Key takeaways
- Instructional design is the process of turning a performance gap into training that closes it. It is not the act of building slides. That's development, which happens after the design is settled.
- The instructional designer's job is to figure out what the learner needs to be able to do, then design experiences that build that capability. Content is a means to that end, not the end itself.
- ADDIE, SAM, Bloom's Taxonomy, Dick and Carey, and Merrill's Principles are all useful frameworks. Good practitioners pick the one that fits the project rather than forcing the project to fit the framework.
- Most failed corporate training is a design problem, not a content problem. Subject matter is rarely the bottleneck. Structure, sequencing, and practice usually are.
- A 95% completion rate tells you almost nothing about whether the training worked. Capability, behavior change, and on-the-job performance are the metrics that matter.
Most corporate training quietly fails. Not because the subject matter experts don’t know their stuff, and not because nobody’s putting in the work. The training gets built, people click through, the completion reports look fine, and then the same mistakes keep happening on the job. That gap between training and performance is the thing instructional design exists to close.
This guide covers what instructional design actually is, what an instructional designer does, the major models you’ll come across, and how to tell whether the training your organization is producing is doing its job.
What is instructional design?
Instructional design is the process of figuring out what a learner needs to be able to do, then designing the experiences that build that capability. It happens before any course gets built, and it’s what decides whether the eventual course actually works.
More fully: instructional design starts with a performance gap, meaning people can’t do something they need to be able to do, and works backward from there. What does success on the job look like? What’s getting in the way of it now? What knowledge, skills, or judgment does someone need to close that gap? And what kind of practice will actually build those things rather than just expose people to them?
A few common misconceptions worth clearing up.
Instructional design isn’t the same as content creation. Writing the words and recording the videos is content development. Building the course in an authoring tool is eLearning development. Instructional design is the work that decides what those things should be in the first place. It’s the architecture. The rest is construction.
And it isn’t “adding interactivity to slides.” Quizzes, drag-and-drop activities, and click-to-reveal interactions are useful when they’re tied to a clear learning goal. They’re useful when they let learners practice a real decision or recall a key concept at the right moment. Where they fall short is when they’re added late, decoratively, to a course that wasn’t designed around what learners need to be able to do.
What does an instructional designer do?
An instructional designer’s day-to-day work falls into four buckets, roughly in this order on most projects.
Diagnose the actual problem
Before designing anything, the instructional designer figures out whether training is even the right answer. Sometimes the gap isn’t a learning gap. It’s a process problem, or a tools problem, or a management problem, or a motivation problem, and no amount of training will fix it. A good instructional designer will tell you that early, before you spend money building something that won’t help.
When training is the right answer, the next job is to be specific about what kind. A new policy that everyone needs to be aware of is a different design problem from a complex skill that takes months to develop. Lumping those together produces training that does neither well.
Extract what the experts know
This is where most of the real work happens, and it’s the work that’s hardest to outsource to a junior writer or an AI tool. Subject matter experts hold knowledge in patterns and instincts that they often can’t articulate cleanly. Asking them “what do new hires need to know?” gets you a list of facts. Asking them “tell me about the last time a new hire surprised you” gets you the stories where the actual judgment lives.
A skilled instructional designer can take a few hours of unstructured SME conversation and pull out the decision points, the common mistakes, and the rules of thumb that experts rely on without realizing it. That’s the raw material that makes training useful.
Design the experience
Then comes the design itself. Learning objectives that describe what the learner will be able to do (not just “understand”). A sequence that builds capability in the right order. Practice opportunities that mirror the real situations people will face. Feedback that actually teaches rather than just scoring. Assessments that measure whether the design worked.
This is the stage where storyboards get built and decisions get made about modality (eLearning, video, ILT, blended), interactivity level, and assessment strategy. The output is a blueprint detailed enough that a developer can build it without having to guess at intent.
Iterate after launch
Good instructional designers don’t disappear after delivery. They look at the data. What learners struggled with, where they dropped off, what changed on the job. First versions are rarely the best versions, and treating training like a finished product instead of an evolving system is one of the quietest reasons it stops working.
The instructional design models you’ll hear about
The major instructional design models are ADDIE, SAM, Bloom’s Taxonomy, Dick and Carey, and Merrill’s First Principles. None of them is “the right one.” They’re frameworks, each with strengths, and most experienced practitioners draw from several depending on the project. Here’s what each one is for.
ADDIE
The most common framework, and the one most people have heard of. ADDIE stands for Analyze, Design, Develop, Implement, Evaluate. It’s a linear, phased approach: figure out the problem, design the solution, build it, deliver it, see how it worked.
ADDIE works well when the requirements are clear and stable, when you have time to do each phase properly, and when the project is large enough that the structure prevents people from skipping steps. It’s the most widely-taught framework in the field, which means most instructional designers can work fluently within it and most stakeholders recognize the phases.
