Author Phil Hill Archives - Page 2 of 5

US Higher Education Enrollment Trends by Distance Education Type, 2012-2018

Earlier this week I shared a profile of US online education enrollment based on the new IPEDS Fall 2018 data set. To put the numbers in context, we should look at the trends since IPEDS started tracking online enrollment in Fall 2012.

As a reminder on the data usage:

There are multiple ways to filter and select data. For this set (as with previous analyses for consistency’s sake), I have limited to U.S. degree-granting institutions in six sectors – public 4-year, private 4-year, for profit 4-year, public 2-year, private 2-year, and for profit 2-year. For undergraduate totals I have included degree-seeking and non-degree-seeking students (degree-granting institutions can offer non-degree programs). This will give different totals than other reporting methods. In particular, note that the IPEDS data view summary includes less than 2-year degrees and also includes non degree-granting institutions, leading to slightly higher numbers than shown below.
For the most part distance education (DE) and online education terms are interchangeable, but they are not equivalent as DE can include courses delivered by a medium other than the Internet (e.g. correspondence course). In this post I use both terms.
Exclusively DE is for students taking all courses online; Some DE is for students taking some courses online but not all; At Least One DE, or ALO DE is a combination of exclusive and some DE.

In this view showing total enrollments by DE type, we can see the persistent decline in total enrollment for US higher education, dropping from 20.7 to 19.7 million from 2012 to 2018. Beyond that, we have consistent trends where the number of students taking no online courses has dropped even further, from 15.3 to 12.7 million, while the number of students taking exclusively online courses has risen from 2.6 to 3.3 million, and students taking some but not all online courses has risen from 2.7 to 3.7 million.

Since the total enrollments are changing, it is worth looking at how the various sectors contribute to these changes by viewing the enrollment differences from 2012 data.

The enrollment changes over the past six years is highly correlated with sector. Public 4-year institutions have increased enrollment by 861 thousand while public 2-year institutions have decreased by 1.2 million – even more than the for-profit sectors.

In this view we look at the percentage of total enrollment per year, and we combine the exclusive DE and some DE categories into students taking at least one online course. The percentage of students taking online courses has risen from 26.1% to 35.3% since 2012.

I noted in the Fall 2018 profile post that the rate of increase for online enrollments has slowed down for the some DE category.

In this view you can see that exclusive DE category is rising more than some DE category over the past year (after the reverse being true from 2014 – 2017), but note that this view combines grad and undergrad degree types. If we separate into each component, there are significant differences.

For undergrads, it is far more common to have students in a mixed mode – taking some online courses and some face-to-face courses – and a small amount of decrease in the growth rate.

Now we see the real source of slowing growth for the some DE category – for grad school students. This is a radical change starting two years ago, with the number of grad students taking some but not all online courses suddenly hitting a plateau. In fact, the numbers are 59,099 for 2016, 57,546 for 2017, and 57,544 for 2018. Is this some artifact of data collection? I find it remarkable that the numbers across the US have only changed by 2 students in the past year and only 1,555 in the past two years. If someone can explain this sudden change and consistency in numbers over the past two years, please let me know in the comments.

Update 12/12: Clarified language about exclusive DE rising faster than some DE over the past year.

Fall 2018 IPEDS Data: New Profile of US Higher Ed Online Education

The National Center for Educational Statistics (NCES) and its Integrated Postsecondary Education Data System (IPEDS) provide the most official data on colleges and universities in the United States. I have been analyzing and sharing the data since the inaugural Fall 2012 dataset, and the Fall 2018 data were just released. Let’s view the numbers for US-based online education enrollment.

Update 12/11: See this post for data trends 2012 – 2018

Data Notes

Please note the following:

There are multiple ways to filter and select data. For this set (as with previous analyses for consistency’s sake), I have limited to U.S. degree-granting institutions in six sectors – public 4-year, private 4-year, for profit 4-year, public 2-year, private 2-year, and for profit 2-year. For undergraduate totals I have included degree-seeking and non-degree-seeking students (degree-granting institutions can offer non-degree programs). This will give different totals than other reporting methods. In particular, note that the IPEDS data view summary includes less than 2-year degrees and also includes non degree-granting institutions, leading to slightly higher numbers than shown below.
For the most part distance education (DE) and online education terms are interchangeable, but they are not equivalent as DE can include courses delivered by a medium other than the Internet (e.g. correspondence course).
Exclusively DE is for students taking all courses online; Some DE is for students taking some courses online but not all; At Least One DE, or ALO DE is a combination of exclusive and some DE.
There are two tabs in the interactive graphic – the first shows totals for the US by sector and by level (grad, undergrad); the second shows a map view allowing filtering by sector.

Initial Observations

The headline appears to me that we are seeing the same trends – growth of online, decline of fully face-to-face (no online) enrollment – but that the trends might be slowing down for the Some DE category. More than 35% of all US postsecondary students take at least one online class as of Fall 2018. From 2016 – 2017, however, the percentage of all students taking at least one online course rose from 31.4% to 33.7% (2.3%), but from 2017 – 2018 it rose from 33.7% to 35.4% (1.7%). I plan to do a separate analysis of DE trends over time, but the initial data show a slowdown in growth for online education enrollment, at least for the Some DE category.

