The Events Page

The events page is a great tool for debugging the relationship between your code and the Web Console. To access it, select Events from the sidebar and choose All.

The Events List

When opening the events page, you will be presented with a graph of your incoming events and a list of the most recent ones. Events can be split into two categories - Exposures and Goals.

Exposure Events

Exposure events are sent by our SDKs when the treatment function is called. The treatment function asks our platform which variant a participant should see and returns a number between 0 and 3. When this question has been asked, we say that the user has been exposed to the experiment.

Goal Events

Goal events are sent by the SDKs when the track function is called and they include any properties that were passed along with it.

Raw Event JSON

Clicking on a single event will bring up a dialog box with the event's raw JSON data:

Event Filters

For debugging it can be useful to filter the incoming events to narrow them down to the specific events that you are looking for.

The available filters are as follows:

Application

Select the application(s) that you want to see events from. Applications are created and edited in the Dashboard Settings.

Unit Type

Select which unit types you want to see events from. Common unit types are user_id or anonymous_id, but they could be anything depending on your setup. Unit Types can be created in the Dashboard Settings.

Event Type

Filter your events to only show exposure events or goal events, as described above. This filter can also be preselected by choosing a type when clicking on the Events sidebar navigation.

Event Name

The event name can be extremely useful for debugging. For exposure events, the event name will be the name of the experiment that was exposed. For goal events, the event name will be the name of the goal that was tracked.

Environment Type

Choose whether to see events from either only Development environments or only Production environments. Both types of environment can be set up in the Dashboard Settings.

Unit UID

The Unit UID is the most specific form of filtering in the events page. As mentioned above, each event will be sent with a unique unit - often user_id or anonymous_id. These values, when passed into the SDKs, are hashed and sent to our platform in an encrypted form. If you only want to see values that have been sent by yourself, you can copy and paste your Unit UID hash into this box and only your own events will appear in the list.

The Exporter

ABsmartly also has functionality for exporting your events. To do this, click on Events in the sidebar and select Exports.

Here, you will be presented with a list of any previously created exports. For more information on any particular configuration, you can click on it.

Exporting Events

To set up a new export, click Export in the top-right hand corner.

Here you will be able to set up an export configuration that runs either once, or recurrently.

Configure Export

The first step on the exporter form is to give your configuration a name and to select how often you want it to run. This can be once, hourly, daily, weekly or monthly.

Filter Events

As mentioned in the Event Filters section, you can filter your events to fine tune what kind of events are exported. The Start At field is required and is the date and time of the earliest event that you want to export. For recurring exports, after the first export, the Start At event will be the next event after the last one that was previously exported.

Configure Storage

The Configure Storage section is where you will input the details of your storage bucket. The fields are described in the following table:

Field Name	Description
Bucket	The name of the object storage bucket to export into.
Prefix	The prefix of the exported filename, which can include / to denote "sub-folders".
Access Key ID	The access key id of for the object storage bucket.
Secret Key	The secret access key for the object storage bucket.
Endpoint	The S3-compatible endpoint to connect to the object storage bucket. Examples: AWS: https://s3.amazonaws.com GCP: https://storage.googleapis.com
Format	The format of the exported data. CSV and TSV include a header row with column names. JSON is a newline-delimited JSON file. Parquet is a very efficient column oriented storage format.
Compression	The compression method to use. GZIP is widely supported and provides good compression. ZSTD is similar to GZIP in compression level but faster, and less widely supported. LZ4 is much faster but provides slightly less compression. None is no compression.

Interoperability with GCP to S3

1. Go to the Cloud Storage Settings page in the Google Cloud Platform Console.
2. In Settings, select Interoperability.
3. If interoperability has not been set up yet, click Enable Interoperability Access.
4. Click Create new key.

Once all of these fields have been filled, you must use the Test Configuration button to make sure that the fields have been filled correctly. If the test is successful, you can click the Export Events button to start exporting!