E-learning completion rates: what the research really says

What is the completion rate?

The completion rate is the share of people enrolled in a course who finished it. The figure depends heavily on what goes into the denominator: all enrolled learners, those who actually started, or those who intended to finish. Two rates calculated differently are not comparable.

It also depends on what "finishing" means in the LMS: having opened every page, reached a minimum score, or passed a final assessment.

What do studies on MOOCs say?

MOOCs (massive open online courses) are the only field where large-scale completion data is published.

Study	Scope	Main result
Jordan (2015), IRRODL	221 MOOCs	Completion from 0.7% to 52.1%, median 12.6%
Reich and Ruipérez-Valiente (2019), Science	All MIT and Harvard MOOCs on edX, 2012-2018: 5.63 million learners, 12.67 million enrolments	3.13% completion among participants in 2017-2018 (nearly 6% in 2014-2015); 46% among learners who paid for a certificate

Jordan observes that completion falls when the course is long and rises when assessments are exclusively auto-graded. The first two weeks are decisive: after that, the share of active learners stabilises.

Reich and Ruipérez-Valiente show that completion did not improve between 2013-2014 and 2017-2018 (it actually fell), and that newcomers rarely come back: the share of first-time learners who took another MOOC the following year dropped from 38% in 2012-2013 to 7% in 2016-2017, according to Inside Higher Ed's report.

Do these figures apply to corporate training?

Not directly. A MOOC is free, open to all, and many enrolled learners simply come to browse. Corporate training is assigned, often taken during working time, sometimes mandatory. The context, audience and stakes differ.

To our knowledge, there is no public, methodologically comparable statistic on corporate e-learning completion. Figures that circulate without an identifiable source should be set aside. Good practice is to measure your own rate, separating mandatory and optional training, and setting a stable denominator.

The gap between 3.13% and 46% in the Science study does suggest that initial commitment matters. It is a correlation: those who pay are also those who were already motivated.

Which levers are actually supported by evidence?

1. Shorter formats. Jordan (2015) associates shorter courses with higher completion. Breaking a long module into short units is consistent with this finding.
2. Get the start right. In Jordan's data, the share of active learners stabilises after the first two weeks: onboarding, the first pieces of content and the first reminders are when engagement is decided.
3. Frequent, auto-graded assessments. They are associated with higher completion (Jordan, 2015) and, according to Roediger and Karpicke (2006), testing yourself improves retention at one week compared with rereading.
4. Space out sequences. The synthesis by Cepeda et al. (2006) of 317 experiments shows that spacing improves retention. It is a learning lever more than a completion lever.
5. Link training to the job. According to the meta-analysis by Blume et al. (2010), applying training at work depends in particular on motivation and a supportive work environment.
6. Do not overestimate automated nudges. Kizilcec et al. (2020, PNAS) tested behavioural interventions on around 250,000 learners in 247 courses at Harvard, MIT and Stanford: effects were much smaller than in earlier studies, and self-regulation interventions increased early engagement without improving final completion.

Should you aim for a high completion rate?

A high rate is a usage signal, not proof of learning. Mandatory training can reach near-total completion without the skill being acquired. The useful indicator combines completion with a measure of acquisition and on-the-job application, as explained in Measuring skills acquisition beyond the completion rate and The 7 indicators every L&D department should track. The limits specific to the LMS are detailed in The LMS is no longer enough.