> 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/favorite-product-game-game-type-and-game-provider.md).

# Favorite product, Detailed Sport, Detailed Casino

Smartico is making a statistical analysis of user activity and bringing the following information in the reporting and as properties available for segmentation:

* **Favorite product** - categorisation of users by preferences in products: casino only, sport only, more sport/less casino, etc.
* **Top 3 favorite games** - 3 games ranked by user preference.
* **Top 3 game types** - 3 game types ranked by user preferences
* **Top 3 game providers** - 3 game providers ranked by user preferences
* **Top 3 favorite sport types** - 3 sports ranked by user betting preferences (requires the **Favorite sport types and leagues** setting in **Label settings**; ask your Success manager to change it).
* **Top 3 favorite sport leagues** - 3 sport leagues or tournaments ranked by user betting preferences (requires the same setting).

{% hint style="info" %}
Note that populating these user properties requires activation. Please contact your Success Manager for more details
{% endhint %}

All these categorization types use a time-decay approach, meaning that if a user changes their preferences, the model captures those changes. For example, if the user played a lot of "*Golden slot*" a year ago, that will be less important than what they play now.

{% hint style="info" %}
"Favorites" of the player can be used in dynamic missions, with each player having a dynamically set personal goal. Read more about [dynamic missions](/welcome/products/gamification-basics/missions/dynamic-missions.md).
{% endhint %}

## Favorite product

The "**Favorite Product**" feature enhances marketing and gamification by identifying and categorizing user preferences based on their betting activity.

This feature analyzes recent betting behavior to help you understand which products—Casino, Sports, or Lottery—your users prefer. This allows for more targeted and personalized marketing strategies.

<figure><img src="https://77049817-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FfS5hl0PiysHtKAKMsQTe%2Fuploads%2Fgit-blob-d17e35b5e1638a7f9010aa791f4e5d01368f5787%2Fsmartico.ai%20%40%20EU2%202024-08-22%2017-26-41.png?alt=media" alt=""><figcaption></figcaption></figure>

### Logic Behind

The feature calculates user preferences by analyzing their betting activity over the last 30 days. Each bet the user places is assigned a weight, where more recent bets carry more influence than older ones.

This is done using a time decay logic, which smoothly decreases a bet's weight as it ages. The weighted bets are then aggregated to determine the percentage share of each product (Casino, Sport, Lottery) that the user engages with. This approach ensures that the shares accurately reflect a user’s current preferences, leading to more precise and timely user categorization.

### Available user properties

The following user properties are calculated daily for each active user and can be used in the segmentation for marketing and gamification activities.

<figure><img src="https://77049817-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FfS5hl0PiysHtKAKMsQTe%2Fuploads%2Fgit-blob-28607447bb0d671b3bfcd22bead97e4e9db37aff%2Fimage.png?alt=media" alt=""><figcaption></figcaption></figure>

**core\_fav\_product\_type** - indicates the user’s preferred product category based on their engagement with Casino, Sports, and Lottery products. Possible values:

1. Casino Only
2. Sport Only
3. Lottery Only
4. More Casino, Less Sport
5. Casino and Sport
6. More Sport, Less Casino
7. Mixed Preferences

**core\_fav\_product\_casino\_p** - the percentage of the user’s engagement attributed to Casino products.

**core\_fav\_product\_sport\_p** - the percentage of the user’s engagement attributed to Sports products.

**core\_fav\_product\_lottery\_p** - the percentage of the user’s engagement attributed to Lottery products.

### Viewing Favorites in the Back-Office

In the Back-Office User Profile under the **Favorite product** section, operators can view a player's calculated preferences: Product Preferences category, Casino/Sport/Lottery engagement percentages, Top 3 favorite casino categories, games, and providers, Top 3 favorite sport types and leagues, as well as AI-recommended bet and deposit amounts.

### Example of usage

Let’s say an operator plans a new promotional campaign focused on Casino products. Using the "Favorite Product" feature, one can segment his/her users based on their preferences and engagement.

**Case 1: Segmenting Users**

* Target Segment: Users categorized under Casino Only, More Casino/Less Sport, Casino and Sport, etc.
* Why: These users clearly prefer or strongly engage with Casino products, making them the ideal audience for your Casino promotion.