It struggles when requirements shift mid-project, when the design needs early prototyping to figure out what’s even going to work, or when the timeline is too tight for a full sequential pass.
SAM (Successive Approximation Model)
SAM is the iterative cousin of ADDIE. Instead of one long sequential project, you build small prototypes early, test them with real learners, refine, build, refine again. The design emerges through iteration rather than getting locked in upfront.
SAM works well for projects where you don’t fully know what’s going to land with learners until you try something. It’s particularly useful for soft skills, behavior change, and anything involving complex judgment. Those are the kinds of training where the design choices matter most and are hardest to predict in advance.
Bloom’s Taxonomy
Less a process and more a framework for setting learning objectives. Bloom’s organizes thinking skills from basic (remember, understand) to advanced (apply, analyze, evaluate, create). It gives instructional designers a common vocabulary for what level of thinking a piece of training is targeting.
Most useful for keeping objectives honest. “Understand the company’s data privacy policy” is not the same as “be able to identify a privacy risk in a customer interaction.” Bloom’s helps you write the second one instead of the first.
Dick and Carey
A more rigorous, systems-oriented model than ADDIE. It treats every component of an instructional system, including the instructor, the learner, the content, the materials, and the environment, as interconnected. Design choices in one part affect all the others.
It’s used heavily in academic and government contexts where the rigor matters. For most corporate projects, it’s overkill. The underlying principle is still worth carrying around regardless of which model you formally use.
Merrill’s First Principles
A short list of principles that most modern instructional designers come back to. Learning happens when the learner is engaged in solving real-world problems, when existing knowledge is activated, when new knowledge is demonstrated, when the learner has a chance to apply it, and when the new knowledge is integrated into what they already know.
Less of a process, more of a checklist for whether the design is doing its job. If a course doesn’t activate prior knowledge, doesn’t demonstrate the new material in context, and doesn’t give learners a chance to apply it, the principles say it’ll struggle to work, regardless of how polished the production is.
For a closer look at the major models and when each one fits, see our guide to instructional design models.
How is instructional design different from curriculum design?
These two terms get used interchangeably and they shouldn’t be. They describe different work at different scales.
Curriculum design is the architecture of a learning program. What courses exist, what order they go in, how they build on each other, what each one is responsible for. A curriculum designer answers questions like: across the full onboarding journey, what does someone need to know by week one, week four, month three? Where do the courses live in relation to each other? What does the learning path look like?
Instructional design is what happens inside each course. Now that we know the new hire needs a course on customer escalation handling, how do we design that specific course so it actually teaches escalation handling?
Curriculum design is at the program level. Instructional design is at the course or experience level. Both matter, and most projects need both. They’re rarely done by the same person, though, and conflating them is one of the reasons training programs end up disorganized.
We have a fuller treatment of this in instructional design vs curriculum design.
How do you know when instructional design is working?
This is the question that matters most, and the one that completion reports can’t answer.
You measure capability, behavior, and outcomes rather than seat time. A 95% completion rate tells you almost nothing about whether the training worked. It tells you that people made it to the end. Whether they can now do the thing the training was designed to teach is a separate question, and it requires looking at different evidence.
Some signals worth watching:
- Performance on the job changes. New hires reach proficiency faster. Mistakes that the training was supposed to address become rarer. The behaviors the training was designed to encourage start showing up in real work.
- The questions stop coming. People stop asking the questions the training was supposed to answer. Help desk tickets in the relevant area drop. Managers spend less time correcting the same thing over and over.
- Spaced reinforcement holds up. Capability that disappears within a few weeks suggests the design didn’t include enough practice or retrieval. Training that holds up at 30, 60, 90 days post-delivery is doing what it should.
- Learners can demonstrate, not just describe. If someone can explain the policy but can’t apply it in a realistic scenario, that’s a design failure. It’s recoverable, but only if you notice it.
If your training isn’t producing these signals, the problem usually isn’t the topic, the LMS, or the production budget. It’s the design.
When should you bring in an instructional designer?
The short answer: as early as you can, before any building starts. The work an instructional designer does upfront is what determines whether the eventual training succeeds, and trying to retrofit good design onto a finished course is much harder than getting it right the first time.
Bringing one in early matters most when the stakes are high. Compliance training where a missed concept creates legal exposure. Sales enablement where the cost of a fumbled conversation is real. Clinical or safety training where mistakes hurt people. In these situations, the cost of getting it wrong dwarfs the cost of doing it properly, and skimping on design upfront is a false economy.
It also matters when the content is complex or scattered. If the source material lives in multiple places, in multiple people’s heads, and isn’t yet organized into anything coherent, that’s exactly the moment when an instructional designer adds the most value.
And it matters when you’re building something that has to scale. Training that will be delivered to thousands of people, in multiple languages, across multiple sites, over multiple years. That’s a system, not a one-off course, and systems need to be designed.
