> For the complete documentation index, see [llms.txt](https://help.smartico.ai/welcome/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.smartico.ai/welcome/products/ai-models.md). # AI Models Smartico's **AI Models** suite provides a comprehensive layer of predictive intelligence, automated segmentation, personalization algorithms, and generative assistance built specifically for iGaming and sports betting operators. By analyzing historical player behavior—including deposits, withdrawals, real-money wagering, game preferences, and promotional responsiveness—these models help marketing and CRM teams automate player lifecycle management, protect operator margins, and deliver individualized player experiences. ### Available AI & Predictive Models The AI suite is organized into four core functional pillars: #### 1. Predictive & Retention Intelligence * [**Churn & LTV Prediction**](/welcome/products/ai-models/churn-and-ltv-prediction.md): Projects player churn probability daily over 7, 14, or 30-day horizons across 6 operational risk ranks and forecasts 15, 30, and 60-day Lifetime Value with statistical confidence bounds. * [**Early VIP Detection Model**](/welcome/products/ai-models/early-vip-detection-model.md): Identifies high-potential First-Time Depositors within their first 14 days at key checkpoints with hourly intraday updates, assigning progressive tier bands for VIP host outreach. * [**Best Time Model**](/welcome/products/ai-models/best-time-model.md): Determines the optimal sending time for SMS, Email, and Push notifications per player, supporting custom delivery rules and an Activities Heatmap report. #### 2. Player Segmentation & Lifecycle Analysis * [**RFM Analysis**](/welcome/products/ai-models/rfm-analysis.md): Automatically categorizes active depositing players into 12 behavioral segments based on rolling 30-day Recency, Frequency, and Monetary quintiles, with 4 selectable calculation methodologies. * [**Value Score**](/welcome/products/ai-models/value-score.md): Ranks depositing players into 4 relative contribution tiers calculated across 30-day and lifetime activity windows. * [**Gamification Penetration Model**](/welcome/products/ai-models/gamification-penetration-model.md): Tracks deliberate player engagement across 5 gamification mechanics, identifies the highest net-deposit value gap, and recommends the next best gamification feature to promote. #### 3. Personalization & Optimization Engines * [**Favorite Product, Game, Game Type, Game Provider, Sport Type, Sport League**](/welcome/products/ai-models/favorite-product-game-game-type-and-game-provider.md): Uses a 30-day time-decay weighting model to determine product preferences, top-3 favored games, categories, providers, favorite Sport Type, and favorite league. * [**Recommended Bet and Deposit Amounts**](/welcome/products/ai-models/recommended-bet-and-deposit-amounts.md): Calculates personalized Conservative and Upsell amounts for Casino Bets, Sport Bets, and Deposits, supporting custom currency rounding rules and dynamic mission tokens. * [**Bonus Eligibility Model**](/welcome/products/ai-models/bonus-eligibility-model.md): Evaluates daily bonusing sustainability per player across four weighted signals to output actionable verdicts. #### 4. Generative & Content Assistance * [**AI Enhancer**](/welcome/products/ai-models/ai-enhancer.md): In-Back-Office assistant for drafting engaging promotional copy, shortening rules and terms, checking spelling, and validating HTML and Liquid syntax in email templates. ### How AI Models Integrate Across Smartico AI model outputs are seamlessly integrated across all operational areas of the Smartico platform: 1. **CRM Automation Journeys**: Trigger automated journeys when a player's predictive status changes (for example, entering Critical churn risk or moving to At Risk in RFM) or branch flows based on eligibility verdicts. 2. **Personalized Communication & Timing**: Schedule emails, SMS, and push messages at each recipient's individual Best Communication Time, and personalize copy using dynamic placeholders (such as favorite games or recommended deposit amounts). 3. **Dynamic Gamification**: Create personalized missions where the required wagering or deposit amount matches the player's calculated target, or tailor tournament invitations to the player's favorite game types. 4. **VIP & Host Operations**: Feed real-time queues for VIP hosts when new depositors are flagged as early VIP candidates, or when high-value players show signs of disengagement. 5. **BI & Reporting Dashboards**: Analyze player base health, track segment migration over time, and evaluate campaign performance across distinct player value tiers in Smartico analytics. ### FAQ **Q: How do I activate AI models for my brand?** Standard segmentation models, such as RFM Analysis and Value Score, are enabled by default across all accounts. Specialized predictive and machine learning models—including Best Time, Early VIP Detection, Churn & LTV Prediction, Favorite Games, Recommended Bet & Deposit Amounts, Bonus Eligibility, and Gamification Penetration—require activation. Contact your Smartico Success Manager or Support team to enable these features for your label, after which configuration options become available in Back-Office Label Settings. **Q: What data prerequisites and historical tracking are required?** AI models rely on continuous transactional and engagement data delivered through your platform integration (including deposits, withdrawals, casino bets, sports bets, game rounds, logins, and bonus redemptions). Most statistical and segmentation models require a rolling 14 to 30-day historical window to calculate baseline metrics. Advanced predictive models (such as Churn & LTV Prediction or Gamification Penetration) typically require 30 to 90 days of consistent transactional history to calibrate model weights accurately. **Q: How often are AI model scores and properties recalculated?** * **Daily / Nightly**: RFM Analysis, Value Score, Favorite Products and Games, Recommended Amounts, Bonus Eligibility, Gamification Penetration, and Churn & LTV Predictions are recalculated during the nightly processing window (typically between 02:30 and 04:30 AM UTC). * **Hourly / Intraday**: The Early VIP Detection model recalculates eligible new depositors hourly throughout the day to ensure host teams can act during live playing sessions. * **Near Real-Time**: When an operator modifies calculation settings in Back-Office (such as changing the RFM methodology or Recommended Bet logic), updated values recalculate immediately. **Q: How can I use recommended amounts in dynamic missions?** When creating a mission task (such as a deposit or wagering requirement), configure the condition to evaluate against the user's recommended amount property and insert the suggested value placeholder in the mission title and description. You can also enable the Lock Recommended Deposit/Bet Amount toggle on the task to freeze the target amount once a player begins progress, preventing subsequent nightly recalculations from changing the goal during an active mission. **Q: Why does a player show an "Unknown" status or empty properties in an AI model?** An "Unknown" or empty property indicates that the model did not have sufficient qualifying data to generate a reliable score, rather than a negative assessment. Common causes include: * The player is newly registered or has not yet completed the required activity (such as placing a bet or making a deposit). * The brand's integration does not cleanly separate real-money wagering from bonus wagering (required for Bonus Share calculations). * The feature was recently activated and the player has not yet passed through the nightly calculation cycle. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://help.smartico.ai/welcome/products/ai-models.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.