The Content Era Was a Good Run. It Is Coming to an End.
If you have been working in enterprise L&D for the past decade, you have lived through the content era. The defining strategic question of that period was some version of: how do we get the right content to the right people at the right time?
The industry responded with energy. Content libraries expanded. Authoring tools got faster. Micro-learning became a category. Video production budgets grew. User-generated content platforms launched. Content curation emerged as a skill set. The market for off-the-shelf learning content became a multi-billion dollar industry.
And it worked, to a point. Organisations that invested in high-quality, well-organised content did see improvements in learner engagement, compliance completion rates, and time-to-competency for structured programmes.
But the returns on content investment have been flattening for several years, and the reason is structural. Content, no matter how good, is passive. It sits in a library and waits for someone to find it, or for an administrator to assign it. It does not adapt to the learner. It does not respond to the context of their work. It does not get smarter over time.
The next era of enterprise learning will be defined by a different organising principle. The question will shift from “how do we deliver better content?” to “how do we orchestrate better outcomes?” And the answer will involve agents.
What Defined the Content Era
The content era was shaped by a reasonable assumption: if the organisation provides enough high-quality learning material, employees will develop the capabilities the business needs.
That assumption drove every major L&D investment decision for years. Budget went toward content production, content licensing, and content platforms. Success was measured in library size, content quality ratings, and consumption metrics. L&D teams were, in essence, running publishing operations.
The model produced real value. Organisations with mature content strategies had better-trained workforces than those without. But the model also had built-in limitations that became more visible as AI entered the picture.
The biggest limitation was personalisation. A content library can be well-organised and well-tagged, but it still relies on either the learner or an administrator to match the right content to the right need at the right moment. In an organisation of five thousand employees across multiple functions and geographies, that matching problem becomes enormous. Most L&D teams dealt with it by segmenting: role-based paths, function-specific curricula, manager-nominated programmes. Reasonable, but blunt.
The second limitation was feedback. Content delivery systems know whether someone consumed the material. They do not know whether it changed their behaviour, improved their performance, or built a capability the business needed. The gap between activity data and outcome data remained wide throughout the content era, and most organisations never fully closed it.
What the Agent Era Looks Like
An AI learning agent operates on a fundamentally different model. Instead of organising and delivering content, it orchestrates outcomes.
The agent knows who the learner is, what their role requires, what their current skills profile looks like, and what the organisation’s priorities are. It uses that information to assemble a development experience that is specific to the individual, drawing from whatever sources are available: formal courses, micro-learning modules, on-the-job tasks, external resources, AI-generated practice exercises, peer connections.
Content still matters in this model. The agent needs quality material to draw from. But content moves from being the centre of the strategy to being one input among several. The centre of the strategy becomes the agent itself and its ability to match the right development to the right person at the right moment.
The shift is already underway. According to Josh Bersin’s 2025 HR Technology report, the fastest-growing segment of enterprise learning technology is AI-powered skills and talent intelligence platforms, not content libraries. Gartner’s 2025 research on corporate learning technologies points in the same direction: the market is moving toward systems that orchestrate outcomes rather than systems that deliver material.
What Changes for L&D Leaders
For L&D leaders, the shift from content to agents changes three things about how the function operates.
The first is where budget goes. In the content era, a large share of L&D spending went toward content production and licensing. In the agent era, a growing share will go toward the intelligence layer: the agent infrastructure, the data integrations, and the skills frameworks that allow agents to make good decisions. Content investment does not disappear, but it rebalances.
The second is how success gets measured. Content-era metrics, completions, consumption hours, library utilisation, lose their primacy. The metrics that matter in the agent era are skills acquired, time-to-competency, capability applied on the job, and the connection between learning activity and business performance. These are harder to track, but they are the metrics that earn continued investment from leadership.
The third is how the team structures its work. Content-era L&D teams were organised around production: instructional designers, content developers, video producers, project managers. Agent-era L&D teams will be organised around orchestration: learning architects, data specialists, skills analysts, and partnerships with IT and HR that are much closer than most L&D functions have today.
None of this happens overnight. The transition will take years, and most organisations will operate in a hybrid model for a long time. But the direction is clear, and the organisations that start planning now will have a meaningful advantage over those that wait.
What Changes for Learners
From the learner’s perspective, the change is welcome. The content era, at its worst, felt like being handed a library card and told to figure it out. Even at its best, it felt like receiving a curated reading list that was designed for your general category rather than your specific needs.
The agent era replaces that experience with something more personal and more useful. Development arrives in context. It adapts as the learner grows. It connects to the work they are doing right now, not to the work someone imagined they might be doing when the course was designed eighteen months ago.
The World Economic Forum’s Future of Jobs Report 2025 estimates that 39% of current skills will be outdated or transformed by 2030. For individual employees, that makes personalised, adaptive development a career necessity, not a nice-to-have. Agents provide it in a way that content libraries alone never could.
Planning the Transition
The shift from content to agents will not happen in a single budget cycle. But 2026 is the year to start planning it deliberately.
That means auditing the current learning stack with an eye toward agent readiness: where is the data, how well is it connected, and what intelligence layer can sit on top of what already exists? It means rethinking how the L&D team is structured and where new capabilities are needed. And it means having honest conversations with leadership about what the next phase of learning investment should look like.
If your organisation is thinking about this shift, we have been building toward it. We would be glad to share what we have learned. Talk to our team
Sources: Josh Bersin. “HR Technology 2025: The Market Reinvents Itself.” https://joshbersin.com/hr-technology-market/ Gartner. “Market Guide for Corporate Learning Technologies, 2025.” https://www.gartner.com/en/documents/5635596 World Economic Forum. “Future of Jobs Report 2025.” https://www.weforum.org/publications/the-future-of-jobs-report-2025/