A practical reference for learning, organizational design, and changed work, curated from more than 20 years and 450+ SDI client engagements.
63 essential terms across 8 categories
Find where the term fits
Filter by the part of the business you are trying to understand or improve.
The map shows how each term relates to a practical question about the work, roles, skills, or support that may need to change.
One connected question
Four connected dimensions
One pressure can expose more than one thing that needs to change.Choose a dimension to focus the collection. Choose it again to return to the full view.
Showing 63 of 63 terms
The practice of designing digital learning content that can be used by all learners, including those with visual, auditory, motor, or cognitive disabilities. Accessible eLearning incorporates screen reader compatibility, captions, keyboard navigation, and other accommodations.
A streamlined instructional design process developed by Cathy Moore that focuses on measurable business goals and the specific actions employees must take to achieve them. Action Mapping challenges traditional content-heavy approaches by designing only the practice and information learners truly need.
A systematic instructional design framework consisting of five phases: Analysis, Design, Development, Implementation, and Evaluation. ADDIE provides a structured approach to creating effective learning experiences by ensuring thorough planning, iterative development, and continuous improvement.
Coordinating multiple AI agents, and the humans around them, so work flows through them as one managed process instead of a pile of disconnected pilots. Orchestration covers routing tasks to the right agent, sequencing handoffs, escalating exceptions to people, and monitoring the chain end to end. It is the operational layer that turns individual agents into an agentic workforce.
Measuring whether deployed AI agents return more than they cost - in hours returned to the team, error rates, cycle time, and dollars - against a baseline captured before deployment. The discipline matters because pilot enthusiasm is not P&L impact: MIT NANDA reported in 2025 that roughly 95% of custom enterprise GenAI pilots showed no measurable P&L impact within about six months. Set the baseline first, instrument the workflow, and report against it on a fixed cadence.
Artificial intelligence systems that can autonomously plan, reason, and execute multi-step tasks with minimal human intervention. Unlike traditional AI that responds to single prompts, agentic AI operates with goal-directed behavior, making decisions, using tools, and adapting its approach based on results. In workforce contexts, agentic AI transforms how organizations structure roles by automating complex workflows that previously required human judgment, necessitating organizational redesign and new skill development strategies.
Generative AI produces content on request, one prompt at a time, with a human driving each step. Agentic AI pursues goals: it plans multi-step work, uses tools and systems, checks its own output, and keeps going without a person in the loop at every step. The practical difference for an organization is that generative AI changes how people do tasks, while agentic AI changes which tasks need a person at all. The first is a productivity question; the second is an organizational design question.
A structured evaluation that maps where AI agents can take on real work inside an organization, task by task and role by role, and what that freed capacity is worth in dollars. It differs from an AI readiness assessment: readiness scores your technology and culture, while an agentic workforce assessment starts from the work itself and ends with a prioritized, board-ready roadmap. SDI Clarity runs this as a fixed five-week engagement for organizations of 500-2,500 employees.
An organization's measured ability to put AI agents to work safely: processes documented at a level agents can follow, tasks segmented, governance boundaries set, and people prepared to manage agents rather than compete with them. Readiness is not the same as adoption - many organizations use AI tools daily but cannot yet hand complete tasks to agents. In a 2026 IBM Institute for Business Value survey of 2,000 technology executives, only 11% said they felt fully ready for the scale of AI agent deployment expected within the next year.
An iterative approach to instructional design that applies agile software development principles to learning content creation. This methodology prioritizes flexibility, collaboration, rapid iteration, and continuous stakeholder feedback to produce relevant, effective training materials quickly.
The decision boundaries, permissions, monitoring, and audit trails an organization puts around AI agents that act on its systems and data. Mature governance treats every agent as a managed entity: it has an identity, an owner, defined limits, and a reviewable record of what it did and why. In Deloitte's 2026 State of AI in the Enterprise survey, only 21% of respondents said their organizations have a mature governance model for agentic AI, which means most are deploying agents faster than they can supervise them.
