Moving From 5% to 85% Completion Rates for Online Courses
MOOCs, shorthand for massive open online courses, have been widely critiqued for their miniscule completion rates. Industry reports and instructional designers alike typically report that only between 5 to 15 percent of students who start free open online courses end up earning a certificate.
This does not necessarily make MOOCs a failure. However, since 2012—when MOOCs gained widespread recognition—instructional designers have made significant strides in designing scalable learning experiences that people successfully finish.
Often these happen in smaller courses, designed with a more intimate learning experience in mind, and sometimes with fees and stricter admissions and registration standards. Today, 2U reports completion rates of up to 88 percent for their online degree programs. Harvard Business School’s online programs claim similar success, with completion rates of 85 percent.
At Acumen, where I design online courses, we’ve also been offering selective cohort-based programs for the past year that achieve completion rates of 85 percent. That’s a far cry from five years ago, when only 5 percent of the students were finishing the MOOCs I was designing.
How have instructional designers collectively moved the needle so dramatically on completion rates? Unsurprisingly, some of the biggest drivers of these improved metrics include making people pay for online programs, increasing the selectivity of courses, and adding program managers and teaching assistants to follow up with learners.
However, there are other meaningful interventions that help more students stick with online educational experiences. Here are seven practices for moving completion rates for online courses from 5 to 85 percent:
1. Make students put skin in the game.
If your goal is having more students finish a learning experience, require students to demonstrate upfront commitment. The most obvious way to do this is to simply charge a fee.
However, another effective strategy can be making students complete an application that asks them to reflect on how they will apply what they learn and why they are prepared for this experience. You can also require that they get a referral from someone else or complete an interview to be selected for a seat.
When you build a sense of selectivity into registration, people generally value the subsequent experience more. At Acumen, we’ve found that payment is the most straightforward proxy for commitment. Learners in our programs who pay their own tuition are almost two times more likely to keep up with assignments in a course, compared to those whose seats are underwritten by a philanthropic funder. We’ve introduced tiered pricing models to make these paid programs equally accessible to students in places ranging from Uganda to the U.K. and have seen that people across the world often value something more when they pay something for it.
2. Avoid on-demand learning and instead impose deadlines.
Rather than making a course constantly available, impose a final deadline and make the learning materials available only for certain intervals during the year. A shockingly low percentage of people actually start an on-demand course if it is constantly available, even if they pay for them. We see that people often purchase on-demand courses on platforms like Udemy or Teachable, but in some courses, only a quarter of those who purchase log in to watch the first video.
In the same way that people idealistically accumulate a stack of New Yorker magazines, but never actually read them, we see that people aspirationally purchase on-demand courses but do not always start them. Instead, you often need to combine payment with a sense of urgency. As the old adage says: What can be done at any time, often gets done at no time.
3. Combine synchronous and asynchronous learning.
If you offer an online course that people can work through at their own pace (asynchronously), it often helps to have a series of “live” or synchronous events that students can collectively tune in to during a course.
In Acumen’s programs, we facilitate weekly “Learning Labs” hosted on Zoom, a video conference platform, where students show up for two hours to discuss case studies and receive feedback from peers. We’ve also hosted “office hours” where students have a chance to interact via video conference with the subject matter expert featured in the course. Finally, we’ve done live sessions using Slack where up to 500 students collectively log in for text-based discussion of assignments in small teams.
Platforms like Facebook Live, Zoom, Slack and Whatsapp that can host synchronous discussions can be combined with course platforms like Coursera or Canvas that offer asynchronous resources. Adding live events helps students feel like they are part of a larger learning community and infuses a sense of energy and urgency into a multi-week learning experience. If you can host periodic live events in person and truly blend the learning experience, your completion rates may be even higher. Face-to-face time, whether in person or via video conference, can help make the course experience feel more personalized and motivating.
4. Use the power of peer pressure.
At Acumen, we’ve been designing MOOCs that can be completed in teams. Even if people just identify one other peer to sign up for a course with, they are far more likely to complete.
Team formation is a tempting feature to automate. Imagine if students could randomly be assigned a team member or accountability partner in the same way that an algorithm randomly pairs you with another passenger in an Uber Pool or Lyft Line.
However, in practice, we’ve found that automated pairing works successfully only for one-time engagements. If you want to have students form teams or pairs that last the entire duration of the course, it works far better to put the onus on students and urge them to find someone they already know (a friend or coworker) to complete the course with. The less the course provider needs to play matchmaker, and the more you can put students in the driver’s seat, the better the results will be.
5. Make students feel seen.
In many MOOCs, students feel like their assignments end up being submitted into a void. They work hard on challenging projects or exercises and upload them to a platform, but then never receive personalized feedback because of the sheer volume of submissions.
To combat this feeling, many selective online learning programs are now hiring a teaching assistant or program manager who can personally send an email to students to let them know that their assignment was received, or reminding them to submit. For larger online learning experiences, you can use a platform like NovoEd that enables segmented emails to be sent to students, customized depending on their progress in the course. It takes time to initially draft all of these segmented communications—and having a real person follow up with students unsurprisingly still drives much better results—but this is increasingly an area where AI could be applied in interesting ways to support online learning.
6. Don’t be afraid to dangle prize money or other incentives at the end of a course.
Online courses generally require a great deal of intrinsic motivation to complete. We’ve found that if you can offer an extrinsic reward—like the promise of seed funding for the best final submissions—students are much more likely to finish a learning experience.
For example, we recently hosted a virtual challenge with Unilever that awarded seed funding and mentorship to the best ideas developed through a 5-week online course. Completion rates tripled from the usual 8 percent average for this course to 24 percent.
The incentives don’t necessarily have to be monetary, but if you can partner with employers or funders to offer students real-world tools and resources to turn their coursework into actual products and services, we’ve found they are more likely to complete.
7. Design assignments that help adult learners complete projects for work.
Most adult learners will value an online learning experience if it helps them accomplish immediate professional goals. If you design open-ended assignments that can be applied to real work projects, people are far more likely to be incentivized to carve out the time to complete them. If you’re optimizing for completion rates, design learning experiences that help people directly accomplish their day-to-day tasks.
Not all online courses necessarily need to be designed to be completed. Some courses should remain free and open for sampling so that learners can pick and choose the parts that are directly relevant to them, or so they have the freedom to learn just for the sake of learning.
However, as we increasingly build online degree programs that students are investing valuable tuition dollars and time into, we have an obligation to use tools and insights from fields like behavioral science, psychology, and pedagogy to help more students finish what they start.