> For the complete documentation index, see [llms.txt](https://docs.jentis.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.jentis.com/jentis-dcp-elements/synthetic-users/create-and-configure-your-model.md).

# Create and configure your model

### What happens when you create a model

When you create a Synthetic Users model, two things happen simultaneously. First, JENTIS creates the AI model configuration for your container. Second, JENTIS automatically generates a set of data-gathering tags that collect the behavioral signals the model needs for training. You don't need to create these tags manually — they appear in your tag management interface automatically.

Understanding what these tags do helps you make informed decisions about your model and troubleshoot issues later, so this page covers both the model creation process and the tags it generates.

### Step 1: Create the model

1. Navigate to Synthetic Users in the left sidebar of your JENTIS UI. Click Create First Model. \
   ![](/files/F4F1IKaniPfcOKJlA6IT)<br>
2. You'll see a model configuration screen. Select the container you want to create the model for. If you manage multiple websites, each container gets its own model — the AI trains separately on each site's behavioral data.\
   ![](/files/KXDr3lHMCX9b4Z4c2EFC)\ <br>
3. Then, choose the relevant tags. You may select all or only the ones relevant for your current tracking setup. If you have any doubts about which tags to choose, please refer to the Synthetic Users tags table below.\
   ![](/files/j1JogXZQCs070Jmm1EYl)\ <br>
4. Define URL parameters to distinguish one marketing campaign from another. In most cases the utm\_campaign URL parameter will suffice, but sometimes you might also encounter different naming conventions (e.g., pk\_campaign).\
   ![](/files/2jSMbeJNp09ITofXsztX)\ <br>
5. Click Create Model to generate the model.\
   ![](/files/9xqFrIclRPj9200QrAQG)

### Step 2: Publish and begin data gathering

Once the model is created, click Publish to activate data collection. The model now enters the GATHERING status — it begins collecting consented user behavioral data through the automatically generated tags.

#### Understanding model statuses

Your model progresses through four statuses during its lifecycle:

<figure><img src="/files/V6vzOvXPwmLE0PrfY5k7" alt=""><figcaption></figcaption></figure>

**NOT PUBLISHED** — The model exists but is not collecting data. This is the initial state after creation, before you first publish.

**GATHERING** — The model is actively collecting behavioral data from your consented users. The AI needs approximately 21 days of data to identify reliable behavioral patterns. During this period, no modeled conversions are sent to ad platforms. Sites with lower traffic (under \~300,000 sessions/month) may need a longer gathering period for the model to reach sufficient confidence.

**READY** — The AI has completed training and the model is ready to activate. At this point, you should review the dashboard metrics (estimated consent gap, projected recovery rate) and configure your advertising tools if you haven't already. When you're satisfied, publish the model again to activate it.

**ACTIVE** — The model is live. Non-consenting users are being assigned to behavioral segments, and modeled conversions are being sent to your configured advertising platforms using pooled click IDs.

{% hint style="warning" %}
Important hint: The new connector "Synthetic Users" will appear within your Tag Manager. It is mandatory to configure the consent as with other connectors, too.

Navigate to the "Legal Hub: Vendors" section and select the Synthetic Users. There configure the consent to be "Essential Mode" and make sure to link the consent to your Consent Management Platform (ie. if UserCentrics is your CMP provider then the according template ID must be configured so JENTIS resolves consent towards the "Synthetic Users" vendor based on a UserCentrics consent information).

The tool will only use consented users data for training purposes. In case of denied consent no data for training will be used, but it is important to configure the Essential Mode option still, as that makes sure that later (after the training period) the Synth. Users data can be used to randomly pick a click-id (for later conversion purposes).
{% endhint %}

### What tags are created automatically

When you create a Synthetic Users model, JENTIS automatically generates nine data-gathering tags in your container. Three are mandatory (the model cannot train without them) and six are optional (they improve model accuracy when available).\
\
Besides this, a new set of variables available for use when configuring your tags in Essential Mode will be added:

* A universal Synthetic User Click ID variable that provides a relevant artificial click ID value based on the tag in which it is used.
* Tool-bound Synthetic User gclid and Synthetic User fbclid variables that provide artificial click ID values for Google Ads and Meta to avoid data duplication when conversion is recorded with both IDs. More tools are coming soon.

