Your course shows a hundred enrollments and that's worth being proud of. But check how many of those hundred people actually logged in over the past two weeks. Odds are that number stings a little. Learners who formally unsubscribe are easy to spot, your dashboard flags them right away. Learners who simply go quiet, who stop logging in without ever officially dropping out, are much harder to catch. And that quiet group is exactly where you have the most to gain.
Completion rates don't tell the whole story
An enrollment doesn't guarantee anything. Data from the EdTech sector shows that in some markets only 15 to 20 percent of people who sign up for an online course actually finish it (2). Most of the rest don't dramatically quit, they just fade. A large share stays technically "enrolled" long after anyone has stopped logging in. That skews your own numbers too: your completion rate or your active-learner count can look healthier than reality, because the system keeps counting someone as active weeks after they stopped thinking about your course.
The silent countdown: day 7, day 30, day 60
Dropout data across online courses follows a surprisingly consistent curve. By around day 7, a sizeable chunk of new sign-ups, somewhere between 30 and 40 percent, has already gone inactive. By day 30, active engagement on many courses has fallen below 20 percent of the original group. By day 60, a large part of that group is effectively inactive, even if nobody formally cancelled (2). This isn't a law that applies identically to every course, but it's a useful benchmark. Put your own login data from your last course next to this curve and you'll see right away whether your dropout is normal, or whether something in your course structure needs fixing.
You already have the signals, you're just not using them yet
The good news is you don't need to hire a data scientist to take this seriously. Most course platforms, even smaller ones, already track exactly what you need: the date of someone's last login, the percentage of lessons completed, how many times they've attempted a quiz, and how much time has passed since their last activity. Research into predictive learning analytics shows that login patterns in the first week often predict who finishes a course more reliably than their score on an early quiz (1). In other words, don't just watch for people who score badly. Watch for people who quietly pull back.
Three risk levels, three kinds of action
You can roughly split your learners into three groups, borrowing from how larger platforms handle this (2):
- A bit too quiet (no login for 7 to 10 days): this group hasn't really checked out yet. A short, personal message is usually enough. Name the course and the exact spot they left off, and offer something concrete, like a quick recap of where they were.
- Properly quiet (inactive for 14 to 30 days): one message usually isn't enough here. Pair a personal note with a follow-up email that proposes one small, specific step, a recap of the last lesson, or access to a short refresher.
- Almost gone (inactive for 30 to 60 days or more): this group needs a human, not an automated sequence. A phone call or a direct message from you as the trainer, asking a genuine question about what's holding them back, usually beats any automated flow at this stage.
The channel you pick matters
Not every channel performs equally well for this kind of message. Data on EdTech communication shows that messages sent through WhatsApp can reach open rates of 80 to 90 percent, compared with 15 to 25 percent for email (2). That gap is large enough to rethink where your learners are actually paying attention. For an independent trainer, this doesn't mean you need a full marketing stack overnight. It means that before you type another reminder email, it's worth asking: does this person even check their inbox, or do they mostly read WhatsApp and texts these days?
What's better left undone
A few habits tend to backfire. Blanket discounts like "come back, 20 percent off" teach learners that waiting pays off, which quietly undermines your pricing over time. Generic "we miss you" emails with no specific content barely move the needle, they read as spam and get treated as such. And maybe the biggest missed opportunity is never asking why someone stopped. One simple question in your last message, "what's keeping you from continuing?", tells you more than any metric, because it shows you exactly where your course or your support is breaking down.
A simple starting point for this week
You don't need expensive software to start. Pick one running course, export the list of last login dates per learner, and sort them into the three groups above using a basic spreadsheet. Write one message per group that refers to the exact lesson where someone left off. Send it this week and track who logs back in within seven days. One larger organisation that automated this approach saw the share of employees missing a mandatory training deadline drop from 18 to 4.4 percent, just by flagging 82 at-risk learners in week two and offering them a short refresher module (1). You don't need that scale for the same principle to work. A spreadsheet and one well-timed, personal message already go a long way.
Sources
- D2L, "Predictive Learning Analytics: The Ultimate Guide for 2026" - https://www.d2l.com/blog/boost-corporate-learning-predictive-analytics/
- CampaignHQ, "EdTech Student Reactivation Automation: How Indian EdTech Teams Win Back Inactive Learners with WhatsApp and Email (2026)" - https://blog.campaignhq.co/edtech-student-reactivation-automation-how-indian-edtech-teams-win-back-inactive-learners-with-whatsapp-and-email-2026