Content stops getting used after launch because it's designed as a one-time event, not a reference someone can return to. The fix isn't more training, it's structural: place content at the right stage of the learner journey, build it to be searchable rather than sat-through, and track whether people come back to it, not just whether they finished it once.
Most learning content gets built for launch day. A course goes out, a guide gets published, a training gets scheduled — and for a week or two, people show up. Then it goes quiet, not because the content was wrong, but because nothing about it was designed to survive contact with real work.
The problem is rarely the training. It's what was supposed to happen after it.
This edition traces why good content stops getting used once the launch is over, and what changes when teams design for the return visit instead of the first one.
What you'll learn:
- Why completion rates and attendance are the wrong signal for whether content is actually working
- Where in the learner's journey a piece of content belongs, and what happens when it lands in the wrong place
- What the return-visit signal looks like when the design is actually working
About Field Notes
Each week, we speak with industry experts and iorad users around the world to understand how they're solving challenges in Enablement and L&D. We share those stories through The Adoption Curve.
But across those conversations, patterns have started to emerge.
Field Notes is where we capture them: distilling insights from interviews, customer feedback, and real-world experience into the trends that matter most.
Why Most Training Doesn't Survive Past Launch
Launch day always looks like success. The room is full, the completion rate is high, the survey scores are good.
Then six months pass.
Talk to enough L&D and enablement leaders and a pattern shows up. The content that goes unused after launch is rarely broken; it’s just built for a moment that already passed. Somewhere between the kickoff and the sixth-month mark, the assumption shifts from "people will need this" to "people already got this," and nobody actually checks which one is true.
What happens to a course after the kickoff week ends?
Why does a perfectly good tutorial stop getting opened?
Is the problem the content itself, or where it lives?
Every team that has solved this has run into the same answer, and it comes down to one thing: Design.
Why good content still goes unused
Eileen Macdonald Cooke, a talent and learning leader who has scaled enablement inside large, multi-division organizations, has a name for the mistake even strong teams make.
"People love to learn new things. They don't like to do new things. That's the difference."
Teams build for the moment someone is willing to sit through a lesson. Then they assume the doing takes care of itself. It doesn't. People leave the room, get pulled back into their actual workload, and revert to whatever they were doing before, not out of resistance, but because nothing in their day reminded them otherwise. The information didn't disappear. It just stopped being in front of anyone at the point where they needed it.
Attendance is not adoption.
Cooke has watched this play out across large, multi-division organizations where the instinct is always to run a bigger kickoff, a longer workshop, a more polished launch event. None of it addresses the actual gap, because the gap doesn't live at launch. It lives in the weeks and months afterward, when the content either has a way to resurface or it doesn't. Cooke points to what has to change about how content gets designed in the first place.
"You design for the forgetting, not just the knowing. It's not what I know that matters. It's what I can do that matters."
She likes to make the point with sports: nobody gets great at golf from one lesson. The lesson matters, but only if there's a way to come back to it, practice it, and get corrected when it doesn't stick. Training content works the same way. If forgetting is the default outcome, the content has to be built for someone to find again, not just sit through once. Which raises the next question: found again, where?
Where should content training content actually live?
Sean Adams walks through this in the show's 5D framework, specifically the deployment stage, where documentation and training actually get placed into someone's day.
"Placing the right content at the right stage of the learner journey. We don't wanna throw an in-app guide over if they're never seen this process before... Inverse is also true."
The mismatch runs both directions. Someone brand new to a process gets handed an in-app pop-up assuming context they don't have yet, and they're lost. Someone who already knows the process gets routed into a 30-minute e-learning module they didn't ask for and don't need, and they tune out. Neither failure is about the content being bad. It's about the content showing up at the wrong stage of someone's actual competence.
Wrong content, right intent, still doesn't land.
This is where a lot of otherwise well-built training goes to die. A team spends real time and real budget documenting a workflow correctly, then deploys it as a single format for everyone, regardless of where each person actually sits on the journey from first exposure to full competence. The content was right. The delivery moment was wrong.