Putting a defensible dollar figure on what AI agents could be worth to a specific organization before anything gets built. Honest sizing works bottom-up - hours per task, fully loaded labor cost, a realistic adoption rate - not top-down from market-survey percentages that describe other companies. The output is a ranked list of opportunities with payback estimates, which is what a board actually needs in order to approve spend.
A systematic evaluation process that analyzes an organization's roles, processes, and workflows to identify opportunities for agentic AI augmentation and automation. AI workforce assessments go beyond technology readiness to examine organizational design, task segmentation, skill requirements, and change management needs, providing a strategic roadmap for responsible AI integration that enhances rather than displaces human capability.
Software applications used to create eLearning content, including interactive courses, assessments, and multimedia presentations. Popular authoring tools like Articulate Storyline, Adobe Captivate, and Lectora enable instructional designers to develop engaging, SCORM-compliant content without programming knowledge.
An instructional design approach that begins with defining desired learning outcomes and assessments before developing content and activities. Also known as Understanding by Design (UbD), this methodology ensures alignment between objectives, assessments, and instruction for more effective learning experiences.
An instructional methodology that combines traditional face-to-face instruction with online digital learning activities. Blended learning leverages the strengths of both modalities to create a comprehensive learning experience that maximizes engagement, flexibility, and knowledge retention.
An approach to organizational design that starts with the capacity question - how many hours of existing work could AI agents absorb - rather than the headcount question. Capacity-first design treats agents as a way to give current teams more throughput on higher-value work, not as a layoff instrument. The sequence is fixed: measure capacity task by task, redeploy it against strategy, and only then revisit structure.
A structured approach to transitioning individuals, teams, and organizations from a current state to a desired future state. Effective change management addresses the human side of organizational change, ensuring employee adoption, minimizing resistance, and sustaining new behaviors.
A systematic evaluation process that measures an individual's knowledge, skills, and abilities against defined competency standards. Competency assessments identify skill gaps, inform training needs, and track development progress to ensure workforce capability alignment with organizational requirements.
Mandatory training programs that educate employees on laws, regulations, and company policies applicable to their job roles. Compliance training covers topics such as workplace safety, harassment prevention, data privacy, and industry-specific regulations to mitigate organizational risk.
Digital learning delivered through electronic devices such as computers, tablets, or smartphones. eLearning enables self-paced, scalable training that can be accessed anytime, anywhere, and typically includes interactive elements, assessments, and progress tracking through a Learning Management System.
The structured process of integrating new hires into an organization, encompassing orientation, training, and socialization activities. Effective onboarding accelerates time-to-productivity, improves retention rates, and establishes the foundation for long-term employee success and engagement.
The application of game design elements and mechanics to non-game learning contexts to increase engagement and motivation. Gamification incorporates points, badges, leaderboards, challenges, and rewards to make training more compelling and drive desired learning behaviors.
The share of a role's work performed by AI agents versus the human in the role, measured at the task level. A 70/30 role keeps the human on judgment, relationships, and exceptions while agents carry routine execution. Tracking this ratio across roles, instead of asking whether AI can replace a job, shows leaders where capacity actually sits today and how each role should be redesigned as agent capability improves.
Traditional classroom-based training where an instructor delivers content to learners in a physical setting. ILT enables real-time interaction, immediate feedback, hands-on activities, and social learning through group discussions and collaborative exercises.
The systematic process of designing, developing, and delivering instructional materials and experiences. Instructional designers apply learning theory, pedagogical principles, and multimedia design to create effective training solutions that facilitate knowledge acquisition and skill development.
The deliberate work of capturing how an experienced operator decides - what they weigh, what they refuse, when they escalate - and encoding it into the systems, documentation, and instructions that AI agents run on. As agents absorb execution, judgment becomes the scarce input, and organizations that transfer it explicitly can delegate more work without losing quality. Judgment transfer is the difference between delegating work to agents and abdicating it.
Training delivered at the precise moment of need, providing learners with relevant information exactly when they require it to complete a task. This approach maximizes knowledge application and retention by connecting learning directly to real-world performance situations.
A widely-used training evaluation model developed by Donald Kirkpatrick that measures training effectiveness across four levels: Reaction (learner satisfaction), Learning (knowledge acquisition), Behavior (on-the-job application), and Results (business impact). This framework provides a comprehensive approach to demonstrating training ROI.
