Why Adults With ADHD Leave Online Courses—and How Some Stay

At 10:47 on a Tuesday night, Anna, a 34-year-old project coordinator in Warsaw, was staring at a progress bar stuck at 12 percent, while three browser tabs hummed with the usual promises of self-reinvention: Python for beginners, a UX certificate, and a note-taking system that was supposed to rescue the rest. She had paid for all of them during moments of genuine conviction, and she had opened each course often enough to feel busy, yet not often enough to move. “I’m not avoiding the material,” she told me over Zoom, rubbing one eye and laughing in the tired way people laugh when they have explained this too many times. “I’m avoiding the moment where I have to figure out where to restart.”

That sentence, more than the usual clichés about discipline, gets close to the practical trouble adults with ADHD describe in online learning: the friction is rarely confined to understanding the lesson itself. It accumulates around re-entry, task initiation, working memory, time-blindness, and the small administrative moves that course designers tend to treat as negligible. In 2024, when moinaki surveyed 312 adult users who identified ADHD-related attention and executive-function difficulties as a major obstacle in study, 94 percent reported abandoning at least one online course in the previous year, and 61 percent said they had dropped three or more. The striking number was not the abandonment rate alone, which mirrors the brutal completion figures seen across MOOCs for more than a decade, but the pattern underneath it: learners were not leaving because the subjects lacked value; they were leaving at the points where a course demanded self-management that the course itself refused to scaffold.

The broader literature points in the same direction, even when it uses different language. Research on online-course completion, from HarvardX and MITx reports in the mid-2010s to later work in learning analytics, has repeatedly found that persistence depends less on declared motivation than on behavioural supports, pacing, feedback timing, and social accountability. Russell Barkley, the clinical psychologist whose work on ADHD has shaped the field for decades, has argued that ADHD is fundamentally tied to deficits in executive functioning and the management of time toward future goals; when an educational product assumes that a learner will effortlessly convert intention into sequence, it is quietly selecting for the people who already can. What most productivity advice misses is that educational design often externalises the very functions that some learners are spending all day trying to assemble from scratch.

The moinaki survey was small and hardly a national census, but it was specific enough to be useful. Respondents ranged from 23 to 52, with a median age of 31; 68 percent were employed full-time, 17 percent were caregivers to children or older relatives, and 42 percent said they usually studied after 9 p.m., when medication had worn off for some and ordinary decision fatigue had set in for almost everyone. Asked for the single most common point of dropout, 37 percent named “after missing several days and not knowing how to resume,” 24 percent chose “when lessons became longer and less structured,” and 18 percent pointed to “when I had to plan my own practice schedule.” Only 9 percent selected “material too difficult” as the main reason. The old story about laziness survives because it is narratively convenient, not because the evidence is especially kind to it.

Where online courses lose people on an ordinary Wednesday

A great many courses are built around an imaginary learner who has a quiet desk, stable energy, and a reliable capacity to convert a syllabus into action. That learner reads the module overview on Sunday, blocks time for Tuesday and Thursday, returns after interruptions without emotional residue, and treats “self-paced” as a gift rather than a trap. For adults managing ADHD, work, family obligations, and the ambient noise of digital life, the phrase “self-paced” often means that pacing has been outsourced to the least dependable system in the room: an overburdened mind trying to remember itself later.

There is a useful distinction here between difficulty and drag. Difficulty belongs to the content; drag belongs to the transitions around it, the tiny but consequential moments when the learner must decide which tab to open, what counts as enough progress for tonight, whether to rewatch a video, where to store notes, and how to recover from a missed week without feeling that the course has become a monument to personal failure. Gloria Mark, whose work at the University of California, Irvine has documented the fractured nature of attention at work, has shown how costly interruptions are even for people without executive-function challenges. Add time-blindness and weak task initiation to that ecology, and the educational problem starts to look less like a matter of grit than of interface, pacing, and memory support.

One respondent, a 29-year-old accountant in Manchester who asked to be called James, described the pattern with painful precision. He would watch two lessons in a burst on Monday, sketch half a practice exercise, then skip Tuesday because a meeting ran late; by Wednesday the skipped day had acquired a moral weight all out of proportion to its size. “The dashboard says continue module three,” he said, “but module three is forty-eight minutes, and I can’t tell whether that means forty-eight actual minutes or forty-eight plus notes plus confusion plus trying to remember what module two was doing.” By Friday he was looking for a new course, because a new course offers what an interrupted one does not: the temporary fantasy of a clean slate.

That fantasy matters more than course marketers admit. Behavioural economists from Daniel Kahneman onward have shown how people are poor judges of future effort when friction is hidden inside a task, and online education is full of hidden friction. A lesson labelled “12 minutes” may require downloading files, setting up software, interpreting vague instructions, and tolerating the discomfort of partial understanding without immediate feedback. For learners who are already spending cognitive energy on regulating attention, the mismatch between advertised effort and lived effort can become a breach of trust, and once trust goes, the course begins to feel adversarial.

The survey captured this in a blunt statistic: 72 percent of respondents said they were more likely to continue when the next action was visible in one sentence on the same screen where the previous lesson ended. Not in a PDF, not in a weekly email, not in a buried syllabus. On paper this looks trivial; in practice it is the difference between momentum and administrative fog. Educational designers like to discuss content architecture, but for this population the architecture of re-entry may be the more decisive craft.