For Undergraduates, in absolute numbers exclusively online enrollment increased by 93k, some online enrollment increased 126k, and no online enrollment decreased by 362k since Fall 2017.
For Graduates, exclusively online enrollment increased 65k, some online enrollment remained unchanged, and no online enrollment decreased 31k since Fall 2017.

Table View

Below is a profile of online education in the US for degree-granting colleges and university, broken out by sector and for each state for the most recent, Fall 2018, data.

Map View

Below is the map view of state data colored by number of, and percentage of, students taking at least one online class for each sector. In the interactive view, if you hover over any state you can get the basic data.

Interactive View

For those of you who have made it this far, you can find the interactive graphic here. Enjoy the data.

Using Arizona nonprofit institutions as an example of data available in the interactive view:

Update 12/8: Added clarification that most of the slowdown in growth comes from the Some DE category.

Update 12/9: Added specific description of IPEDS data view summary.

Seven Things We Mostly Know About the Planned Instructure Acquisition and Three We Don’t

Author Phil Hill, Blog December 4, 2019/6 Comments/by Phil Hill

There is obviously a great deal of interest and speculation about what the Instructure news means for Canvas and the LMS Market moving forward. See this morning’s post for a summary of the announcement. ¹Disclosure: Instructure, Blackboard, D2L, Schoology, and Moodle are all subscribers to our LMS Market Analysis service. I doubt we will get any official answers from either Instructure or private equity firm Thoma Bravo before the expected completion of the sale in January or February of 2020, but we can look at Instructure’s recent history as well as Bravo’s history to get some clues. Here are seven things we mostly know already.

Seven We Know

Canvas is the market leader in the academic LMS market in North America and the fastest-growing LMS provider globally, although market activity has slowed significantly since 2018. Our market data shows the Canvas win rate for LMS adoptions has hovered at 80 – 90% for years, but since this summer it has dropped to 60% due to an increase in wins for D2L Brightspace.
Before the strategic review process that led to the sale agreement, Instructure had already been making moves to go beyond the LMS, creating Arc as a video platform in 2016, buying Practice in 2017, and MasteryConnect and Portfolium in 2019. Instructure even rebranded the product suite under the Canvas name in spring of this year – Canvas no longer means just the LMS, Arc is now Canvas Studio, etc.
Also before the review process, Instructure was making plans to separate out the corporate learning Bridge LMS – both in operations and in code base – into a subsidiary unit or a separate company (possibly as a sale). We don’t know the final form of this separation yet.
The agreement is for Thoma Bravo to acquire Instructure for $47.60 / share, above historical prices but 10% lower than this morning’s public market pricing. This share price leads to roughly a $1.79Bn market cap, which interestingly is very close to Blackboard’s $1.88Bn market cap (in 2019 dollars) deal when they were taken private in 2011. Unless I am missing something, the announced “$2 billion” sale seems like a generous rounding [update: I think the issue is net stockholders’ equity of $150 m in addition to outstanding shares.]
The deal is not complete, and the agreement includes a 35-day “go-shop” period where Instructure can solicit higher bids. Most investment analyses that I have seen view the $47.60 price to be a good offer and do not expect a bidding war. In either case, the deal is not yet complete.
Thoma Bravo is widely know for the “buy and build” acquisition strategy, where a platform company with solid customer base is purchased (often for high price), enabling subsequent acquisitions of smaller companies that have lower price multiples. This is not the same as buying company A and B and combining them; rather, this strategy is based on multiple acquisitions tied to the platform company.
Thoma Bravo (or its predecessor) have been involved in education before. They backed the management buy-out of Datatel in 2005 and then sold that company to JMI Capital and Hellman & Friedman in 2009. Bravo more recently bought Frontline Education in 2017, which is an administration management software provider in the K-12 market. Frontline has been on a acquisition spree, similar to PowerSchool and its Vista Equity Partners owner, following the “buy and build” strategy.

Three We Don’t

What does this mean for Instructure’s future? One guess is to expect the Bridge separation with increased focus on the Canvas / academic business to happen quickly. Another guess is to expect Instructure to ramp up their corporate acquisitions starting in 2020. But there are at least three main questions we don’t know and will have to watch.

What we don’t know yet includes how long the current management team will remain in place. Instructure made a point in press releases that the “Instructure management team, led by CEO Dan Goldsmith, will continue to lead the Company in their current roles”, but in internal FAQs this is qualified as “for now, our senior management team will remain as is”. It is commonly assumed that most of the current Instructure executive team was brought in primarily to make Bridge work in the corporate learning markets while maintaining Canvas, and this increased focus on Canvas and the academic markets is a change in strategy largely based on investor pressure. CEO Dan Goldsmith, Chief Strategy Office Jennifer Goldsmith (Dan’s sister), EVP of Sales Frank Maylett, SVP Strategy and Operators Amanda Buckley, Chief Marketing Officer Marta DeBellis – all of them come from corporate markets with no apparent education experience. And it’s worth remembering Blackboard’s history upon the 2011 acquisition, with public plans to keep the same management team, but within 15 months there was a mostly new executive team. None of this is to say that you have to have education experience to succeed in the education markets, but I consider this an area to watch closely, as it will impact corporate strategy and culture.
What we also don’t know is whether the planned acquisition will lead to price changes for the Canvas LMS. This seems to be one of the most asked questions I’ve heard since the strategic review announcement, but I don’t think the answer is clear. The “build and buy” strategy does not imply maximizing revenue from existing customers on the core platform – the real bet is more often on a combination of organic growth (new customers on existing platform) and corporate acquisition-based growth. Prices could go up more than current annual increases, but we do not know yet. I would expect increased cross-selling of related products, however.
One additional question is what impact this sale announcement will have on the LMS market. I suspect that D2L and Blackboard are busy revising their marketing and sales messages, and Instructure cannot just assume the same retention and win rates. Corporate transitions in the academic LMS space tend to cause big market impacts and Instructure has its work cut out for itself.

