Principles For Better Online Learning
Within the past 15 years, multiple online learning platforms emerged. In 2006, one year after YouTube’s invention, Salman Kahn uploaded a video to explain math to his cousin . The video went viral and two years later, he founded the well-known Khan Academy. In 2010, Udemy, one of the first massive open online course (MOOC) providers emerged, and soon thereafter other online course providers such as EdX, Coursera, and Udacity followed. In 2015, Masterclass brought online learning to the masses, and in the past years, cohort-based-course providers such as Godin’s AltMBA or the Maven platform have been on the rise. Working as an EdTech entrepreneur over the past decade, I’ve had a unique perspective for observing numerous online learning platforms and long-term trends that make me believe online education is still in its infancy. While more people gained access to devices, connectivity, and educational materials, the means of learning haven’t yet been transformed.
The Digitization Of Education
Traditional education has mostly been digitized—in the case of Masterclass, for example, with highly polished video and well-known instructors. According to the Masterclass founder, instructors design their classes . But education researchers agree that “masters” might not be the best teachers. Likely, they’re beginners when it comes to Instructional Design and the science of learning. Masterclass and many other online education providers build on the thesis that online low-touch courses are for skill-building. But with lectures remaining the dominant teaching method, many online learning providers face completion rates as low as 5% . Online education platforms will need to evolve further, specifically tackling three aspects that have been neglected so far.
First, learning platforms will need to include learning science, which has evolved a great deal in the past years. Research pointed out that brains don’t work like recording devices. Learning is a three-step process that involves acquisition, encoding, and retrieval. Learning scientists such as Barbara Oakley, Roediger et al. , and Lieberman  have explored evidence-based learning techniques.
We learn best by connecting new material to prior knowledge, understanding its broader implications (elaboration), attempting retrieval over increasing intervals (spacing) and in different contexts (variation), using testing as a learning tool, and offering real-time feedback on students learning. Yet, most online platforms have not yet integrated these evidence-based learning strategies. Future online education platforms need to include technology that builds upon these insights. They might include testing as a tool, or built-in spaced repetition features, thereby improving students’ learning outcomes and completion rates.
“Learning is deeper and more durable when it’s effortful. Learning that’s easy is like writing in sand, here today and gone tomorrow.”
Second, online course providers will acknowledge the importance of pedagogy in teaching. While online platforms such as Udemy have enabled anyone to become a teacher, quality Instructional Design is still scarce. Future online platforms must factor in that being a “master” at a skill doesn’t equate to being an expert teacher. They must break with the underlying assumption that everyone who’s mastered a skill can become an excellent teacher because it is flawed.
Teachers go through years of teacher training, in both didactics and means. A camera and a platform don’t turn skill experts into great instructors. Online platforms will have to implement some form of quality insurance or instructional training and have to be designed around pedagogy principles. Likely, eLearning platforms will include some form of teacher training that equips instructors with the skills needed for effective course design and delivery.
“Good teachers introduce new thoughts, but great teachers introduce new ways of thinking.”
Lastly, next-generation online learning providers will integrate learning analytics to help students achieve their desired outcomes . While learning analytics is still relatively new, data will be used for supported decision-making in teaching and learning. It will potentially transform how we measure learning outcomes by providing in-depth insights into the learning process. Moreover, learning analytics can serve as a tool to increase students’ motivation.
“Learning analytics is ‘the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimising learning and the environments in which it occurs’.”
Online education has an enormous potential that hasn’t yet been fully unlocked. To build high-quality online education, learning platforms will need to integrate evidence-based learning technology, acknowledge the role of Instructional Design, and harness the potential of learning analytics.
 Peter C. Brown, Henry L. Roediger III, and Mark A. McDaniel, Make It Stick: The Science of Successful Learning. Cambridge, MA.: The Belknap Press of Harvard University Press, 2014.
 David A. Lieberman, Human Learning and Memory. Cambridge: Cambridge University Press, 2011.