Developing a learning data strategy is all about the outcomes. It’s about the change that will happen in your organization as a result of design decisions that you implement for putting data to work. The best way to create a learning data strategy is to think about not only what you have today, but to focus on where you want to go and what it is that you want your learners, managers, and stakeholders to be able to learn from and to do with data you make available to them.
In the following article, explore the methodology organizations are using to strategically connect their investments in learning data to business outcomes with a sustainable, future-forward approach.
Planning: The foundation for a successful data strategy
As with any successful strategy, the planning and research phase is the most important part. When it comes to designing a learning data strategy, organizations are working through planning as a four-step process—preparation, evaluation, evidence, and presentation.
- Preparation. The first stage is simple. Look around your organization and take inventory of all the tools, spreadsheets, rosters, platforms, and any other possible data sets that you already have in your learning ecosystem. Aim to answer the question “What data do we have?”
- Evaluation. Once you know what you have, you then need to evaluate the quality, usefulness, and impact of those data assets. This will allow you to build a solid understanding of the value of what you have, why you like what you like, and what you think needs to go. From this stage, you’ll be able to define “How do we use our data now?”
- Evidence. Now that you’ve evaluated the value of the learning data sources you already have, you’re able to start mapping out the availability of data as they relate to your organizational goals. During the evidence stage, define your critical questions or business objectives, work to identify the data that helps you answer those questions, and, if you can’t answer all of your questions, identify new data assets you’ll need to fill in the gaps. Here you’ll determine “What do we want to do with our data in the future that we can’t do now?”
- Presentation. If you’re going to make your data useful and strategically connected to the organization, you have to determine how to surface it to the right people in the most effective way. In this stage, you’ll need to determine how, where, and at what level you want your stakeholders and learners to be able to access information relevant to their needs. Don’t overlook this critical step. Identify and plan to have an answer for “How will each stakeholder in my organization put learning data to work daily?”
If you’re interested in seeing more best practices for planning and lessons learned, check out this Learning Data Strategy Workbook or these resources from SAS and O’Reilly that look at enterprise data strategy development. The latter two are not scoped specifically to learning data but provide excellent guidance and best practices from application in IT departments.
Implementation: Turning data strategy into reality
While the planning phase of a learning data strategy is always personalized to the organization, implementation can use out-of-the-box technologies, custom solutions, or some combination of both. A good place to start is to choose one or two well-defined business objectives that you focused on in the planning phase. From your data strategy planning, you should know what data assets you have that will support those business objectives. During implementation, your focus will be first on unifying the relevant data assets in interoperable formats, such as xAPI, and a central data store, such as a learning record store. If your data is not already in an interoperable format, you may need to complete xAPI integrations or data transformation processes to get your data in interoperable formats.
After your data is unified and standardized, you can begin to implement the presentation stage of your learning data strategy. Whether your organization is working with out-of-the-box business intelligence tools or designing custom dashboards and data visualizations, it’s critical to connect the existence of the data to the ability for stakeholders to apply the data.
For the learning and development team
Let’s look at a situation where an organization decided that one of their business goals for implementing a learning data strategy was to optimize their investment in a modular, modern learning ecosystem. Specifically, the organization had invested in several different learning platforms in order to provide a variety of learning resources and formats to their employees including an LMS, a learning experience platform, intranet, and discussion forums. However, the organization did not have access to analytics that provided transparency into if, when, and how their employees were engaging with the various learning platforms. By developing a data strategy that tied the investments they had made to the change they wanted to see, the organization was able to use a specific set of success metrics to understand and optimize platform adoption over time.
The following visualization examples (Figures 1 through 4) illustrate how the organization presented this unified data to their stakeholders.
Figure 1: This platform adoption data visualization illustrates adoption of each platform based on how much time employees spent engaging with content on the platform. The data card can be configured to illustrate adoption based on activity on the platform, as well as time spent.
Figure 2: This timeline data visualization illustrates learning activity on each platform over time, aggregated from across a group of learners
Figure 3: The platform calculator enables the stakeholder to monitor the actual adoption rate and cost per user based on near-real-time utilization counts
Figure 4: The activity hotspots data visualization illustrates when and how much employees are engaging with each platform in the learning ecosystem
With this kind of information readily available in near-real-time, the organization was able to sunset platforms that weren’t producing results, incentivize employee utilization of effective learning platforms and provide data-driven business cases for platform investment at the executive level.