Disclosure: Instructure, Blackboard, D2L, Schoology, and Moodle are all subscribers to our LMS Market Analysis service.

Private Equity Firm Thoma Bravo to Acquire Instructure for $2 Billion

Author Phil Hill, Blog December 4, 2019/1 Comment/by Phil Hill

That was a short process. Less than three weeks after Instructure announced that it was considering strategic alternatives that could involve an acquisition, the company has agreed to be acquired by a private equity firm, Thoma Bravo. ¹Disclosure: Instructure is a subscriber to our LMS Market Analysis service.Bravo also owns Frontline Education, a school administration vendor for the K-12 market. From Instructure’s press release:

Instructure (NYSE: INST) today announced that it has agreed to be acquired by Thoma Bravo, LLC, a leading private equity investment firm, in an all-cash transaction that values Instructure at an aggregate equity value of approximately $2 billion. As part of the terms of the agreement, Instructure stockholders will receive $47.60 in cash per share. The price per share represents an 18% premium to the Company’s 3-month volume-weighted average price as of October 27, 2019, the day prior to the Company’s third quarter earnings call at which it announced a strategic review for its Bridge business.

The main emphasis in the initial press releases is the continuity of the management team.

The Instructure management team, led by CEO Dan Goldsmith, will continue to lead the Company in their current roles. Thoma Bravo will support Instructure as it increases investment in education technology innovation and expands internationally.

“Instructure believes the opportunity to become a private Company will provide additional flexibility and position us to invest more strategically to drive innovation for our customers,” said Goldsmith. “We look forward to working closely with all parties to complete this transaction and enter into our next chapter of growth and industry leadership.”

Obviously this is big news for the LMS market, and it is worth noting that in the press release Thoma Bravo specifically mentions Canvas but not Bridge, which should give a hint about future plans. And perhaps more importantly, the press release explicitly mentions future growth that will or could include M&A – Instructure buying other companies.

“Instructure’s Canvas product is the gold standard for learning management systems in the global education market,” said Holden Spaht, a Managing Partner at Thoma Bravo. “We are excited to partner with Dan and the senior management team to support continued investment and innovation in the Company’s market leading products and world class customer support.”
Brian Jaffee, a Principal at Thoma Bravo added, “We’ve followed the impressive Instructure growth story for many years and believe Canvas is a highly unique vertical market SaaS leader with exciting scale and future growth potential. We look forward to building on the strong momentum in the business and accelerating growth and product investment both organically and through M&A.”

Also worth noting is that there is a “go shop” period of 35 days that allows Instructure to receive other bids.

Stay tuned for more coverage on the news and the likely impacts on the LMS market in particular and EdTech market in general.

Disclosure: Instructure is a subscriber to our LMS Market Analysis service.

Christensen Scorecard Update: Data through Fall 2019

Author Phil Hill, Blog December 1, 2019/by Phil Hill

In April I used data from our partners at LISTedTECH to look at the predictions made by Clayton Christensen and Michael Horn about upcoming college closures and mergers. We now have updated data for 2019 as well as new views based on the total number of students affected by schools going out of business. You can see more analysis over at LISTedTECH’s blog post, where Justin breaks out a lot of data.

Prediction Redux

As a reminder, in 2013 Christensen made a bold prediction based on his ubiquitous innovation theory that maybe half of all postsecondary institutions could close within 10-15 years.

(source: https://youtu.be/KYVdf5xyD8I, starting at 6:25)

The scary thing is that 15 years from now, maybe half of the universities will be in bankruptcy, including the state schools. But in the end, I’m excited to see that happen.

Christensen then doubled down on his predictions in 2017, humorously saying it might take nine years instead of ten.

Q. Do you still believe, as you’ve said before, that as many as half of colleges and universities will be bankrupt or closed within a decade?
A. Um, yes. [snip] Whether the providers get disrupted within a decade — I might bet that it takes nine years rather than 10. Maybe I’m too scared about the Harvard Business School to be rational about it. But we should worry.

Read the original post to get a deeper examination of whether the predictions are realistic given the data since 2015. The gist of the data analysis was that the vast majority of closures have been from the for-profit sector (which did not seem to be in scope of Christensen’s comments, but we can’t be sure), but the predictions could be realistic if including for-profits but not if applied only to nonprofit institutions.

Updated Data

Since April we have had quite a few news stories about new closures or mergers. Marlboro College, Southern Vermont College, Green Mountain College, etc. How do the new data affect the trends? Below we share updated running totals of closures and mergers since 2011 and running through Fall 2019.

If you include all degree-granting institutions (i.e. for-profits as well as private nonprofits and publics), then the current trends lines show that the 50% closure prediction by 2028 certainly seems feasible. Note, however, that there are fewer than 1,000 for-profit institutions remaining as of Fall 2017 IPEDS data, and the rate of for-profit closures cannot continue more than another 8-10 years (best case / worst case, take your pick).

