Imagine sitting down to develop a learning design strategy and already knowing the optimum length, medium, style, and tone, and even the best day of the week and the time of day to launch your content. Contributor Lori Niles-Hofmann explains why data-driven learning design needs to be part of your strategy.
While there is no crystal ball for learning, with data we are getting increasingly closer to accurately predicting how to best engage our learners. Meaning: less intuition, more metrics, and delivering more value.
One data point = one big difference
The good news is that high-end modern learning tools and management systems make tracking data really easy. They even spit out ready-made visual reports to give instant insights about your learning content. That is a lot of power.
Take this true story as an example of how even one small piece of data can make a big difference. As part of a blended learning solution, a project deliverable was an hour-long video. A simple report from the LMS revealed that the average viewing tolerance for previous video content was around four minutes. Clearly, 60 minutes was not going to launch well. The solution was to divide up the content into segments of two to four minutes and launch the series alongside the full-hour version. No surprise that the shorter videos resulted in an overall increased viewing length, by 79%.
This example leveraged only one simple data point: optimal video lengths. Consider the benefits of other metrics, such as what devices are most used, who by and where, and granular-content-level data, such as how learners answer key questions. The latter could highlight splits in opinions between leaders and team members, for example. Likewise, the former could rapidly streamline design modality decisions.
Data-driven learning design is hugely empowering. So what’s stopping everyone from doing it?
Stop looking at the rearview mirror
It is fair to say that L&D is a bit late to the Big Data party. Sure, we started out strong, waving our glow sticks in the air as we tracked completions and Kirkpatrick evaluations. Yet somewhere along the way we failed to recognize that data was more than just bums in seats and Likert scales. Yes, these data points are important, but by the time they are collected, the learning is already over and the budget spent. Those metrics are like driving down a highway looking in the rearview mirror.
With the right tech, data is now available before you even start your journey. To bring this party back to epic Coachella status, we need to place data analysis at the start of a project and continually mine it for insights along the way.
Does data give us an edge over face-to-face?
I do not like to admit this often, but I have been in the industry long enough that I can remember when elearning was first introduced as an alternative to face-to-face delivery.
**relaxes into rocking chair, and puffs slowly on a pipe**
At the time, the argument against online content was that the connection between the learner and facilitator was lost. In a classroom, a facilitator could see each and every learner and know if the person was paying attention. How would we know if learners were engaged with the online content if we could not see them?
The truth is, as some savvy learning management systems have evolved, the data we have about learner behaviors is more than we could ever see in the classroom. Online, the learner is navigating about as an unobserved individual. It would be rude to simply get up and walk out of a classroom (although it has been known to happen). A learner, however, thinks nothing of instantly shutting down a video or jumping past content when it is of no interest. The result is a very finicky and discerning audience to engage.
The truth can hurt
Yes, it can be ego-bruising. Sometimes we really, really, love that drag and drop exercise that 53% of learners skipped over, but the learner did not see the value. When you go beyond the completions, this type of data analysis on how learners truly interact with each facet of digital content can yield valuable insights. And if you can get past the wounded pride, there are opportunities to refine design to increase engagement. Or you might have a design that totally rocks, and you can learn what worked for your next project!
Which train to take?
Consider the following analogy: a transit planning department is designing the timetables for the subways, trains, and buses in a major city. They start off with some basic assumptions, such as ridership is most likely higher during the start and end of workdays. There are now two directions that the planning can take. The planners can use their expertise to design a timetable, or they can combine their knowledge with an analysis of data from transit systems that are similar in size and population. The chances of success are much more likely with the latter. And even more so if the planners continue to monitor usage and refine schedules and timetables based on insights. I know which commute I would prefer.
As we use data to better understand our learning audiences, we start to move the design process from what we think will work, based on individual expertise, toward what we can predict will work, based on metrics. This ultimately evolves learning design from an art to science, with better outcomes to meet all-important business goals.
About the author
Lori Niles-Hofmann is a senior learning strategy consultant with nearly 20 years of experience in L&D. She believes in the power of data to revolutionize the way we design digital content, which often makes her a nerd at parties. Lori is the author of “Data-Driven Learning Design,” a free e-book available from her blog: https://www.loriniles.com/ebook
Latest posts by Lori Niles-Hofmann (see all)
- Data: Evolve Your Learning Strategy From Art To Science - October 26, 2016