We're making progress on getting the Sakai conference keynote videos online, but while we wait for those to be ready for the kick-off to the conference post series, I'd like to take advantage of the unexpected lull to write a bit about a new LMS entrant that I had an opportunity to learn about recently. Instructure's Canvas product is one of a new generation of LMS's being created by start-ups that seem to be suddenly popping up everywhere. It has some of the Web 2.0 features that you would expect, like easy personal profile integration with external social networking sites and easy video embedding. But unlike, say, NIXTY---another start-up that really emphasizes open education---Instructure's big theme seems to be getting back to basics---in a good way.
First, a little background. The company was founded by two BYU graduate students who were unhappy with how clunky and out-of-date their school's LMS felt. They decided to do something about it. Now, BYU is a good place to be if you want to rethink educational technology. For starters, it is home to such out-of-the-box thinkers as David Wiley and, until recently, Jon Mott. It also has a lot of support for students who aspire to start their own businesses.1 So these two guys mocked up an LMS design and business model for an entrepreneurialism class and, as a result, were put in touch with serial entrepreneur Cory Reid, who became the company's CEO. They then went on a further requirements-gathering tour of schools in their area and used the data they got to design their LMS and their business. The result is Instructure.
If I had to summarize Instructure's strategy in one sentence, it would be "They use the lessons learned by consumer web companies to clear the clutter out of LMS software design and business model." They're not focusing particularly on open education or analytics or any other hot topics in online education, although they are aware of these and do pay some attention to them. Rather, they are looking at core use cases and trying to make them as simple as possible, throwing out some outdated LMS design assumptions in the process. For example, now that we live in a time where we have pretty sophisticated needs for managing information being pushed to us---and sophisticated tools for addressing those needs---the old email alert check boxes that are sprinkled unevenly throughout LMSs seem pretty inadequate. Why not build a global notifications panel in which users can control the different kinds of notifications they can receive both in terms of frequency (e.g., instantly, daily, weekly, etc.) and in terms of notification channel (e.g., email or text message)? This is something that dotLRN had years and years ago but which most LMSs are only kinda sorta coming to in fits and starts, partly because it's one of those things that's hard to bolt on well if it hasn't been designed into the product from the start. Likewise, if you know that users are going to be using popular Web 2.0 tools as part of their coursework, why not build account integration for things like Google Docs and Twitter right into the personal profile page? Some of the simplification is not driven so much by lessons of Web 2.0 as by just plain good UX design work. For example, they've worked hard to simplify the grading workflow and reduce it to as few clicks as possible, so that teachers can power through grading of a couple dozen student assignments as quickly as possible. In fact, Cory spent a lot of time in our conversation talking about reducing the time burden that the LMS places on teachers and on getting the technology out of the way as a core value proposition of Canvas.
The end result feels quite nice:
Interested in the LMS market? Sign up to receive more information about our LMS Market Analysis service, including a free sample newsletter!
Instructure also works hard to simplify the business arrangements with their customers. Individual faculty can use the system---with help desk support---for free. This helps the company build a support base within a school that might consider adopting. Institutions sign a relatively straightforward contract at what Cory claims is a pretty competitive price. The company uses a couple of strategies to make a decent profit at a decent price. First, by building a SaaS-only platform using modern web technologies (presumably including a lot of open source, although we didn't talk about this), the platform development and maintenance costs are significantly lower than those of a previous-generation on-premise enterprise-style app. Second, they claim to be able to keep the customer support costs low, despite having offering help desk support even for non-paying teachers, by focusing hard on ease of use. One Instructure customer claims to have reduced faculty training time from six hours on a previous LMS platform to four short overview videos.2
Those of us who work in educational technology typically spend a lot of time hand-wringing about how inherently bad the LMS might be and how much we need to throw out. What's intriguing about about Instructure and Canvas is that they raise the question of whether some of the things that teachers and students dislike about LMSs are ontological or contingent in nature. We may just need to refresh the product design (including both technical architecture and assumptions about how users will tend to want to work) based on lessons learned in the past decade. After all, Google didn't have to throw out SMTP and POP in order to bring a significant modernization of email in the form of GMail. The real problem with the LMS may simply be that there hasn't been a refresh of basic design and architectural assumptions...well...certainly in the last seven or eight years, and arguably since the product category came into existence. It's no wonder people are dissatisfied.
- Passions for both education and entrepreneurialism seem to be woven into the fabric of the Mormon culture. [↩]
- Incidentally, there has been some debate in the Sakai community about whether Sakai 3 embedded help can be reduced to this level of just a few overview screencasts, at least for most users. Instructure clearly believes that they have done it. [↩]