If you ignore the for-profit sectors, however, then the trend line for private nonprofit and public institution closures & mergers remains far below that needed to hit the 25% level described by Horn or the 50% level described by Christensen. None of this is to say that the trends moving forward will be linear, however.

Christensen does get credit for the timing of the original prediction, as closures and mergers really accelerated starting in 2014.

Adding Enrollments

Campuses vary in size, and counting the number of schools does not capture the full picture. LISTedTECH has updated their enrollment analysis to look at both numbers of institutions and size of total enrollments in the US over time, based on IPEDS data. We are working to generate a running total view of closures that is based on enrollments (i.e. how many students have been directly impacted by schools going out of business) and will share that with you early next year. Below is a chart shared by LISTedTECH. Read through over at LISTedTECH, as they also break down by degree type and sector.

Update on Chegg Data Breach: Decrypted credentials now leading to multiple campus security attacks

Author Phil Hill, Blog November 25, 2019/by Phil Hill

In September 2018 on e-Literate I wrote about a data breach at Chegg that potentially affected up to 40 million student accounts. The breach itself occurred in April of that year, with the discovery of the breach occurring a week before the notice, as described in the company’s disclosure to investors:

On September 19, 2018, Chegg learned that on or around April 29, 2018, an unauthorized party gained access to a Company database that hosts user data for chegg.com and certain of the Company’s family of brands such as EasyBib. The Company understands that the information that may have been obtained could include a Chegg user’s name, email address, shipping address, Chegg username, and hashed Chegg password. The investigation into the incident, which is supported by third-party forensics, is ongoing. To date, the Company understands that no social security numbers or financial information such as users’ credit card numbers or bank account information were obtained.

Mainstream channels focusing on stock price effects (Bloomberg, CNBC, TechCrunch, etc) and security-specific channels (Security Today, DataBreaches, etc) covered the story.

In a subsequent post on e-Literate, I noted the lack of transparency from Chegg.

What is problematic with the Chegg data breach is that no further information has been made public and there has yet to be any interest from the broader ed tech press to dig up answers. We have no idea how serious this breach is, and I do not believe that the users with compromised personal information have had any updates since the initial email blast and associated post.

This behavior continues despite a new exposure resulting from the 2018 breach.

Information Decrypted and Exposed

In September of this year the data appears to have been decrypted (reversing the hash-based password encryption) and posted online, exposing student email and password combinations as described in a Tulane University IT announcement: [emphasis added]

In 2018, Chegg.com, a company that offers textbook rental and tutoring services suffered a data breach resulting in the loss of 40 million customer records which included full names, email addresses, usernames, and passwords. Now the results of Chegg’s 2018 data breach have been publicly exposed online, which poses a substantial threat. If users happen to reuse or slightly modify passwords across multiple services, publicly exposed credentials can be exploited.
Several universities across the nation have reported malicious use of these leaked email addresses and passwords which were contained in the Chegg breach. This is of concern for anyone who may have signed up for Chegg using their Tulane email address and a password similar or identical to their Tulane.edu password. Several Tulane students and alumni have reached out to the IT service desk affected by this breach of Chegg data.

An article from Saint Mary’s College describes that the discovery of this decryption was not described to schools or users by Chegg, but rather by a separate security agency.

Junior Sophie Koeppl, a Chegg user since her freshman year of college, said she was alerted to the breach by the College and was not contacted directly by the textbook provider.
“I never got an email from Chegg confirming the security breach that happened last year,” Koeppl said.
Kathy Hausmann, associate director for technical support services at Saint Mary’s, said the information obtained in the 2018 breach potentially included a Chegg user’s name, email address, shipping address, Chegg username and hashed Chegg password.
“Saint Mary’s College received a notification from REN-ISAC (Research and Education Networks Information Sharing and Analysis Center) ‘that some credentials from your institution have appeared in a credential dump related to the Chegg data breach,’” Hausmann said in an email. “The information obtained from the Chegg data breach had been shared online for others to do further damage beyond the initial data breach of Chegg.”

The University of Central Florida’s IT advisory described how Chegg’s method of hashing passwords appears to have led to the decryption and release of comprised information online.

According to a news alert sent by UCF INFOSEC Wednesday morning, the weak encryption used to store user data was the primary cause of cyber criminals gaining access to the information of thousands of users.

What about Chegg?

Where is Chegg in all this year’s news? It appears that the company is continuing its approach of trying minimize the news about security breaches and primarily notify the investment community. I could find nothing in public where Chegg is notifying people of the decryption and comprise of username / password combinations.

The only public mention of the 2018 data breach, ironically, came from an investor call in early November where an analyst asked about a similar data breach at Thinkful, a company that Chegg had just acquired. CEO Dan Rosenweig did not take the opportunity to describe the decryption of the 2018 data and even described the incident as having no impact. [emphasis added]

Let’s start with the news. So it was — so no, we were not cognizant ahead of the announcing of the [Thinkful] data breach.
We obviously were made aware, the moment that they knew. And fortunately, similar to ours, which, as you know, amounted to nothing. There was no pertinently identifiable information, no credit card information. And they don’t have — their customer base is much smaller than ours, obviously, and they charge a lot more.
So it was a non-variable as best as we can tell. And everything that we looked at afterwards, we’ve asked them to follow the exact same protocols that we did, which was complete and open transparency, communicate immediately, notify everybody. Openly notify people because, unfortunately, it’s a fact of life in technology that these things are going to happen. But fortunately, neither one of us has the PI or the credit card information.