The systematic process of capturing, organizing, sharing, and applying organizational knowledge to enhance performance and innovation. Knowledge management ensures critical expertise is preserved, accessible, and leveraged across the organization to support informed decision-making.
The organizational function responsible for developing employee knowledge, skills, and capabilities to improve individual and organizational performance. L&D encompasses training design, delivery, technology management, and strategic alignment of learning initiatives with business goals.
A software system for creating, storing, managing, and reusing learning content at a granular level. An LCMS differs from an LMS by focusing on content authoring and management rather than learner administration, enabling organizations to efficiently develop and maintain training materials.
The collection, analysis, and reporting of data about learners and their learning contexts to understand and optimize learning outcomes. Learning analytics leverages data from LMS platforms, xAPI statements, and other sources to provide actionable insights for improving training effectiveness.
A software platform for administering, documenting, tracking, and delivering educational courses and training programs. An LMS centralizes learning content, automates administrative tasks, tracks learner progress, and provides reporting capabilities for organizational training management.
A data storage system that receives and stores learning records from xAPI-enabled content and applications. An LRS serves as a repository for all learning experience data, enabling comprehensive analytics and reporting across multiple learning platforms and activities.
A learner-centric platform that uses AI and social learning to personalize the learning experience. Unlike traditional LMS platforms, LXPs aggregate content from multiple sources, recommend personalized learning paths, and emphasize self-directed, continuous learning over compliance-based training.
A training approach that delivers content in small, focused segments typically lasting 3-7 minutes. Microlearning targets specific learning objectives and is designed for just-in-time performance support, higher retention rates, and compatibility with modern attention spans and mobile devices.
Learning delivered through mobile devices such as smartphones and tablets, enabling access to training content anytime, anywhere. Mobile learning supports flexible, on-demand access to microlearning modules, job aids, and performance support resources for today's distributed workforce.
An AI strategy built so the organization can swap the underlying model - GPT, Claude, Gemini, or open-weight alternatives - without rebuilding its workflows. Model-agnostic design keeps the durable assets in the organization's hands: process maps, task segmentation, governance rules, and evaluation data. The model is treated as a replaceable component, which protects against vendor lock-in and against betting the roadmap on whichever AI lab is ahead this quarter.
Animated visual content that combines graphic design, animation, and audio to explain concepts, processes, or data. In training contexts, motion graphics create engaging explainer videos that simplify complex information and capture learner attention through dynamic visual storytelling.
A hands-on training method where employees learn job-specific skills while performing actual work tasks under supervision. OJT provides practical experience in real work environments, enabling immediate application of skills and contextual learning that formal training cannot replicate.
Proof drawn from running the actual work: real deployments, real baselines, real numbers, including the failures. It is the opposite of surveys of other people's work or vendor benchmarks. Operator evidence answers "what happened when you did it?" rather than "what did survey respondents say?" In an AI market full of secondhand claims, it is the standard buyers should hold any vendor to: show the workflow, the before-and-after measurement, and what broke along the way.
The deliberate process of configuring an organization's structure, processes, roles, and systems to align with its strategic objectives. Organizational design determines how work flows through an enterprise, how decisions are made, and how teams collaborate. In the age of AI, organizational design has become critical for identifying which roles and processes can be augmented by agentic AI while preserving human-centered value creation.
Performance support is guidance people can use while doing a task, such as a checklist, decision guide, or short walkthrough. It helps them find the right step without stopping to search a long manual. It complements training; it does not replace practice or judgment when the work requires them.
Documenting how work actually flows - steps, inputs, decision points, exceptions - at the level of detail an AI agent needs to execute it. Conventional process maps were written for humans, who fill gaps with context and common sense; agents do not. Mapping for AI forces the undocumented judgment calls inside a process into the open, which is where most automation attempts fail and where most of the redesign value hides.
The financial return on investment from training programs, calculated by comparing the monetary benefits of improved performance against the total cost of training development and delivery. Measuring training ROI demonstrates the business value of learning initiatives and justifies L&D investments.
