Throughout the past decade, Knewton’s adaptive learning technology has been backed by some of the biggest names in the both the publishing and venture capital community. Now one of its most high-profile content partners and investors, Pearson, is pulling back.
Pearson will no longer use Knewton’s adaptive learning engine for some of its digital offerings. The retrenchment delivers a setback to Knewton, for whom Pearson was the first—and most visible—partner. The publisher co-led the New York City-based company’s $33 million Series D round in October 2011, hailed at the time as a strategic investment and a major vote of confidence in Knewton’s potential. (The company has raised more than $157 million in venture capital.)
Now, Pearson is “investing heavily in product development and is developing its own in-house adaptive learning capability,” Scott Overland, Pearson’s Director of Media and Communities, wrote in an email statement to EdSurge. “These innovations have allowed us to in-source our adaptive learning capability, and thus we are phasing out our agreement with Knewton for select areas of our global product portfolio.”
Among the Pearson products that will phase out Knewton’s technology include those in the MyLab and Mastering suite, which cover a wide range of higher-ed courses in math, economics, English, sciences and engineering. (Knewton technology is also used in enVisionMATH2.0, a Pearson offering for grades K-8, but neither company offered specific details on the status of this agreement.)
“Our understanding since last summer is that Pearson is building its in-house capabilities around adaptive learning and is planning to gradually reduce its use of Knewton. The partnership is a long and important one for us; while a result of this will be that we may stop powering some Pearson products in the future, we expect to remain a key feature in others,” said a Knewton spokesperson.
Knewton also provides the adaptive technology backbone to other companies and traditional publishers—including Houghton Mifflin Harcourt and others that directly compete with Pearson. Through these partnerships, Knewton tags digital content with metadata and maps them onto a learning path. As students work through the materials, Knewton’s algorithms detect where they are struggling and surface the appropriate content from the publisher.
Adaptive learning capabilities have become a key value-add for many content publishers as they transition from print to digital offerings. The value proposition that Knewton and its competitors have offered to these companies is that by collecting user data, the system can infer what a student knows—or doesn’t.
“The notion that adaptive technology is the reason why one school should choose one company’s content over OER (open educational resources) or other options” has become a staple of many publisher’s marketing claims, says Trace Urdan, an education market analyst.
Knewton’s website currently lists 32 partners that include blue chip tech companies (Microsoft, HP), online schools (Florida Virtual Schools) and non-U.S. content providers. Not all publishers hopped onboard. McGraw-Hill has built its own adaptive platform, LearnSmart, in-house through its acquisition of ALEKS and Area9. The Telegraph estimates that McGraw-Hill has spent more than $700 million on these efforts.
According to Jose Ferreira, Knewton’s founder and former CEO, Pearson had been reviewing the business relationship for some time. “Every few years Pearson will tell us that they’re considering pulling back on our partnership, and they’ve done so for higher ed,” he told EdSurge during the ASU+GSV Summit, an education business conference.
At a meeting with investors last summer, Pearson CEO John Fallon said: “Everything that sits in the MyLab products going forward and in the next generation” will use technologies developed in-house. Tim Bozik, Pearson’s President of Global Product, followed up: “The implications of which is [that] we’ll be phasing out Knewton.”
Not all publishers have been willing to commit the time and resources required to use Knewton effectively. Ferreira offered this vehicular analogy: In the eyes of some publishers, he says, “Knewton is a Ferrari, but we’re in a Kia market. Ferraris require more maintenance. It’s more complicated to use Knewton.”
Ferreira left Knewton in December 2016 but remains on its board of directors. He is now CEO of a stealth startup, Bakpax, but did not offer any detail on the new company other than “it’s a big data and education play.”
Pearson’s decision to pull back from its Knewton partnership coincides with a broader retreat across the entire company. Earlier this month the publishing giant announced it is putting its U.S. K-12 courseware business under “ strategic review.” (That’s usually the corporate jargon for trying to sell.) It has already scaled back support for its learning management system tools.
“Over the last few years Pearson has been simplifying and streamlining every part of the company. This has involved not only our business footprint and our geographic footprint, but also our technology footprint,” said Overland.
Pearson has already courted another company to enhance its higher-ed digital learning products: IBM. “Through conversation and the use of natural language, the Watson-based adaptive learning technology will help students gain a better understanding of the subject material,” reads the press release about the partnership announced last October.
Partnering with IBM to build adaptive capabilities seems “more plausible than Pearson trying to do it entirely on its own, given its current internal restructuring,” says Richard Garrett, Chief Research Officer at Eduventures.