**Case 2: Enhancing Engagement**

* Track the campaign’s success by monitoring changes in core\_fav\_product\_casino\_p. If this percentage increases after the campaign, it indicates successful engagement.

**Case 3: Product Adoption**

* Target players who have a strong focus on Casino games to participate in sports betting, build promotional campaigns for such users, address mini-games, missions, and tournaments

## Detailed Casino & Sport Preferences

In the Games & Sports reports, you can now **identify the top favored casino games** by name or type over a specifically set time window. The report is based on user preferences and engagement, and it is based on the number of unique users, total bets, and share of favorite games.<br>

<figure><img src="https://77049817-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FfS5hl0PiysHtKAKMsQTe%2Fuploads%2Fgit-blob-e310aacd239ae74959e5359aac5c09a653942859%2FScreenshot%202025-06-19%20175207.png?alt=media" alt=""><figcaption></figcaption></figure>

While the report is showing the general behavior of your users, you can also use this information on a personal level.

### **Using game type, name, and provider in segmentation**

<figure><img src="https://77049817-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FfS5hl0PiysHtKAKMsQTe%2Fuploads%2Fgit-blob-856fd3cf66456349b98392e213e31ed4f577581d%2FSegments%20%5C%2023800%20-%20smartico.ai%20%40%20EU2%202025-07-03%2016-19-56.png?alt=media" alt=""><figcaption><p>Users with Baccarat as most popular game type</p></figcaption></figure>

<figure><img src="https://77049817-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FfS5hl0PiysHtKAKMsQTe%2Fuploads%2Fgit-blob-d4290b64972d3e1c29fe6e4a7e0ed6872b930d1e%2Fimage.png?alt=media" alt=""><figcaption><p>Users with BOOK OF CAMELOT or BOOK OF FROSTY in top 3 favorites games</p></figcaption></figure>

So, you can use properties like:

* **core\_fav\_game\_type\_top3** - indicates the user’s top 3 most-played game types, ranked by importance (slot, table...)
* **core\_fav\_game\_top3** - indicates the user’s top 3 most-played or preferred games, ranked by importance
* **core\_fav\_game\_provider\_top3** - indicates the user’s top 3 most-played or preferred game providers, ranked by importance
* **core\_fav\_sport\_type\_top3** - indicates the user’s top 3 most-played sports (e.g. Soccer, Basketball, Tennis), ranked by importance
* **core\_fav\_sport\_league\_top3** - indicates the user’s top 3 most-played leagues or tournaments (e.g. Premier League, NBA), ranked by importance

And, a special operator in conditions like "has on pos.1 any of" gives you the possibility to find the most popular game, type, and provider.

### **Using game type, name, and provider in the missions or tournaments**

You can target missions according to users' preferences, for example, to target "Blackjack" missions to users who like to play Blackjack

In dynamic missions, you can use dynamic UI tags in the mission title, sub-header, or description to display personalized recommendations in the player's preferred language: use `{{suggested_games}}` for their top casino game, type, or provider, and `{{suggested_sport}}` for their top sport type or league.

<figure><img src="https://77049817-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FfS5hl0PiysHtKAKMsQTe%2Fuploads%2Fgit-blob-e09e74d77a9664478e7d7f53d907d70663d1a0c5%2Fimage.png?alt=media" alt=""><figcaption><p>Mission on blackjack for users playing blackjack</p></figcaption></figure>

You can use standard segment filtering to target tournaments, jackpots, or other gamification mechanics to players with specific favorite games, sports, or leagues (dynamic property-to-property event matching and the {{suggested\_games}} and {{suggested\_sport}} tags are exclusive to Missions).

For example, a tournament targeting players of Gold Quest:

<figure><img src="https://77049817-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FfS5hl0PiysHtKAKMsQTe%2Fuploads%2Fgit-blob-c1f5733f6a18c1a1abcb56a57e011387c38c9d70%2Fimage.png?alt=media" alt=""><figcaption></figcaption></figure>

### Mission Locking for Favorite Casino & Sport Preferences

When creating dynamic missions based on favorite casino or sports preferences, you can enable the **Lock Favorite Game/Type/Provider** or **Lock Favorite League/Sport** toggle in the mission task settings. Because player preferences are recalculated nightly, a player's top games, sports, or leagues could otherwise shift while a mission is active. When locking is enabled, the player's favorite values are snapshotted at their first qualifying action or bet selection and frozen for the remainder of that mission, ensuring their target does not change mid-progress.