Amounted to nothing? We now have new exposure of decrypted information with documented increase in malicious attacks. While I am not aware of further loss of information from this situation, I would not call it nothing. As described in the Saint Mary College notification:

“After hashed passwords are decrypted, the passwords can be used to sign into affected accounts if the passwords were not already changed,” Hausmann said. “There is also the concern that the released e-mail addresses and passwords could be used to try and gain access into accounts unrelated to Chegg, including e-mail, social media and finance-related websites.”

Complete and open transparency? Chegg waited a week to send an email to users about the breach discovery last year and only later made a public post. And the company has made no “complete and open transparency” attempt based on the September 2019 decryption and exposure of information, according to the news articles.

This evolving story should be a wakeup call to the EdTech community to take data security more seriously, and for institutions to demand greater transparency from vendors.

Why Instructure’s News Matters: Big Tech

Author Phil Hill, Blog November 22, 2019/by Phil Hill

This is part 2 of a short series examining the significance of Instructure’s announcement that they are going “to explore strategic alternatives in order to maximize shareholder value”, where the alternatives could include selling the company to financial buyers (e.g. private equity) or strategic buyers (e.g. big technology). Of course, one alternative is to not make changes and remain a publicly-traded company. In part 1 I looked at the historical context of the LMS market, and in this post I’ll explore the context of Big Tech (i.e. involvement of Amazon, Google, and Facebook) in this story.

There are at least three relevant parts of the Big Tech story – the rise of cloud hosting, the international expansion of AWS and Canvas, and the entry of Google Classroom.

Disclosure: I will note more prominently than usual that Instructure, Blackboard, D2L, Moodle, and Microsoft are all subscribers to the MindWires LMS Market Analysis data service, and that we have a number of investment firms who are also subscribers to the service and pay for in-depth market data and research.

Cloud Hosting

To understand the significance of cloud hosting to the rise of Canvas, we can start with Bill King’s comments in the first post.

One market dynamic that I’m sure you are aware of, but is worth including here, is why Canvas could take off so quickly (I was in sales at WebCT, and then a non-LMS edtech vendor, who moved from on-prem to cloud deployment, and then Canvas, before leaving academic IT system sales). While the ease of use and faculty-friendly interface and design was important, so was AWS (Amazon Web Services). CIOs by nature are conservative — generally, they don’t like change, because all the change falls on their shoulders, with limited staff, and usually with a bigger demand that they get the SIS, ERP and other admin systems correct. This left little time for academic systems. Even when there was a strong academic tech department, Academic tech was usually classroom tech. The area of servers (and network traffic) still had to go through the CIO office. Which reminds me – at the time, another big IT project was upping the bandwidth capacity across campus. IT had LOTS to do.
Canvas came along with a cloud-based deployment scheme that eliminated the need for the CIO to source servers and stand up instances (let alone fail-over and backup systems). Even when early vendors moved to “managed hosting” (where a third party ran your server license for you), often times the CIO required proof of the safety of the data — at rest and in transit. You also had to prove the up time commitment (I remember early boasts of 93%, which is laughable today). So, you had to sell your product AND you had to sell the hosting service. With Canvas, you could just point to AWS’s terms of service, and EVERYONE knew Amazon was rock-solid reliable and secure. Whether that was true or not (I generally think it is), the point is that Amazon was like IBM from a few decades ago. Everyone trusted them and, even if AWS went down, it impacted everyone, so your IT group would not be blamed.
This, to me, is the key why faculty and academic technology were free to choose their LMS platform. The entire IT services equation was taken care of, and AFAIK, Canvas had first mover advantage here. Then you throw in their culture at the time, and the product, and it was never a question of “if” – it was always a question of “when.” (e.g. I remember talking with one major private research university in 2014, that had planned their Canvas move for 2017, once all their legacy contracts expired and a few other internal issues were taken care of.)

I think Bill gets this description just right and ties together these first two posts, but I’ll add two comments.

Instructure’s competition made strategic mistakes in dismissing the allure of Canvas, and also to cloud hosting (also described as Software-as-a-Service, or SaaS). But nine years after the launch of Canvas, the situation has changed dramatically. D2L has recently completed its rearchitecture of the Brightspace LMS to be fully deployed on AWS – a three year effort. Blackboard is roughly a third of the way through its changes to cloud hosting. The significance here is not just having SaaS deployment options; it is also on the freedom to manage a unified code base and the ease of regression testing, introducing new features on one configuration, and ensuring clients are using the latest and greatest versions. In higher ed, Canvas is no longer the sole fully cloud-based LMS – add D2L Brightspace. In K-12, add Schoology and Google Classroom as two additional fully cloud-based LMS solutions. While there are variances in how well each company manages its cloud deployment, Canvas no longer has a completely unique approach.

