Sparks, Learning Locker and automation

February 3, 2019 - Dave Tosh

I've been working around learning data and xAPI off and on for nearly 5 years. When I started experimenting with Ben at HT2 on a new platform in 2013 we were trying to use xAPI to build personal learning plans for individuals, they would control their data and based on insights from that data, build out their plans and / or have plans suggested to them.

For various reasons it didn't work however the time was not wasted; out of that experiment came Learning Locker which is now the leading Learning Record Store (LRS).

Since I started working in this space, my focus has never been on visualising the data, while interesting, I think bigger potential lies in putting the data to work. Thanks to xAPI we have the data in a standardised format and in a powerful LRS (Learning Locker), so, the next step is using the data to power things like:

  • Contextually relevant feedback and support nudges.
  • Working out who might need extra assistance in subject X and automatically nudge them with helpful content.
  • Automate reporting for L&D managers as and when they need it.
  • Working out when and what to action for learners (personalisation) to maximise benefit for their development.

This is why Sparks exists, I wanted to build a tool that could make use of the data being gathered and put it to work helping those learning, training and teaching.

How does it work?

At its simplest, Sparks listens out for xAPI statements that meet some criteria at which point events trigger. Events consists of one or more "actions" such as sending a personalised email.

There are 4 different trigger types, each one handling xAPI data in a slightly different way: incoming, fetch, aggregate and analysis.

  • Incoming trigger: this is when an event listens out for a single incoming xAPI statement before triggering.
  • Fetch trigger: instead of waiting to receive an incoming xAPI statement, this option fetches data from Learning Locker and then triggers an event once the data has been fetched and parsed.
  • Aggregate trigger: this option allows you to listen out for multiple xAPI statements that meet and / or do not meet given criteria. This could be someone completing X, attempting Y but who has not done Z within a given timeframe.
  • Analysis trigger: this option consumes xAPI statements and runs analysis on the data before triggering the event. An example could be carrying out sentiment analysis on discussion board comments and then triggering the event if they are trending negative.

These trigger types provide powerful options to automate follow on actions which we will cover in more detail in future blog posts.

To find out more about Sparks and arrange a demo, head to - if you would like to find out more about Learning Locker or xAPI, visit

We’d love to hear from you

Sparks is in beta. Sign up if you would like to receive a demo and try things out knowing it is early days!

By supplying my email address and clicking "Sign up", I give Sparks permission to contact me with more information.

© 2020 PedalRed, Inc. All rights reserved.