifications improve mobile app retention?

Push notifications can improve mobile app retention when they are permission-based, timely, behaviorally relevant, and linked to a useful in-app destination. Excessive or irrelevant messages can increase opt-outs and churn, so frequency limits and exit conditions are essential. Push should support a relevant user action within a broader retention system.

How do you measure whether re-engagement reduces app churn?

Measure meaningful actions after a user returns instead of treating an app open as the final success metric. Compare post-message conversion, repeat visits, later retention, and opt-outs against an aligned baseline or holdout cohort where possible. This measurement shows whether re-engagement restored useful product behavior beyond a single return.

How to Re-Engage Inactive App Users Depending on Your Audience explains how to measure reactivation, downstream behavior, and longer-term retention.