The six percent do not study harder; they study in a different environment

Among surveyed users, 6 percent said they completed more than three-quarters of the online courses they started during the previous year. Their lives were not magically calmer, and their attention did not sound more obedient. What they had, instead, were systems that reduced ambiguity before it could metastasise. Several kept one active course at a time, not because they lacked ambition, but because parallel enrolments multiplied restart costs; several studied in sessions capped at 20 or 25 minutes, with the explicit aim of stopping before depletion rather than after it. Almost all of them used some external cue—a calendar block, a body-double session, a standing appointment with a friend, or a visual progress tracker that lived outside the learning platform itself.

One of them, Marina, a 41-year-old HR specialist in Saint Petersburg now living in Tbilisi, had finished a data-analytics certificate after years of partial starts. She did not describe a breakthrough in willpower. She described, with the calm of someone who has stopped negotiating with fantasy, a set of constraints. The course lived on the first screen of her phone; she studied only at 7:30 a.m. or during a lunch break, never “later tonight”; each lesson had to end with a handwritten card answering three prompts—what this was about, what I would do next, where the file lives. “The card is for future me when she is suspicious and tired,” she said. “If I leave her a mystery, she won’t come back.”

That line could serve as a design principle. Barkley’s work has long emphasised the value of externalising what neurotypical learners may handle internally: time, sequence, reward, reminders, and consequences. Good ADHD-oriented learning design does something similar. It shortens the distance between intention and action, makes the next move concrete, reduces the penalty for interruption, and replaces vague self-monitoring with visible cues. The learners in the six percent were not proving superior character; they were benefiting from environments, some self-built and some platform-supported, that respected the economics of attention.

This is where the phrase “ADHD-friendly curriculum” earns its keep, provided it is treated as a design framework rather than a branding flourish. In moinaki’s survey, respondents were asked which course features would most increase the odds of completion. The top answers formed a coherent pattern: shorter lesson units with honest time estimates based on total task time, restart summaries after breaks, visible next actions, scheduled check-ins, low-friction note templates, and deadlines soft enough to absorb disruption without dissolving into abstraction. A curriculum built on these elements does not dilute intellectual demands; it redistributes cognitive load so that the learner can spend more of it on the subject itself.

The point is easy to miss because traditional educational prestige still clings to a certain romance of autonomy. Serious learners, the myth goes, should be able to organise themselves around the material without hand-holding. Yet the same institutions that prize autonomy also provide timetables, seminar rooms, office hours, cohort pressure, and the social embarrassment of not showing up. Strip those away, move the course into a glowing rectangle at the end of a workday, and then blame the learner for failing to simulate a university around themselves, and one has not defended rigour so much as confused it with neglect.

An ADHD-friendly curriculum begins with re-entry

If I were commissioning an online course for adults with ADHD tomorrow, I would start with the moment after the learner misses two days. That is where many courses quietly become unusable. The platform should greet return not with a pristine dashboard implying uninterrupted progress, but with a restart lane: a two-minute recap, a note saying what can be skipped, one clearly marked next step, and a small estimate of total effort that includes setup and practice. Shame thrives in vagueness; momentum returns when the path becomes legible again.

The second principle concerns granularity. Barbara Tversky’s work on how people map action and space helps explain why abstract instructions create unnecessary cognitive burden; the mind handles sequences better when they are spatially and temporally concrete. A lesson called “Understand regression” may be accurate but unhelpful at 9:40 p.m.; a lesson called “Load the sample file, run one model, and compare two outputs” gives the learner a shape to hold. In the survey, 66 percent of respondents said they were more likely to continue when assignments were broken into units that could plausibly be completed in one sitting without guessing what “one sitting” meant.

The third principle is accountability without theatrical pressure. Public streaks and gamified guilt work for some people and repel others, especially those who already associate learning with chronic underperformance. Better results came, in the survey and in interviews, from modest forms of human contact: weekly check-ins, asynchronous feedback that arrived quickly enough to matter, and cohort structures small enough that names could be remembered. Thomas Joiner’s work on belonging is not about online courses specifically, but the lesson transfers neatly: people persist more readily when their participation feels witnessed in a low-drama way.

Finally, the curriculum should acknowledge energy as a variable rather than a moral verdict. Learners reported far better adherence when courses offered “minimum viable study” options for difficult days—a recap audio, a five-minute practice prompt, a quick quiz that preserved continuity—alongside deeper sessions for stronger days. This may sound indulgent to purists, yet the alternative is familiar and wasteful: one missed evening turns into a lost week, then a quiet withdrawal, then another purchase made in hope. The educational product that survives contact with real life is usually the one that accepts fluctuation before the learner is forced to.

There is, underneath all this, a broader argument about adult education and dignity. Many adults with ADHD arrive at online courses carrying a private archive of abandoned planners, half-read books, impulsive sign-ups, and other artefacts that have been interpreted by schools, employers, or families as evidence of weak character. A well-designed course does not flatter them with empty reassurance, and it does not patronise them by lowering the ceiling. It does something harder and more respectful: it builds a structure in which effort can reliably attach to action, and in which interruption is treated as a design condition rather than a personal scandal.

That is why the 94 percent figure matters, even as a modest platform survey rather than a grand academic study. It tells us that dropout is not an embarrassing edge case; it is the central fact around which serious online education for adults with ADHD should be organised. The six percent who stay are not mysterious unicorns. They are often using external supports, constrained choices, and kinder re-entry paths that courses could have offered in the first place. Moinaki is one of the places trying to build with that reality in view.