Building an Advanced Data Pipe
Data Pipes are among the most powerful features of Datazoom.
They allow you to build data collection routing maps ranging from the simple, to the particularly complex.
In this tutorial we’ll expand upon what we covered in the last video and create a more Advanced Data Pipe which provides for data collection from multiple video players, wiring them to multiple destinations, and focusing on specific data points.
Here we are selecting two Collectors as the basis for our data collection strategy. Once you add a Collector to a data pipe you have the ability to select specific data points to capture. Data points fall into three categories, events, metadata, and FluxData.
Event data is collected every time the selected event occurs. Think along the lines of user interaction events or other events, like buffering, that occurs during playback.
Metadata is static data for an individual playback experience. Think of the type of device, asset title or user’s IP address.
FluxData includes special events that change throughout playback. Think along the lines of the video’s bitrate, playhead position, or available bandwidth. We set the interval for collecting these events when we create Connectors, and can adjust said interval on the fly.
Select as many or as few data points as you require. What’s more, you need not worry about being locked-in to these selections. You can edit combinations at any time, putting you in complete control of your data collection strategy.
Similarly, you can add as many Connectors as you require. In this case, we are adding three. Like data point selections, you are not bound to this configuration. You can add or remove Connectors at will, without having to re-deploy any code.
That’s it! Thanks for joining us for this series of Datazoom Data Pipes tutorials.