Percentage of Blackboard Learn in Higher Ed and K-12 on three different hosting options

Bill and I both mention AWS and not just as an example of Big Tech cloud infrastructure. With little notice, EdTech infrastructure has moved en masse to AWS. Canvas, D2L Brightspace, Bb Learn, Schoology, and most other EdTech platforms such as adaptive learning tools and competency-based platforms. Microsoft is making a bigger play in higher ed with Azure and Teams and business intelligence, but for Infrastructure-as-a-Service or Platform-as-a-Service, AWS is king in EdTech.

AWS and International Growth Opportunities

A corollary to the importance of Big Tech-based cloud hosting is that much of Instructure’s growth – past and future – is strongly tied to the expansion of AWS data centers worldwide. When AWS was purely hosted in the US, there were regions that were out of reach for Canvas or any other AWS-based system. There are policies and regulations in many world regions that prevent their core data from being processed outside their boundaries, particularly in the US (thanks, Mr. Zuckerberg).

While the first AWS region in Europe was opened in Ireland in 2007, it wasn’t until there was a mainland AWS region in Frankfurt (opened October 2014) that Canvas really had a chance to expand. The University of Birmingham was the first major university win for Canvas in Europe, and that was when Canvas gained credibility and started their expansion. Now, Europe is pretty well covered, and the Middle East opened a new region just this year.

AWS Regions and Availability Zones as of November, 2019: https://aws.amazon.com/about-aws/global-infrastructure/regions_az/

South America, meanwhile is quite limited with only two locations, both in Brazil. Canvas is slowly building a customer base there, but . . . slowly. I suspect AWS regional coverage is part of that reason. Keep in mind that much of Latin America does not view Brazil as similar to them.

Blackboard and Moodle have both had long-term regional coverage in the Middle East, Latin America, and in Asia Pacific, but a big part of the reason is their availability for self-hosting or regional managed-hosting options. It will be interesting to see how well Blackboard can migrate these customers to Learn SaaS (on AWS) and how that impacts Canvas future growth trajectories. Moodle is similar in having these other options available and a long history in all regions, but there are no plans even considered for the open source options. No migrations forced there.

Higher Education LMS market share in four global regions for primary LMS usage

Google Classroom

While Google Cloud itself is not a big player in EdTech, Google Classroom and G Suite for Education are very important – G Suite is huge in K-12 and Higher Ed, but Google Classroom is almost exclusively in K-12. In fact, Google Classroom has enormous interest and is growing quickly, far surpassing Canvas in the excitement factor. The issue is that Classroom does not fully solve enterprise LMS issues and tends to be used as secondary system (where the school district uses another system as the primary LMS, i.e. faculty- or department-based adoption) or for school districts not looking for a full enterprise solution with all of the reporting, advanced grade book, assessments and assessment management, and complex roster management.

You can see from the chart below that there is a Big Four of K-12 – Moodle, Google Classroom, Canvas, and Schoology – for primary systems in North America. But for secondary systems, it’s not even close.

Instructure does not make as much money in K-12 markets as they do in higher ed, but with the investor pressure to focus more on Canvas at the expense of Bridge, the K-12 market dynamics are important to their future. A strategic buyer (another, larger technology company) would likely value Instructure and Canvas for its total footprint and not just revenue potential, and a financial buyer might be interested in combining Instructure with another company.

In the next post we’ll deal with the context of how Instructure is responding to news of the strategic alternatives review and to the broader context around LMS history and Big Tech, along with privacy & data usage opportunities and concerns.

Why Instructure’s News Matters: Market history

Author Phil Hill, Blog November 18, 2019/10 Comments/by Phil Hill

Last Thursday I shared Instructure’s news release that they are going “to explore strategic alternatives in order to maximize shareholder value”, where the alternatives could include selling the company to financial buyers (e.g. private equity) or strategic buyers (e.g. big technology). Of course, one alternative is to not make changes and remain a publicly-traded company.

Why is a potential change as mundane as having Instructure’s shares traded in public markets vs. being owned by a larger company newsworthy? After all, Instructure is growing, an established LMS market leader, and is not in financial trouble. Their stock price is even at its all-time high.

The reason this announcement matters is all about context. The history of the academic LMS market, the general environment of distrust of big technology and its data usage, and Instructure’s public marketing and communications are all key to understanding today’s situation. In this first post I’ll deal with market history.

Disclosure: I will note more prominently than usual that Instructure, Blackboard, D2L, and Moodle are all subscribers to the MindWires LMS Market Analysis data service, and that we have a number of investment firms who are also subscribers to the service and pay for in-depth market data and research.

Market History

When looking at the early history of the academic LMS market, from its inception in the mid 1990s through roughly 2011, the market was dominated by Blackboard. Most LMS market dynamics could be understood through the lens of Blackboard’s market position, its corporate acquisition strategy, its poor reputation for reliability and usability, its product pricing, and the market’s reaction to Blackboard. This was the situation that led to the original LMS ‘squid’ graphic, which was my attempt to help US higher education institutions understand how to interpret market dynamics.

