Learning Ecosystems: How to add data sources & iterate [Guide]

     

Last week, Tim Dickinson explained how to create a strategy for your modern learning ecosystem. In this post, we’ll dig more into the weeds of getting started and look at how to choose a good starter data source, connect it, and build from there.

So, what's a data source again?

What is an xAPI data source?In this instance, a data source is any system that generates or stores data about a learner’s interactions. This includes learning systems, operational systems, and performance-monitoring tools. For example, your learning ecosystem might include:

  • an LMS,
  • a Learning Experience Platform,
  • external content providers, and/or
  • a performance observation checklist tool.

Your ecosystem also might be integrated with the sales CRM and the organization’s data warehouse for business KPI data.

Remember: Your organization already has a learning ecosystem (or several), but it just might not be integrated or planned. Data sources are normally applications already in your organization that need to be connected.

How do I choose a good starter data source?

In the United Kingdom, there’s a brand of pet hamster cage (Rotastak) that allows you to connect various modules with plastic tunnels, which lets hamsters climb and explore every aspect of their environment.

There are sleeping modules, exercise wheels, and larger cage elements for storing food, holding water, or burrowing in the deep sawdust. And, because you have all these options, you can create quite elaborate setups. The idea is that you keep adding modules and tunnel sections to your hamster’s cage piece by piece.

How to choose learning ecosystem data sources

A learning ecosystem is a bit like this example in that there can be many different yet connected elements. You can’t just start with any module, though. For instance, a hamster won’t be happy if he only has an exercise wheel, but no bed compartment or water bottle—which is why Rotastak sells starter hamster setups that include all the basics. Similarly, it’s important to start with the right components when connecting your learning ecosystem.

So, what makes a good starting point when connecting up your ecosystem? In short, you’ll need:

  1. an access point for learners,
  2. a system to track data about learner activities,
  3. and a database to hold that data for analysis.

Real-World Example: Visa University

Visa is a great example of starting with one data source and iterating. (NOTE: We’ll go more in depth with their learning ecosystem in an upcoming post.)

Visa Learning Ecosystem

Visa chose a Learning Experience Platform (LXP) as a starting point. Starting with an LXP (i.e. Pathgather) was a good move for Visa because:

  1. From a learner perspective, the LXP is the heart of the learning ecosystem. It’s the central place where learners go to access all their learning content. When setting up a learning ecosystem, an important principle is ensuring you include the places where people go for learning—which is exactly what Pathgather is for Visa.

  2. Pathgather is already connected to all the different elements of the ecosystem. It’s integrated with Workday for HR information, with all the external content vendors, and Visa’s internal learning systems. In fact, by integrating Pathgather, Visa also was able to access some basic data about utilization of these other systems because they could track when people launched these systems from Pathgather. This basic data was then augmented with more detailed data in a later iteration.

  3. Finally, Pathgather’s xAPI implementation process is easy, especially when connecting to a Learning Record Store (LRS). This ease of connectivity is super important at the early stages of developing your ecosystem because you can get a quick win and start collecting and making use of data as soon as possible.

Integrate xAPI into your learning ecosystem.

In Visa’s case, they were in the process of implementing Pathgather as a new system, so they were able to require that Pathgather implement xAPI as part of the RFP.

This is really important: If you’re even considering implementing an integrated learning ecosystem in the next five years, make sure xAPI is a requirement for every new system you bring in. We recommend adding the following requirement to your RFP:

Sends detailed data about learner interactions via xAPI to an external Learning Record Store (LRS) that passes the ADL conformance test suite.

(NOTE: Every Watershed Certified Data Source conforms to xAPI.)

How do I get connected?

Not all applications support xAPI, though, and other options may include importing CSV database dumps or building connectors to translate the data. It’s common to use a mixture of methods to integrate different data sources.

If you’re building a connector, working with CSVs, or partnering with a vendor that implements xAPI, you don’t need to implement tracking for everything up front. For example, Caterpillar, a global manufacturer, features both Kaltura MediaSpace™ and Inkling as part of their learning ecosystem.

Both of these products first implemented tracking for basic events, but now include tracking for additional learner interactions, which is based on feedback from stakeholders. So, if you find a vendor that’s willing to grow its xAPI implementation with you, you’re onto a good thing.


NOW PLAYING: What's a Learning Ecosystem?

Need a quick refresher? Watch this "pop-up book" video from our friends at xapiapps.


Iterate! What’s next?

Once you’ve connected your first data source, it’s time to iterate! This might mean adding another data source, or going deeper into one you’ve already connected. Either way, try to iterate in a logical way, while adding tracking methods and data sources related to what you’ve already connected.

Some organizations will iterate in a mostly linear way, adding one data source after another. Other organizations start to bring in more people and teams and accelerate the rate of growth as they add more systems. Both approaches are valid, depending on the unique needs of your organization.

This also might be a good time to think about gaps in your learning ecosystem. Avoid adding new products just for the sake of it—especially if people are using systems that are already in place. If you have genuine gaps in your learning offerings, however, consider how you might fill them in complementary ways. Maybe you need a video platform such as Kaltura that can add video streaming and webcasts to your toolkit. Or, perhaps you need a performance observation checklist such as xapiapps to start collecting data about real world performance. Whatever the need, remember to include xAPI in your requirements. 

Up Next: How one company replaced its LMS with a learning ecosystem

In our next post, we'll show you another real-world example of how a large organization replaced its LMS with a learning ecosystem. Don’t miss the next update in our Learning and Data Ecosystem blog series—sign up for Watershed Insights to get the latest blog posts and updates sent right to your inbox!


Need some help choosing data sources?

Download this handy list to learn more about different types of data sources and find ones that fit into your organization's learning ecosystems. 

Learning Ecosystem Data Source Guide

Andrew Downes

About The Author

As one of the authors of xAPI, Andrew Downes has years of expertise in data-driven learning design. And, with a background in instructional design and development, he’s well versed in creating learning experiences and platforms in both corporate and academic environments.