Your Learners Drop Out Before Finishing the Course: What to Do

An instructor launches her first course with twenty paying learners. A month later she opens the dashboard: eleven have watched the first lesson, six reach the third, two have finished the first module. Nobody asked for a refund and nobody complained. They simply stopped showing up. She wonders what she did wrong. The uncomfortable answer is that she probably did nothing different from almost everyone else.

This article looks at the available data on dropout, separates what is known from what is assumed, and proposes concrete measures.

What the data says, with caution

The most cited study on completion is Katy Jordan's, published in 2014 in the International Review of Research in Open and Distributed Learning. She compiled public data from massive open online courses (MOOCs) and found that the average course enrolled around 43,000 students, 6.5% of whom completed it. She also found that completion was lower the longer the course.

Read it carefully. These are free university courses with mass enrolment and lots of people signing up out of curiosity. A paid course from an instructor with her own method is a different thing: someone who pays has a clearer intent. So do not expect 6.5% or measure yourself against it. What is useful in the study is the trend (longer, fewer finish) and the warning that dropout is not a rare failure but the default behaviour of self-paced learning.

An invented example to think in numbers: if twenty out of a hundred paying learners finish, you have eighty people who paid and did not complete the result you promised. Some were fine with that. Others felt alone, stuck or guilty, and those are the ones who do not recommend the course.

Why they leave: four causes that repeat

There is no single cause, and the following are the ones instructors and platforms describe most often. Treat them as hypotheses to test against your own dashboard, not as measured truths.

1. They get stuck and nobody notices. An unanswered question in lesson 3 becomes dropout in lesson 4.
2. They lose the rhythm. Two weeks without logging in and coming back feels heavy. The course shifts from a habit to a debt.
3. They cannot see progress. Without a clear sign of advancing, there is no reason to continue on a Tuesday night.
4. They already got what they wanted. Some learners only wanted one technique and learned it in module two. That is not a failure; counting it as dropout without asking is.

What it takes to solve this

Three layers, from cheapest to most expensive:

See who is leaving. Before intervening you need to know who has gone days without practising. A list of at-risk learners with a clear criterion (for example, a week without activity) lets you write to ten people instead of staring at a chart of a hundred.

Give a reason to return without manipulating. A lot of gamification works in the short term and tires people: points, badges, leaderboards. A practice streak with a kind rule (a rest day does not break it) and small challenges with a clear celebration can sustain the habit without pressuring anyone.

Resolve doubts in time. This has the most impact and costs the instructor the most time. Helpful here: an honest preview before buying (a learner who knows what to expect drops out less), an assistant that answers the repetitive questions, and a route for what it cannot answer to reach you.

And a fourth of a different kind: a closing with value. A certificate the learner can show, and that someone can verify, turns finishing into something with an outside reward.

How LearnSuite does it

Each piece exists in LearnSuite today, with its limits:

• Challenges and streaks. The streak counts practice days in the learner's own time zone: one day without practice does not break it and two empty days end it, but the record is kept. Default challenges are seeded per course (days in a row, complete a module, first 1:1 session) and the instructor edits them. By design there are no points, badges or public leaderboard. More in learner challenges and streaks.
• What counts as a finished lesson. Watching 90% of the video, passing its module quiz or, if there is no video, marking it by hand.
• At-risk learners. The progress dashboard flags anyone who has gone seven days without practising and shows completion rates. For the learner, "you are ahead of X% of the group" only appears with cohorts of ten or more.
• Quizzes and certificate. Server-side grading, a minimum score configurable per course, a unique number, a public verification page, PDF, an Open Badges 3.0 credential and a button to add it to LinkedIn. It is issued when all modules with a quiz are passed. There is no exam proctoring. Details in verifiable certificates.
• Free lesson. Any lesson can be opened without buying. It is the antidote to dropout caused by wrong expectations: the free preview lesson.
• Tutor. A chat assistant per academy that answers questions from what is published and, if it does not know, hands over to the instructor. See the AI tutor per academy.

What it does not do: it does not guarantee your learners finish, it has no community among learners (LearnSuite is not designed as a community), and automatic emails to people who cool off are not available today, so the follow-up on an at-risk learner is yours. Plans and prices are on the pricing page.

A seven-day protocol

A simple routine for an instructor with twenty learners:

• Day 0 of each week: review the at-risk list. If it has fewer than ten people, write to each with a personal message and a concrete question, not "is everything OK?".
• After two attempts without a reply: stop pushing. Note the reason, if they gave one.
• Every month: count how many finished each module and find where most people fall. That lesson gets redone, not defended.

How each option in the market solves it

Data from the official pages read on 15 October 2026; what is unverified is marked. Overview in the online course platform market.

• Teachable. Auto-graded quizzes; course certificates only on the Growth plan (189 dollars a month, 139 on annual billing). Comparison: LearnSuite vs Teachable.
• Kajabi. Private community on all plans and live cohorts from Growth (249 dollars a month, 199 annual); certificates not verified. Comparison: LearnSuite vs Kajabi.
• Thinkific. Quizzes on all plans and certificates from Start (107 euros a month, 79 annual on the page we read). Comparison: LearnSuite vs Thinkific.
• Skool. Community is its core product, with unlimited members, courses and live calls from 9 dollars a month; certificates not verified. If peer commitment is your lever, it may fit better than LearnSuite. Comparison: LearnSuite vs Skool.
• LearnWorlds. Certificates from Pro Trainer (79 a month on annual billing, currency not verified) and AI assessments on all plans. Comparison: LearnSuite vs LearnWorlds.

LearnSuite can be started with no fee on the revenue share (20% of sales), or with a fee from Starter (€99 a month plus 8%) and Professional (€297 a month on annual billing): several of these options cost less.