For learners and other stakeholders
Once data is collected and understood at the aggregate level by the L&D team, many organizations will take the next step by making that data accessible to managers or individual learners directly. This most often will take the form of individualized data dashboards that have been scoped directly to the learner, manager, or executive using the permissioning structure from the organization’s single-sign-on system (Figure 5).
Figure 5: Creating a dashboard of visualizations that enable individual learners to see their own learning data can help them understand how they compare to their peers and the expectations for their role
By unlocking this level of data transparency through data visualization access directly to learners, organizations are able to significantly improve learner agency while increasing the rate and frequency of assignment completion. Managers are able to use near-real-time data on a daily basis to make interventions with direct reports, saving time, money, and headaches. Empowering learners and managers directly with learning data distributes accountability for effective talent development across all levels of the organization.
One important consideration in the effort to share learning data at all levels of the organization is the benefits that can come from a metered cascade of dashboard access. It can be quite effective to start by surfacing learning analytics at the administrative and executive levels first, then opening access to managers, and finally to individual learners. Through this metered approach, you will be able to demonstrate the value of your data strategy as early as possible to key stakeholders and executives, while allowing you to build internal subject matter experts and data evangelists, all of which will help you drive adoption and buy-in to your learning data strategy across the organization.
Evolution: Sustaining your data-driven learning ecosystem
Designing and implementing a data strategy is not a one-time effort. Just as the needs and goals of an organization change over time, so will the learning technologies available to you and the expectations of your workforce. Once you’ve implemented your learning data strategy and experienced the impact of the resulting data-driven outcomes, it’s time to develop an ongoing maintenance and enhancement plan that will ensure you evolve your learning data strategy alongside your evolving organization.
Here are some key items to consider as you evolve your learning data strategy:
- Revisit your strategic plan annually. The goals of your organization, and the resulting learning and development objectives, will change over time. A best practice is to evaluate your learning data strategy annually to ensure that the data assets you’re gathering and the success metrics you’re monitoring remain closely aligned to your current business goals. Including your IT and finance partners in this review process can help secure the appropriate resources to keep implementation needs up to date with your strategic plan.
- Evaluate emerging technologies. The availability and accessibility of new learning technologies improves constantly. Just think how far virtual reality technology has come in terms of becoming a standard part of the instructional designer’s toolkit! Staying apprised of emerging technologies is an important part of a learning and development professional’s responsibilities. But not every technology is going to suit the learning needs of your organization. Pilot testing a new technology can be a great way to assess fit with your organization. Defining the success metrics associated with the pilot test and building real time data flows from the new technology into your existing set of performance metrics will help you quickly determine value to your organization and build a data-driven business case for further investment.
- Optimize platform adoption and effectiveness. Optimization of learning platforms, and the associated investment, is also an ongoing effort. At least annually you should evaluate how each platform is contributing to learning engagement and business outcomes, assess current technologies against emerging technologies, and analyze your cost and ROI on each platform. The results of this assessment may alter specific success measures you have in place for each individual platform and/or the success metrics you are measuring across the entire ecosystem.
- Make a plan to implement advanced analytics. Through the implementation of your data strategy you will have created a pool of unified, standardized xAPI data. This pool of high resolution, time series event data represents a tremendous asset to your organization. As you gather more and more data over time, the value of that data only increases. A rich, relevant, and maintained data set enables advanced analytical functions and tools such as recommendation engines and predictive analytics. For an L&D professional, this can mean providing automated and personalized learning curriculums for your employees, or the ability to identify high performers who can be funneled into advanced leadership training, or even using smart analytics to more quickly identify employees experiencing performance declines and implement interventions earlier and more effectively. When you build a solid learning ecosystem foundation through a well-planned data strategy, the possibilities are powerful.
Just as every learner is unique, every organization is unique. From business goals to culture to the way employees like to work together, our job in learning and development is to take that uniqueness and translate it into learning experiences that drive business results. Which means that the process you use to develop your learning data strategy must tap into and be customized to that uniqueness. If you build your learning data strategy with a well-designed plan, implement piece-by-piece, and continue to evolve, you’ll be putting data to work for learners, managers, and executives each and every day.