In today’s business climate, there is a feverish, almost unique urge to transform into a Skills-Based Organisation (SBO). This movement, often fuelled by the strong influence of suppliers, industry analysts and even “propaganda” or the fear of missing out, promises a future of talent democratisation and organisational agility.
However, many of the multinationals with whom we worked are discovering a harsh reality: a “skills-first” strategy is certainly mandatory, but often not sufficient on its own. A piece of this puzzle is missing and often overlooked: the quantifiable, urgent world of tasks.
The proportional dilemma
The renewed focus on tasks is increasingly present, yet this shift is not without resistance. HR departments have invested heavily in skills-first strategies over the past few years and are understandably hesitant to dilute that direction to flip the system from being skill-oriented to task-oriented, making the task the unifying factor where skills will become the output of the task definition, not the starting point.
However, the current and real risk lies in skills being isolated into an isolated doctrine. When organisations become entangled in overly complex taxonomies and the laborious collection of fragmented skill data, resistance grows, and impact diminishes. Marc Ramos, one of our trusted advisors and author of Task vs Skills, describes this tension as the proportional dilemma. Many existing frameworks fail not because skills are irrelevant, but because they are disconnected from actual work performance and concrete business challenges.
The dilemma is straightforward. Developing strong skills is essential for future readiness. But an overreliance on a skills-only mindset creates a structural gap between what people are taught and what they are expected to deliver. Therefore, closing the circle requires balance: top-down strategic skills aligned with bottom-up operational tasks.
Tasks as the Grassroots Drivers of the Learning Ecosystem
In this strategy, tasks are the “grassroots heroes” of productivity in an AI-driven learning ecosystem. They are grassroots because a task is a discrete unit of work activity with a specific beginning, end and measurable output.
Unlike skills, which are qualitative, fluid and context-dependent, tasks are quantitative and binary; they are either completed correctly or not. This makes it possible to objectively and measurably identify where productivity gains can be made and where capability gaps exist.
- Productivity gains: Access to task-specific AI support has been shown to increase productivity by an average of 14%, and by as much as 34% for new employees.
- Accelerated skill development: Task-based AI support can enable employees with only two months of experience to perform at the same level as employees with six months of experiences
Task as a grassroots initiative is therefore the starting point for the search for continuous improvement by capitalising on the added value of Agents in learning ecosystems, or cognitive learning ecosystems. The starting point here is, of course, to achieve measurable results, and although only 4% of companies report on the business results of their skills development programmes, tasks make it possible to clearly track resolution times, error rates and cycle times.
From zero-shot to a buddy for capability building
One area for improvement that often comes up in our customer conversations and research is zero-shot prompting: asking a single, isolated question to a chatbot or consulting a static FAQ page, even for critical processes such as onboarding.
Today, we are increasingly seeing the benefits of agentic workflows that go beyond this reactive approach. In onboarding, for example, we see advantages as the system evolves from a simple information engine to a proactive ‘buddy’. As a proxy agent for capability building, it no longer provides isolated answers, but performs autonomous actions. By iteratively refining its own output, the agent guides a new employee through complex sequences, from setting up systems to navigating cultural nuances, ensuring fundamental success from day one.
Using this workflow as a “buddy” gives a new employee a reliable teammate that interprets the environment to perform essential tasks, such as scheduling meetings and clarifying workflows. This transforms vast datasets into tailored development paths, turning a simple process into a foundation for long-term success and well-being.
5 strategies for success in the cognitive learning ecosystem
To move beyond the proportional dilemma, several strategic considerations deserve attention. For a deeper analysis, Marc’s full report provides comprehensive insights.
1. Clarify the taxonomy and framework Organisations must dare to strive for open-source skill taxonomies. This standardisation allows AI systems to process and analyse data from different sources more accurately, leading to better training datasets for AI models and more effective workforce planning.
2. Modernise task analysis with multimodal AI Traditional ‘over-the-shoulder’ monitoring is being replaced by Task Mining and Multimodal AI. Multimodal AI can process video, audio, and images to perform real-time task analyses, such as identifying safety risks on a construction site, and then automatically generate punch lists or tracking systems for completion.
3. Implement Retrieval-Augmented Generation (RAG) To reduce AI errors, organisations are switching to RAG, which combines AI’s internal knowledge with external, task-specific data. This ensures that AI-generated instructions are based on the company’s specific, current procedures rather than on general, potentially outdated information.
4. Focus on “routinisation” for automation Routine tasks – tasks in which a fixed procedure is methodically repeated – are the prime candidates for automation. By performing a task audit, companies can identify these everyday activities and remove the “grunt work” from human workflows.
5. Strategically pair people and AI The goal is not to replace people, but to create augmented workers. Research shows that “human-in-the-loop” approaches, where AI handles the processing power and humans provide domain expertise and final control, reduce review cycles by 20-60%.
Human first by squaring the circle
The productivity gains of task-oriented AI are undeniable, yet the human impact must be addressed with equal seriousness. In liminal times, uncertainty fuels fear, especially when no one knows what tomorrow will bring and innovation appears to move faster than our ability to adapt. Many employees are already exhausted, while others worry that time saved through AI will simply be replaced by more work at a faster pace, or ultimately lead to the loss of human roles.
Across many organisations, this moment is increasingly used to redefine productivity by shifting the focus from quantity to quality. By squaring the circle, balancing the qualitative depth of skills with the quantitative precision of tasks, organisations can ensure that the time gained through AI is reinvested in innovation, creativity and employee wellbeing.
Services and experiences for lasting Impact
Amplify your employees’, teams’, and organisation’s Learning Mindset in liminal times through a comprehensive portfolio of services and experiences. From keynotes and events to in-depth development journeys, leadership coaching, executive advisory, and technology support, every initiative is powered by data-driven insights from the Learning Mindset Amplifier and translated into meaningful experiences that create lasting impact. Each programme is tailor-made and delivered with the support of a global network of experts
Leave a Reply