On the left side of each band is the origin of systems, and these LMS solutions mostly originated as projects at specific universities, with moves to commercialize the products happening after the fact.
The bands coming together show how Blackboard’s core strategy was acquisition of their competitors, eventually moving clients onto what is now known as Blackboard Learn. After Blackboard went public in 2004, their biggest acquisitions were of WebCT (#2 in market at the time) in early 2006 and of ANGEL (a growing #3 or #4 in the market) in 2009. Not shown is the 2006-2009 patent lawsuit against D2L (#3 and known as Desire2Learn at the time), where I (along with many others) believe that Blackboard tried to put that company out of business or soften it up for an acquisition. D2L ended up fighting and winning this patent lawsuit battle, but the other companies and their products are gone.
Based on the intense pushback against this corporate behavior, the open source market became the perceived safe haven, a way for schools to retain control and not be forced to migrate. Sakai originated in 2004, with Moodle before that, but the real grown of open source LMS adoption occurred in 2007 – 2011 in reaction to Blackboard corporate moves.

Why does this history matter? Because the academic market has a long memory, and many educators (faculty, staff, and administrators) deeply remember the days of corporate finance driving the market. Losing control of university projects, being forced to migrate from chosen LMSs, rising prices and constant cross-selling ¹e.g. Bb Collaborate, created by acquisition of Elluminate and Wimba and upselling ²e.g. move from Bb Basic to Enterprise, to Academic Learn suite, with constantly rising prices, systems going down during Finals week, clunky frustrating systems, etc.

Change in Market Dynamics

The market changed in 2011 for two reasons. One was that Blackboard was taken private by a private equity firm, leading to the early 2013 change in executive team and the end of corporate acquisition of competitors. Blackboard acquired Moodlerooms in 2012 (now known as OpenLMS), but note that this product is still alive – there was no forced migration, and Blackboard has contributed heavily to the open source community. Note: Moodlerooms was the largest Moodle Partner and is not the same as Moodle HQ, the company that leads the development of the Moodle open source product. ³See Michael Feldstein’s post at e-Literate to get a description of the Moodle ecosystem.

The second reason is that Canvas emerged as the fastest growing LMS solution – an intuitive, cloud-native product with a company with a unique corporate culture. From 2011 through the present, Canvas began to dominate the market dynamics – either through its constant growth with no losses, or from competitors’ reactions to Canvas. We now have much more reliable systems with better usability built in, not just in Canvas but in particular with D2L Brightspace and increasingly with Blackboard Learn itself. While many people are frustrated with the pedagogical limitations of most LMS solutions, the market is much healthier and more focused on educators’ needs than it was a decade ago.

One related change in the market is that the primary buyers of an academic LMS – the most important people in a decision on which LMS to adopt – has shifted from the CIO to the faculty and academic technology support. This shift happens along a spectrum and differs among schools, but there have been dramatic changes. A decade ago, LMS migration mostly occurred in spite of faculty wishes. CIO-type needs, or product end-of-life coming from acquisitions, forced schools to change, and academic types (faculty, academic tech support) at best had input in reaction to these needs. Today, quite often it is the faculty or academic technology support staff that are driving LMS adoptions, requesting changes with the CIO and IT staff having to react to these demands.

Underlying Tension

There has been an underlying tension this whole time, however, with many educators wondering whether and when the market might shift back its previous dynamics. Would corporate finances come back and take priority over customer satisfaction, would Canvas become just like corporate players of the past market?

It is this underlying tension that still exists and in my opinion is increasingly coming out in reaction to Instructure’s corporate moves recently. Many of the details of the market history are forgotten or jumbled, but the underlying tension is deeper than the surface language. The sentiment is more intense for US higher education than for K-12 markets or for global regions outside of North America, partially as these other market segments never viewed Canvas quite as intensely as the market champion. But the tension is still there.

Why does this news matter? From the market history perspective, there is a risk that the potential sale of Instructure will unleash more of this market sentiment against the company, impacting future market growth. And this sentiment is particularly intense among the academics that have driven the growth of Canvas. At the same time there is a chance that Instructure will come out of the process better able to serve the academic markets. The result of the risk and opportunity mostly depends on whether Instructure is sold, to whom, and how the company responds in public during this process. But there is a lot on the line and underlying questions are being raised. In this market, I would not discount sentiment based on history.

In the next post we’ll deal with the broader context of Big Tech.

Instructure Considering Sale Options

Author Phil Hill, Blog November 14, 2019/7 Comments/by Phil Hill

Instructure, maker of the popular Canvas LMS, announced this morning that they are reviewing strategic alternatives.

Instructure (NYSE: INST) today announced that in response to interest received from multiple third parties, its board of directors, supported by management, has commenced a process to explore strategic alternatives in order to maximize shareholder value. These alternatives may include continuing as a standalone public company, going private, or being purchased by a strategic partner. The board has retained J.P. Morgan as its financial advisor and Cooley LLP as its legal advisor in connection with the review.
“Consistent with its duties and in response to interest from several parties, our board of directors has determined that it is prudent to undertake a review of alternatives to identify the best way to maximize shareholder value,” said Josh Coates, Chairman of the Board at Instructure. “As the board conducts its review, we remain focused on executing our strategy with an increased focus on the education market as well as the previously-announced strategic review of Bridge.”

This news comes on the heels of several articles about investor pressure being applied on the company.

April 2019 – Activist investor Praesidium takes 5+% stake in Instructure and pushes thesis that Instructure would be more valuable by selling off Bridge (corporate learning market) and concentrating on Canvas (academic market).
Early October 2019 – Instructure announces investor day on December 3rd where “management team will describe the company’s new strategy and operating model, key decisions about future financial goals, and the plan for focused capital allocation”. (couldn’t find direct link anymore, so using Google cache)
Late October 2019 – Activist investor Sachem Head takes a stake in Instructure and pushes company “to explore alternatives, including selling itself, two people familiar with the matter said”.
November 2019 – Praesidium joins Sachem Head in call for Instructure to explore a sale.
Today – Jana joins activist investor pool.

