Snowplow - Kameleoon Documentation

Snowplow is a Behavioral Data Platform. With Snowplow, you can:

Generate a high-quality behavioral data set.
Control you data’s quality and compliance with governance tooling built at every stage of the solution.
Model behavioral data with customizable incremental web and mobile models that shorten time to value, all managed securely in the Snowplow BDP Console.

Key benefits:

Zoom in to individual customers and events.
Zoom out to compare behaviors between cohorts and over time.
Segment your audience by behavior.
Develop recommendations and personalizations engines.

With the Snowplow integration, you can now use Kameleoon’s campaigns events in any Snowplow Collector.

Activate Snowplow on Kameleoon

The first step is enabling Snowplow. To do this:

Log in to the Kameleoon app.
Click Admin > Integrations > Install the tool.
Select the projects on which you want to configure Snowplow.
Click Validate.

The integration will then appear in the Installed tab with the number of projects where you configured the tool.

The data will be automatically sent to Snowplow using Snowplow’s JavaScript Tracker.

If you need to change the behavior of the native bridge, you can create a custom analytics bridge and use the Snowplow sample code on the Kameleoon Github repository.

Kameleoon Snowplow event dispatcher

If multiple sources produce your Kameleoon events, the Snowplow event dispatcher integration is better suited for these use cases. This integration plugs directly into the event streaming layer. It singles out the exposure events affiliated with your project ID and dispatches them via an HTTP call to your Snowplow Stream Collector. To set up the dispatcher for a given project, provide your technical account manager with the following information:

The URL of your Snowplow Stream Collector
The host of your Iglu schema registry and your Snowplow API key.

Kameleoon uses the schema registry and API key to inform your Snowplow pipeline of the Kameleoon exposure event type’s custom schema, enabling your system to correctly interpret the data Kameleoon then sends. Kameleoon uses your API key only once. The schema of a Kameleoon exposure event that Kameleoon will push to your registry is the following:

    {
        "$schema": "http://iglucentral.com/schemas/com.snowplowanalytics.self-desc/schema/jsonschema/1-0-0#",
        "self": {
            "vendor": "kameleoon",
            "name": "kameleoon_experiment_exposure",
            "format": "jsonschema",
            "version": "1-0-0"
        },
        "type": "object",
        "properties": {
            "nonce": {
                "type": "number"
            },
            "visitor_code": {
                "type": "string"
            },
            "timestamp": {
                "type": "number"
            },
            "experiment_id": {
                "type": "number"
            },
            "variation_id": {
                "type": "number"
            }
        },
        "required": ["nonce","visitor_code","timestamp","experiment_id","variation_id"],
        "additionalProperties": false
    }

Where:

nonce is equivalent to the event UUID
visitor_code is a unique user identifier that serves as the user_id in the JSON event sent to Snowplow
timestamp is the timestamp of the event
experiment_id is the unique ID of the Kameleoon experiment the user encountered
variation_id is the unique ID of the variation the user encountered within the given experiment

Kameleoon batches exposure events into the HTTP call body as custom self-describing JSON. This format contains a reference to the descriptive schema, described in the preceding section, that Kameleoon pushed to your Snowplow Iglu schema registry during setup. You may have set up a Mapping Identifier for your project to, for example, attach the visitor ID that you use for your standard Snowplow events to your Kameleoon visitor sessions. If so, Snowplow uses this Mapping Identifier as the user_id parameter in the self describing event JSON sent to Snowplow instead of the Kameleoon visitor code. For more information on the HTTP format for sending events to Snowplow, you can refer to the Snowplow Tracker Protocol specification.

Associate Snowplow with a Kameleoon campaign

In the Graphic/Code editor

Click Integrations in the finalization panel.
Select Snowplow.

The summary pop-in will mention Snowplow before going live. Kameleoon will automatically transmit the data to Snowplow, and you can view your results directly in the tool.

On the personalization creation page

You can select Snowplow among the reporting tools for a personalization in the same way you would in an A/B test.

On the Results page

Once you set Snowplow as a reporting tool for an experiment, you can select (or clear) it via the Kameleoon Results page. To do this:

Click Reporting Tools > Edit.
Select or deselect Snowplow.
Click Validate.

On a feature flag

Learn how to manage third-party analytics integrations

​Activate Snowplow on Kameleoon

​Kameleoon Snowplow event dispatcher

​Associate Snowplow with a Kameleoon campaign

​In the Graphic/Code editor

​On the personalization creation page

​On the Results page

​On a feature flag

Activate Snowplow on Kameleoon

Kameleoon Snowplow event dispatcher

Associate Snowplow with a Kameleoon campaign

In the Graphic/Code editor

On the personalization creation page

On the Results page

On a feature flag