For personalized bet and deposit requirements in dynamic missions, see [Recommended Bet and Deposit Amounts](/welcome/products/ai-models/recommended-bet-and-deposit-amounts.md).

## FAQ

**Q: What factors are taken as the basis for Product preference calculation**

We are taking turnover for casino, sports & lottery.

**Q: What is the time interval of data used in the calculation**

Favorite product shares and favorite casino games, types, and providers use 30 days of historical betting data with exponential time decay. Favorite sport types and leagues analyze 90 days of sport betting activity.

**Q: How often are favorite properties recalculated?**

We aim to recalculate after midnight, within the time window from 2:30 to 4:30 AM UTC.

This time is chosen to have minimal impact on the overall system performance and have fresh betting activities already delivered from the integrated systems to Smartico.

**Q: Are there any weights used in the calculation**

We use an optimized exponential function delivering coefficients for each day, so that the highest weight is assigned to today's bets and a very small weight is assigned to the bets made 30 days ago. This way, we achieve a smart bias towards players' recent activity.

**Q: Does it include events that happen only after the feature activation, or does it take historical events too?**

The model includes all historical bets and is not dependent on the time of feature activation.

**Q: Are Favorite Casino Games and Favorite Sport Types/Leagues activated together?**

No. They are separate features that can be enabled independently. "Favorite casino games, game types and providers" controls top-3 casino games, game types, providers, and recommended bet/deposit amounts. "Favorite sport types and leagues" controls top-3 sport types and leagues. Each requires activation by your Success Manager.

**Q: Can I use favorite game property-to-property conditions in Automation Rules?**

No. Dynamic property-to-property matching (where each incoming bet event is checked against the user's personal top-3 favorites) is supported exclusively in Mission tasks. In Automation Rules, you can filter players by standard segment criteria (such as players having a specific game in their top 3), but you cannot dynamically match event properties to user favorite arrays.

**Q: What happens if a player has played fewer than 3 games or only one product?**

If a player has bet on only 1 or 2 games, their top-3 array property will hold only those 1 or 2 games, and the remaining positions stay empty. If a player has engaged exclusively with Casino, their favorite product category is set to "Casino Only", the casino share is 100%, and sport and lottery shares remain 0%.

**Q: Why is Core: Favorite Sport League, Top 3 empty while Favorite Sport Type is populated?**

Favorite sport leagues can only be calculated if your sportsbook integration provides league or tournament data alongside bet selections. If your sportsbook only sends sport categories without league metadata, the sport league property remains empty while top sport types are still ranked normally.

**Q: If there were favorite games defined, but then the user places no bets in 30+ days, does the model keep the value that was previously recorded?**

The model will preserve the user's latest favorite preferences, even if the user is no longer active.

**Q: Where do the game names in favorite games come from?**

Game names and categories are extracted directly from player betting and winning transactions delivered to Smartico, rather than static entries in the Game Catalog. If a game has never had bets placed on it by players, it will not appear in favorite rankings.

**Q: Can these AI favorite game recommendations be used to automatically reorder games in our casino lobby?**

No. Smartico's favorite games model powers Smartico's native segmentation, dynamic missions, tournaments, and widgets. It does not directly manage or reorder third-party casino lobby game grids. If you wish to use these preferences to drive external lobby display, your developers can retrieve player user state properties via the Smartico Public API.

**Q: How does the {{suggested\_games}} tag format game names in dynamic mission text?**

When the mission task condition uses "has any of", the {{suggested\_games}} tag automatically formats all available top favorite games into a comma-separated list in title case (for example, "Book Of Dead, Starburst, Sweet Bonanza"). When using position operators like "has on pos.1 any of", the tag resolves to only the single game at that rank.

**Q: How does the {{suggested\_sport}} tag format sport names and leagues in dynamic mission text?**

The {{suggested\_sport}} tag follows the same formatting logic as the games tag. When the mission task condition uses "has any of", the tag formats all available top sports or leagues into a comma-separated list in title case (for example, "Football, Tennis, Basketball" or "Premier League, NBA"). When using position operators such as "has on pos.1 any of", the tag resolves to only the single sport or league at that rank.

<br>


---

# 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/favorite-product-game-game-type-and-game-provider.md?ask=<question>&goal=<user_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.