Get the placement right, and the same piece of content that failed as a mandatory course can succeed as something people choose to open when they actually need it. The difference isn't the material. It's whether it shows up as a lesson or as a reference, and whether that choice matches where the person actually is.
What does content built to be reused look like?
Heidi Kirby, an L&D consultant who has spent years pushing back on how her field measures success, ran into this directly when a stakeholder asked her to track course completions.
"I don't want them to complete these courses... If people don't need all this information, I want them to come to our videos, get what they need and get the heck out and go do what they need to actually do in the software."
That instinct shows up again in how she scopes what actually gets built. Her rule of thumb: teach the 20 percent of a system that people will use 80 percent of the time, and put the rest somewhere searchable, so people can find it later instead of sitting through it now.
Findable beats memorable.
That's a different design goal than most training starts with. It's not "how do I get someone through this," it's "how do I make sure this is still useful the fifth time they need it." The first goal optimizes for a single pass. The second optimizes for the moment, weeks or months later, when someone has genuinely forgotten and needs the answer fast, not a refresher course.
This is also why the format matters as much as the content. A fifteen-minute video someone has to sit through start to finish is not the same asset as a short, specifically titled clip they can search for and land on directly. Both might cover the same workflow. Only one of them survives being needed at 4pm on a Friday, three months after anyone attended the original training.
How do you know if your content is actually working?
The signal that content is landing isn't completion. It's whether people come back to it, and whether anyone can see where they get stuck. William Illingworth, a learning systems manager who has spent his career moving between higher education and corporate L&D, points to the gap most teams are missing.
"What metrics can we track to identify where the breakdowns are? Do they get to a hard quiz and not persist? Do they get stuck getting into a course and they never even enroll in courses? Do they complete two or three courses but never do the fourth?"
Most teams can ask these questions. Very few can answer them, because the tooling in place only tracks who showed up, not where they got stuck or whether they ever came back. Without that visibility, teams end up reinforcing whatever content they already built, instead of fixing the specific point where people give up.
If you can't see the drop-off, you can't fix it.
This is the same gap Illingworth ran into with something as basic as a course completion flag. People were re-taking material they'd already passed, not because they hadn't learned it, but because they hadn't closed the module correctly and the system never marked them complete. Nobody could see that pattern until someone went looking for it. The lesson generalizes well past that one bug: the breakdown is rarely where you'd guess, and you only find it by actually tracking where people stop, not just whether they eventually finished.
The teams that get this right build in a feedback loop from day one: not just did they finish, but did they return, and if not, where did it break down.
Is this a training problem or a systems problem?
Every fix described so far, the placement, the findability, the usage tracking, points to the same underlying change. Content built to be used more than once has to be managed like a system, not staged like an event.
Cooke names this directly when talking about what separates teams that sustain change from teams that just run good workshops.
"It's the systematic approach versus being an events activities organization."
An event ends. A system doesn't. That distinction sounds abstract until you look at what each one actually produces. An event produces a completion number and a survey score, both of which peak on launch day and mean less every week after. A system produces something that's still being used, corrected, and returned to long after the launch is forgotten.
That's not a call to build more content. It's a call to treat what already exists as something that needs upkeep, ownership, and a way to resurface, the same way any other system in the business would.
Where to start
None of this requires rebuilding a training library from scratch. It starts smaller than that.
- Pick your most-used piece of content. If you don't know which one that is, that's the real starting point, since it means usage isn't being watched at all.
- Put a name and a review date on it. Content without an owner doesn't get worse on purpose. It just quietly stops being anyone's job to keep current.
- Check whether it's built for lookup — short, titled specifically, searchable — or built for a one-time sit-through. A course that takes twenty minutes to reach the one answer someone needs is built for the first, not the second.
- Check whether anyone can currently see where people drop off or stop returning. If the answer is no, that's the next gap to close, before deciding to build anything new.
- Fix placement before you fix content. A good tutorial in the wrong stage of the journey will underperform a mediocre one in the right stage. Confirm someone at the reference and reinforcement stage isn't being routed into a first-exposure asset, and vice versa.
Launch is the easy part. What happens in month six is the real test.