As part of today’s announcement [emphasis added]:

The company has not set a formal timetable for this exploration, nor has it made any decisions related to strategic alternatives at this time. While Instructure is pursuing opportunities, there is no assurance that the process will result in a transaction. The company does not expect to make additional public comment regarding these matters until the board approves a specific action or otherwise concludes the process.
The previously scheduled financial analyst day on December 3 has been canceled to allow management and the board to explore these strategic alternatives for the company.

Disclosure: I will note more prominently than usual that Instructure is a subscriber to the MindWires LMS Market Analysis data service (as are many of their competitors), and that we have a number of investment firms who are also subscribers to the service and pay for in-depth market data and research.

There will be more analysis coming in the future, but I would note several items to set context, especially for academic readers:

Bridge is a corporate learning LMS created by Instructure (not a corporate acquisition) starting in 2015, the same year that Instructure went public. This additional product line, with its larger available market, was a key part of Instructure’s IPO story.
When Blackboard went through a similar process and actually went private (i.e. a private equity firm acquired the company, taking them off of the public stock market), the process lasted from early summer 2011 until October 2011. That should give some sense of the timeframe of these decisions.
Blackboard also considered a sale in 2015 but did not actually change ownership – these processes do not always result in a sale.
At the risk of annoying some readers (again), I have pointed out that “there are strong arguments that company financial health in 2018 / 19 for the providers will continue to have an outsized impact on the future of LMS offerings”. Consider PowerSchool’s planned acquisition of Schoology and Blackboard’s recent debt ratings news (more on that in future post). This news is as much about the LMS market as it is about one company.
For academic community, it is worth noting that much of the investor-based pressure is for Instructure to focus more on supporting Canvas, not less. Instructure management has made it a point to say that they are increasing investment in Canvas, but today’s news puts even more emphasis on that need.

Keep watching – this should be interesting for the LMS market.

Update 11/14: Added note about Jana disclosing investment. Also added bullet about Bridge history.

Bb Learn Hosting Models: New global data for LMS market

Author Phil Hill, Blog November 10, 2019/by Phil Hill

Justin Menard from LISTedTECH (our partners in LMS market analysis service for higher ed and K-12) shared today that we now have the ability to track Blackboard’s Learn LMS hosting models. Self-hosting, Managed Hosting and SaaS Hosting (software-as-a-service, aka cloud hosting) are now captured in the data – globally for higher ed and in the US and Canada for K-12. This is a great addition, and to my knowledge we are the only ones to have this data in this scope. Head over to Justin’s blog to read more on the data additions. Also note that we will be validating and updating this new data over time, so treat this as our initial views. ¹Disclosure: Blackboard, D2L, Instructure, and Moodle are subscribers to our LMS market analysis service.

Part of the reason this is so important is that Blackboard, their competitors, and many in the investment community agree that a client moving to SaaS hosting changes their likelihood of remaining with Blackboard, and not just due to the contract extensions involved in the SaaS move. The movement to Learn SaaS and then to the Ultra user experience remains the core of Blackboard’s corporate strategy.

We can now provide independent analysis of the global transition to Learn SaaS, whereas previously we tracked this transition based on public Blackboard statements.

Our data confirms that roughly 30% of clients have shifted to Learn SaaS:

Recently Blackboard announced that they are making significant progress in moving many of their Middle East clients to Learn SaaS, which is impressive given the historical tendencies in this region to strongly favor self-hosting models. From the press release:

Blackboard, a leading education technology company for teaching, learning and student engagement, announced today that many of its clients across the Middle East continue to transition to the Software as a Service (SaaS) deployment of Blackboard Learn, the company’s flagship learning management system (LMS). [snip]
The benefits of moving to a cloud-based delivery model include continuous delivery of new features, zero-downtime updates and quicker delivery of fixes. Once institutions are on SaaS, they can enable Learn Ultra, Blackboard’s next-generation LMS, which offers educators and students a clean, modern interface and intuitive workflows.

Using our data, we see that the Middle East has indeed started a significant shift to Learn SaaS hosting models. In Justin’s post you can see that approximately 20% of Blackboard Learn clients are now on SaaS, and interestingly they shifted directly from self-hosting (i.e. there is no managed hosting usage in this region).

What is interesting is to look at the age of Blackboard Learn implementation for each of the current hosting models. In other words, for each hosting model of Learn, how long have those schools been Blackboard LMS clients.

For higher ed clients, there is a broad range of ages moving to Learn SaaS, yet for K-12 there is a narrow range of mostly 4-6 year clients that have moved. Part of this is likely due to K-12’s smaller support staff and budgets for LMS, which also has lead to a smaller percentages on Learn SaaS than in North American higher ed (19% vs. 33%).

We expect this data addition to open new capabilities for us to analyze the LMS market and its underlying trends. Instructure’s Canvas was created designed on cloud hosting, and D2L’s Brightspace has recently completed its transition to the cloud. The story is broader than Blackboard.

Disclosure: Blackboard, D2L, Instructure, and Moodle are subscribers to our LMS market analysis service.