Look at how someone actually uses your work now. They hit a problem on a Tuesday afternoon, go looking for the one thing that solves it, use that thing, and close the tab. A few weeks later a different situation comes up, and they come back for a different piece. They are not working through your material from start to finish. They pull what is relevant at the moment it becomes relevant, and most of the time they never see the rest.
Think about the person who bought your course a year ago. They watched the first two modules, meant to finish, and never got back to it. Six months later a pricing decision came up, and they returned for the one section that covered it, watched about four minutes, got what they needed, and left. The completion dashboard files that person under failure. They are actually the person your work served best.
This is not a discipline problem in your audience, and it is not a sign your work is weak. People do not use important work in a straight line. They come back to what is relevant when it matters. The behavior changed. Most expert content is still built as though it did not.
How the ground moved under the course model
The course model was a real advance, and it made sense for the problem it was built to solve. In the early years, the hard part was access and delivery. You had expertise stuck inside a room, a keynote, a PDF, a live event, and the job was to get it online and get people enrolled. Courses did that well. For a while, putting the material somewhere people could reach it was most of the battle.
Access got solved. Use never really started. Those are two different problems. Getting the material online turned out to be the easy half. The harder half was whether anyone used it once it was there, and the course format was never built to answer that. It was built to deliver, and it delivered.
The question used to be whether people could get to the material. That question is answered now. Expertise is everywhere, most of it a search away. The open question is whether people use what you made when a real situation calls for it. That is a question about return. The course format was built to answer a different one.
Why intermittent return is closer to how people actually learn
There is a reason the pull-it-when-you-need-it pattern feels normal to your audience. It sits close to how memory actually works. When people take in something once and move on, most of it fades within days. When they come back to the same material across spaced intervals, far more of it stays. Researchers have measured this across hundreds of studies; a large meta-analysis of distributed practice found that spacing learning out over time produces substantially better long-term retention than packing it into one pass (Cepeda, Pashler, Vul, Wixted, and Rohrer, 2006).
So the person who returns to your work in pieces, weeks apart, is not failing at it. They are behaving in a way that fits how memory works better than a single linear run through a course does. The intermittent, need-driven return you see in your own numbers is not a deviation from good learning. In a lot of ways it is the shape good learning takes.
The same pattern shows up at every scale
You can see it inside a single lesson too. A six-minute block that has to be watched in one sitting assumes a straight line through it. The same person who returns to your whole body of work in pieces also returns to one lesson in pieces, for the forty-second answer they actually came back for. The argument for building a lesson as small returnable pieces rather than one long block is the same argument this piece is making, one scale down.
The lesson scale and the relationship scale are the same behavior. Build for one and you tend to build for the other. Put the assumption next to the behavior and the gap is easy to see.
| What the course model assumes | What people actually do |
| Start at the beginning and work through to the end | Open the one piece the moment it is needed |
| Finish it once, and you are done | Use it, close it, and leave |
| It happens in a single sitting | Come back weeks later for something different, when a new situation calls for it |
What actually fits the behavior
So what do you build for an audience that behaves this way? Not a better course. A course is organized around finishing, and finishing is the thing your audience mostly does not do. The more useful move is to give the work a form shaped around how people already use it: access they can return to, arranged so the one piece they need is findable at the moment they need it, sitting somewhere the body of work keeps accumulating instead of scattering across old launches and buried folders.
Two things have to be true for that to work. People have to want to come back, and when they come back they have to find what they need. A body of work that does both is what we mean by a place people can return to. The audience-facing name for the way in is a pass, permission to enter the work and keep coming back to it. It is the model shaped around the behavior your audience already has, and it is the one where the work compounds instead of resetting every time you launch something new.
You do not have to convince your audience to behave differently. They already changed. The work is to build for the behavior they already have.
Frequently asked questions
Why don’t people finish online courses?
Because most people no longer take in expertise in a straight line. They pull the one piece they need the moment a situation calls for it, use it, and leave, then come back later for something different when a new problem shows up. Completion assumes a single linear pass that most audiences stopped making. The low finish rate shows the format no longer matches the behavior. It is not a sign the audience lacks discipline.
How do people actually consume expert content now?
On demand and out of order. They find the specific answer they need when they need it, apply it, and move on, then return weeks later for a different piece when a new problem comes up. Use is intermittent, need-driven, and return-based. Most of the material never gets seen in sequence, and that is normal.
Why is course completion rate so low?
Course completion measures whether someone worked through the material from start to finish, and that is a behavior most audiences no longer have. People treat expert work as something to return to when it is relevant. They do not treat it as something to get through once. A low completion rate usually reflects the mismatch between the format and how people use the work. It says little about the quality of the content itself.
What is replacing the online course?
The move is toward formats organized around return: access people can come back to, arranged so the piece they need is findable at the moment they need it, in a place where the body of work keeps accumulating. The audience-facing version of that access is a pass. It is a model shaped around how people already behave.