The ways in which employees access information, surface answers to questions, and find the right subject matter experts is shifting and drastically improving. With this, so is access to and potential efficiency of formal learning and training.

The question is, how do these elements fuse together in the reimagined ecosystem? What will performance support, formal learning and training, or upskilling and reskilling look like when we combine the best of digital and asynchronous tools, as well as synchronous and in-person endeavors. 

Our series launched with L&D Must Resolve ‘Content Chaos’ to Meet Performance Goals, an overview of the biggest challenges organizational development and L&D face right now.

We continued the series with Can Skills Frameworks Close the Skills-to-Performance Gap?, where we look closely at the challenges and potential values and limits of skills and competency frameworks, including some practical advice and insights gained from working with a large variety of small, medium, and enterprise businesses.

Our third piece, Performance Support in the Flow of Work—Now and Going Forward, explores what employee performance support, in the moment of need, can look like. We describe a future of mega-stores of digital assets that will no longer be seen as impossible to keep up-to-date. No more unmanageable piles of information, impractical at surfacing the right information effectively and efficiently, at the point of need. Quite the opposite; we see these resources forming the backbone of employee performance support.

Personalized or adaptive learning: What we mean

In this article we dive into the lesser-explored role of AI in improving more formal digital learning programs. Our exploration starts with the impact of AI tools on design, build, and curation, via the delivery of hyper-personalized learning journeys for each learner.

In the past, and documented well in a Guild research piece from 2021, different understandings of “adaptive” (or “personalized”) learning have been used in our industry.

We today, and for the purpose of this series, differentiate personalization at the macro-level as opposed to “truly adaptive learning,” adapted or personalized at the micro-level.

  • Macro-level personalized content are pathways, courses, or content pulled together for a specific job role, through social content recommendation algorithms or similar such mechanisms. While the suggested content or courses are chosen specifically for a person or group (personalized), the actual delivery within each course or module is still a one-size-fits-all, linear eLearning course.
  • Truly adaptive learning, adapted at a more granular, or micro-level, consists of a network of content—text, slides, audio, video, practice questions, case studies, worked examples, recall questions, assessments, etc.—that is traversed by the learner in a unique fashion tailored to that learner’s pre-existing knowledge, their rate of progress, and even their confidence levels.

Simply put, macro-level personalization is the librarian choosing the right books for the learner and handing them over, maybe even suggesting an order in which to read them. Truly adaptive learning is—paragraph by paragraph, question by question—adapting the journey, through the content delivered, to the learner continuously, in real-time.

At a granular learning level, we explore the exciting potential of detailed learning analytics. These analytics will derive insights that enable and empower organizations to make data-informed decisions on learning, training, and content strategy that impact everything—from tightly aligning training and learning needs with business goals and outcomes to maximizing the impact and return on investments made for in-person training, coaching, and mentoring.

Efficiently tag and structure content using AI

AI can now power contextualization engines, allowing content to efficiently be organized, tagged, managed, and updated. All of this is key in a world of growing data complexity, where organizing and tagging by hand which, besides being highly error-prone, is simply no longer an option due to the amount of information.

In addition, content can now be mapped, in seconds, with respect to its intrinsic hierarchy. Let’s take the example of a modern language or science textbook. These often have only one way in which the order of pages makes sense. Instructional scaffolding means that new concepts are introduced one at a time, with foundational concepts covered first.

This hierarchy is automatically there, within any body of knowledge or information. AI tools are able to uncover this hierarchy at lightning speed and at scale, or, in the case of gaps in the hierarchy, identify these and inform further instructional design steps or other steps to plug these gaps.

What’s new here is that these mappings aren’t “set in stone” like file and folder structures on an organization’s system. Rather, users can map and remap at will, all without altering original versions, similar to the way that internet search engines provide results at will.

This flexibility produces a multiplier effect, enabling the repurposing and reuse of content, while reducing the duplication of investment and effort. Using AI tools in this way achieves agility, efficacy, efficiency, and effectiveness.

Enrich & expand learning opportunities

Yes, this is extremely powerful for any learning record store, LMS, LXP, or knowledge hub. But what does it mean for formal learning?