<table data-full-width="true"><thead><tr><th width="138.98046875">Tag</th><th width="121.6875">Mandatory</th><th width="650.1640625">Description</th></tr></thead><tbody><tr><td><strong>User Raw Data</strong></td><td><p><img src="https://pf-emoji-service--cdn.us-east-1.prod.public.atl-paas.net/atlassian/check_mark_32.png" alt="" height="16" width="16"></p></td><td>Collects and maintains information about the Synthetic User itself.</td></tr><tr><td><strong>Session Raw Data</strong></td><td><p><img src="https://pf-emoji-service--cdn.us-east-1.prod.public.atl-paas.net/atlassian/check_mark_32.png" alt="" height="16" width="16"></p></td><td><p>Collects and maintains information about the Synthetic User Session. The following session data is collected:</p><ul><li>Navigator Language</li><li>Navigator Platform</li><li>Navigator User Agent</li><li>Window Screen Width</li><li>Window Screen Height</li><li>Referrer</li></ul></td></tr><tr><td><strong>Pageview Data</strong></td><td><p><img src="https://pf-emoji-service--cdn.us-east-1.prod.public.atl-paas.net/atlassian/check_mark_32.png" alt="" height="16" width="16"></p></td><td><ul><li><p>Collects pageview data for non e-commerce page types. The following page information is collected:</p><ul><li><p>Document Location Hostname</p><ul><li>Derived from a Client-Side Variable</li></ul></li><li><p>Document Location Pathname</p><ul><li>Derived from a Client-Side Variable</li></ul></li><li><p>Referrer</p><ul><li>Derived from a Client-Side Variable</li></ul></li><li><p>Content Category 1 (Optional)</p><ul><li>Derived from a <code>_jts.push()</code> property content– see the code example. [contentCategory1]</li></ul></li><li><p>Content Category 2 (Optional)</p><ul><li>Derived from a <code>_jts.push()</code> property content– see the code example. [contentCategory2]</li></ul></li><li><p>Content Category 3 (Optional)</p><ul><li>Derived from a <code>_jts.push()</code> property content– see the code example. [contentCategory3]</li></ul></li></ul></li></ul><p><strong>Code Example</strong></p><pre><code>_jts.push({
  "track" : "pageview",
  ...
  "contentCategory1" : "category1",
  "contentCategory2" : "category2",
  "contentCategory3" : "category3"
  ...
});
</code></pre></td></tr><tr><td><strong>Search Data</strong></td><td><p><img src="https://pf-emoji-service--cdn.us-east-1.prod.public.atl-paas.net/atlassian/cross_mark_32.png" alt="" height="16" width="16"></p></td><td><p>Collects data on completed searches on the website. The following search information is collected:</p><ul><li>Search Term</li><li>Count of Results</li><li>Found Product IDs</li></ul><p><strong>Code Example</strong></p><pre><code>_jts.push({
  "track" : "search",
  ...
  "term" : "shoe black",
  "countresults" : 145
  ...
});
</code></pre></td></tr><tr><td><strong>Productview Data</strong></td><td><p><img src="https://pf-emoji-service--cdn.us-east-1.prod.public.atl-paas.net/atlassian/cross_mark_32.png" alt="" height="16" width="16"></p></td><td><ul><li><p>Collects data from product detail page views on the website. The following product information is collected:</p><ul><li>Product ID</li><li>Product Name</li><li>Product Price</li><li>Product Discount</li><li>Product Category 1</li><li>Product Category 2</li><li>Product Category 3</li></ul></li></ul></td></tr><tr><td><strong>Add-To-Cart Data</strong></td><td><p><img src="https://pf-emoji-service--cdn.us-east-1.prod.public.atl-paas.net/atlassian/cross_mark_32.png" alt="" height="16" width="16"></p></td><td><ul><li><p>Collects data on items added to the cart. The following information is collected:</p><ul><li>Product ID</li><li>Product Name</li><li>Product Price</li><li>Product Discount</li></ul></li></ul></td></tr><tr><td><strong>Remove-From-Cart Data</strong></td><td><p><img src="https://pf-emoji-service--cdn.us-east-1.prod.public.atl-paas.net/atlassian/cross_mark_32.png" alt="" height="16" width="16"></p></td><td><ul><li><p>Collects data on items removed from the cart. The following information is collected:</p><ul><li>Product ID</li><li>Product Name</li><li>Product Price</li><li>Product Discount</li></ul></li></ul></td></tr><tr><td><strong>Order Data</strong></td><td><p><img src="https://pf-emoji-service--cdn.us-east-1.prod.public.atl-paas.net/atlassian/cross_mark_32.png" alt="" height="16" width="16"></p></td><td><ul><li><p>Collects data on successful orders/purchases on the website. The following information is collected:</p><ul><li>Order ID</li><li>Order Gross Value</li><li>Order Shipping Costs</li><li>Product IDs</li></ul></li></ul></td></tr><tr><td><strong>Scroll Tracking</strong></td><td><p><img src="https://pf-emoji-service--cdn.us-east-1.prod.public.atl-paas.net/atlassian/cross_mark_32.png" alt="" height="16" width="16"></p></td><td><ul><li><p>Collects scroll depth information from visitors on the website. The following data is recorded for each page:</p><ul><li>Scroll Depth ([10%, 25%, 50%, 75%, 90%])</li></ul></li></ul></td></tr></tbody></table>

### What's next

Your model is now in the GATHERING phase. While the AI trains (\~21 days), you can proceed to configure your advertising tools so everything is ready when the model reaches READY status.

<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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.jentis.com/jentis-dcp-elements/synthetic-users/create-and-configure-your-model.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.