The other side of the question: when do you not need an instructional designer? Genuinely simple information transfer (a brief policy reminder, a one-page job aid, an email update) often doesn’t justify the investment. The honest test is whether the cost of getting the training wrong is higher than the cost of designing it properly. If it is, design work is worth doing.
If you’re trying to figure out whether you need an in-house designer, an agency, or a freelancer for your situation, our guide to instructional design companies walks through how to evaluate the options.
Working with an instructional designer: what to expect
A few practical notes for organizations that haven’t worked with an instructional designer before.
The first conversations will feel slow. A good instructional designer will spend time on diagnosis before proposing anything. They’ll ask about the actual job, the real performance gap, the people involved. This isn’t the designer being difficult. It’s the work that determines whether the eventual training is any good.
You’ll need to make subject matter experts available. The most common reason training projects underperform is restricted access to the people who actually know the material. An instructional designer can do remarkable things with good SME access. Without it, the whole project is built on guesses. Plan for two or three focused conversations of 60 to 90 minutes per major topic.
Reviews matter more than you might think. Designs improve through review and feedback. Skipping or rushing review cycles, or running them with the wrong people, is one of the most common ways good design quietly turns into mediocre design. It helps to agree on the reviewer list at the start of the project so new voices don’t show up in round three with feedback that contradicts what was approved earlier. Aim for consolidated feedback rather than five reviewers each sending different comments.
And source files and ownership should be settled upfront. When the project ends, you should own everything: the storyboards, the source files, the assessments, the assets. If a vendor wants to keep the source files so you have to come back to them for changes, that’s a yellow flag worth raising before you sign anything.
If you’d like a closer look at how a partnership-based instructional design engagement works in practice, our case studies walk through several real projects across different industries and complexity levels.
A note on Neovation’s approach
Our team is 100% in-house, so the instructional designer who scopes your project is also the one running the SME interviews, building the storyboard, and staying involved through development. We use the Custom Learning Points model rather than fixed-bid contracts or hourly billing, because design projects evolve as you learn more about the problem, and the budgeting model should accommodate that. The metric we care about is whether training changes behavior on the job, which shapes how we run interviews, write objectives, and design assessments.
If your situation looks like a fit, we’d be glad to talk. If not, we’re happy to point you toward what does. Sometimes a freelancer is the right call, sometimes an off-the-shelf course will do, and sometimes the right answer isn’t training at all. Request a quote when you’re ready, or browse our case studies to see how this looks across different industries.
Frequently asked questions
What is instructional design in simple terms?
Instructional design is the process of figuring out what a learner needs to be able to do, then designing the experiences that build that capability. It's the work that happens before any course gets built. It's the architecture that decides whether the eventual training will actually change how people work, or just track completion. Building the course itself is a separate step that follows good design.
What does an instructional designer do day-to-day?
An instructional designer diagnoses the actual performance problem (often determining that training isn't the right fix), extracts knowledge from subject matter experts through structured interviews, designs the learning experience including objectives and practice opportunities, and builds detailed storyboards that developers can build from. After launch, they look at how the training is performing and refine the design based on real outcomes.
Is instructional design the same as eLearning development?
No. Instructional design is the work that decides what the training should be. That includes the objectives, structure, sequence, practice, and assessments. eLearning development is the work of building it in an authoring tool. Most successful projects involve both, often working closely together, but they're different disciplines with different skills. A good instructional designer can hand a development team a storyboard detailed enough that they don't have to guess at intent.
What is ADDIE in instructional design?
ADDIE is an instructional design framework that stands for Analyze, Design, Develop, Implement, and Evaluate. It describes a linear, phased approach: figure out the problem, design the solution, build it, deliver it, and see how it worked. It's the most widely-used framework in the field and a sensible default for most projects, though it's not the only option. Iterative models like SAM are often better when the requirements are likely to evolve.
How long does it take to design a training course?
Design itself (separate from development) typically takes anywhere from one to four weeks for a single course, depending on how complex the content is, how accessible the SMEs are, and how many review cycles the design goes through. Larger curriculum-level work, like designing a full program rather than a single course, can take much longer, especially if it includes program architecture, learner journey mapping, and modality decisions across multiple courses.
When should we hire an external instructional designer instead of building internally?
When the project is high-stakes (compliance, safety, sales, leadership), when content is scattered across multiple SMEs and sources, when the timeline is tight, or when your internal team doesn't have spare capacity. Internal teams are a great fit for stable, repeated training that benefits from institutional knowledge, but most organizations underestimate the bandwidth needed to do good design well. If your team is already stretched, hiring externally for design work is often less expensive in total than absorbing the project internally and producing something less effective.
How do I know if our training is actually working?
Look at capability and behavior change, not completion rates. Are new hires reaching proficiency faster? Are the mistakes the training addressed becoming rarer? Are the questions the training was supposed to answer no longer being asked? Capability that holds up at 30, 60, and 90 days post-training is the signal that the design worked. Completion rates and satisfaction scores are useful but lagging. They don't tell you whether anyone is doing the job differently as a result.