An agile instructional design methodology that emphasizes rapid prototyping and iterative development cycles. SAM enables faster course development through collaborative design sessions, early stakeholder feedback, and continuous refinement rather than linear sequential phases.
An instructional strategy that immerses learners in realistic situations requiring decision-making and problem-solving. Scenario-based learning develops critical thinking skills by presenting authentic challenges with consequences, enabling safe practice of complex skills in context.
A set of technical standards for eLearning software that enables content to be packaged and tracked across different Learning Management Systems. SCORM ensures interoperability between authoring tools and LMS platforms, tracking learner progress, completion status, and assessment scores.
U.S. federal law requiring electronic and information technology developed or procured by federal agencies to be accessible to people with disabilities. Section 508 compliance ensures equal access to training materials for all employees, aligning with WCAG 2.0 Level AA standards.
An experiential learning method that replicates real-world environments, systems, or scenarios for practice and skill development. Simulations provide safe environments to practice high-stakes skills, make mistakes, and receive feedback without real-world consequences.
A skills gap analysis compares what people can do today with what their roles require. It helps leaders decide where practice, coaching, hiring, clearer expectations, or better tools are needed. The aim is to improve performance in specific work, not simply create a list of courses.
A strategic process for identifying and developing internal talent to fill key leadership and critical roles within an organization. Succession planning ensures business continuity, reduces talent risk, and creates career pathways that enhance employee retention and engagement.
A strategic approach to developing employee capabilities through learning, career development, and performance management initiatives. Talent development aligns individual growth with organizational objectives, building workforce competencies that drive business success and employee engagement.
Breaking a role into its component tasks and sorting each one by what it actually requires: routine execution, pattern recognition, or human judgment. Task segmentation matters because AI agents take over tasks, not jobs. Very few roles are fully automatable, but most contain tasks that are. It is the analytical core of an agentic workforce assessment: segment the work first, then decide what to hand to agents and what to redesign around people.
A systematic process for identifying the gap between current and required performance levels and determining whether training is the appropriate solution. TNA examines organizational, task, and individual factors to define learning objectives and design targeted interventions.
The sequenced plan that turns an AI assessment into shipped change: which opportunities to pursue in what order, who owns each one, what gets measured, and when the plan gets revisited. A useful roadmap is short, dated, and strict about sequence - quick capacity wins first to build credibility, structural redesign after the evidence is in. If it does not fit on a page, it is a deck, not a roadmap.
An educational framework that proactively designs learning experiences to accommodate diverse learner needs from the outset. UDL provides multiple means of engagement, representation, and action/expression to reduce barriers and optimize learning for all individuals.
The use of video content to deliver training, including talking-head presentations, demonstrations, animations, and interactive video. Video-based learning leverages visual storytelling to explain complex concepts, demonstrate procedures, and create emotional connections with learners.
Live, synchronous training sessions conducted remotely through video conferencing platforms. VILT combines the interactive benefits of instructor-led training with the accessibility of digital learning, enabling real-time engagement across geographical boundaries.
International accessibility standards published by the W3C that define how to make web content accessible to people with disabilities. WCAG guidelines are organized around four principles: Perceivable, Operable, Understandable, and Robust (POUR), with conformance levels A, AA, and AAA.
The strategic process of reshaping an organization's human capital strategy, roles, and capabilities in response to technological advancement, market shifts, or business evolution. Workforce transformation encompasses role redesign, reskilling programs, organizational restructuring, and culture change to ensure employees and teams can thrive in new operating models, particularly those driven by AI and automation.
A modern eLearning specification that captures learning experiences from any source, including mobile apps, simulations, games, and real-world activities. xAPI overcomes SCORM limitations by tracking learning data beyond the LMS, enabling comprehensive learning analytics and experience tracking.
This glossary is curated by SDI Clarity, drawing on over two decades of experience redesigning work, developing leaders and teams, and making complex change usable for Fortune 1000 companies and mid-market organizations. Each definition reflects widely used industry concepts and practical knowledge gained through 450+ client engagements across 14+ industries.
Reviewed and expanded: August 2026
SDI connects work design, clear roles and decisions, the skills people need, and the support that helps change last. Shared language should lead to useful action.