For formal learning, this means we are now able to draw from content across an ecosystem, whether the content was originally designed with learning in mind or not. Learning journeys can easily adapt to and cater for specific learning objectives, outcomes, and levels of expertise and capability building. Once the content items have been gathered together and mapped into the hierarchy tree, learning designers can explore:

  • How far the content covers the objectives and outcomes
  • How densely populated different branches of the tree (or the “modules”) are
  • The balance of instructional methodologies
  • How well the existing questions cover the concepts and all branches of the tree to empower a fully personalized learning journey

This approach, utilizing AI tools, is completely changing the design process and the designers’ agility for building programs of learning. An additional efficiency is that, with accurate hierarchies of content, no SME is required to advise where exactly a certain question should sit; the AI system “knows” where to position the question.

In short, accurate content and assessment hierarchies empower fully adaptive learning.

Adaptive learning is relevant

Adaptive learning engines are now able to navigate the learner through this course, continuously assessing prior knowledge and needs with questions and answers.

The learner experiences a fully adaptive, truly personalized learning journey. Where the learner has preexisting knowledge, proven through answering questions, the course will not require them to view and spend time on content they are already proficient in.

However, where there are gaps and where learning is required, the user will be given the relevant content to build their understanding and confidence. Additionally, via further, questioning, the AI will continuously adapt the pace of progress.

With time often cited as the top barrier to learning, and compliance being one of those main time drains, imagine the potential gains in productivity and employee experience. Adaptive learning also reduces cognitive overload and frees employees to develop and broaden their skills. Additionally, it allows the development of domain expertise and mastery—known as vertical learning. This moves us away from content consumption and surface-level horizontal learning.

Adaptive learning is efficient

The easy availability of asynchronous, fully personalized adaptive learning will drastically improve learning efficiency, in line with Bloom’s famous paper, the 2 Sigma Problem.

Allowing learners individualized access to spaced repetition and practice with relevant content marks the end of one-size-fits-all digital learning. It efficiently supports learners in moving from the basic and pure knowledge (“rote learning”) to transferable knowledge and skill. Furthermore, it allows learners to be more self-aware of their learning and their progress.

Data aids in gauging progress

Early overconfidence in learners, coined the Dunning-Kruger effect, is detected through continuous questioning, alongside confidence measures; and, just like any gap in knowledge or confidence, should be addressed in any meaningful learning environment.

As we noted in Can Skills Frameworks Close the Skills-to-Performance Gap?, using AI and data this way will help eradicate the subjectivity of skills self-assessment by providing robust and reliable data to enable equitable, data-informed career and talent decisions.

However, there’s even more. AI’s role in implementing essential hierarchy trees and ensuring coverage across breadth and depth of content, combined with continuous assessment, means that all learning is measured.

This is true of measuring the strengths and weaknesses for individual learners—and for whole teams, whose proficiency can be mapped, analyzed, and reacted upon.

Together with performance data and job KPIs, we’re realizing a connected learning ecosystem that will empower organizations to strategize and prioritize future learning as well as human capital development and deployment. Learning leaders can combine and overlay team data with individuals’ data to support employees’ long term performance and development.

The efficiency gains are not limited to digital learning. In-person training, coaching, and mentoring can also be deployed in a more target-driven fashion, utilizing rich learning data analytics and performance KPIs side by side and providing a holistic picture.

Maximize performance improvement

Thus, the picture we are painting here is a world that does not purely consist of digital learning, training, reskilling, and upskilling. Rather, we envision a seismic opportunity for utilizing digital tools side by side with in-person learning and training efforts. For every role. For every skill or competence.

Every organization, no matter what its unique context, products, and services, can leverage the combination of these strategies that proves to be the most impactful and effective. The key lies in utilizing all of the tools and methodologies at our disposal, in the most effective and efficient ways possible, to maximize impact on performance improvement.

Envisaging this shift to a more agile and formal learning & development environment, together with the performance support in the moment of need discussed in Performance Support in the Flow of Work—Now and Going Forward, a much richer picture of a connected ecosystem emerges, reimagining how the employee of the future will learn, develop, and be supported in the flow of work—at the point of need.