One of the most popular EdTech themes and start-up ideas has been in the “Gamification of Learning” and the new opportunity for mobile educational gaming apps. Indeed, Educelerate has already organized multiple meetups on the theme in Los Angeles (EdSurge recap here) and the Twin Cities this year. Meanwhile, posts on the subject stretch from EdSurge to TechCrunch to Quora. Knewton has cranked out an Infographic on the subject (graphic to the right), Lynda.com an online course, and eSpark Learning counts over 100,000 third party educational game titles for the iPad alone!
However, there is a dirty little secret lying buried within the App Store: consumer-led educational game apps do not monetize – the business model is inherently flawed.
(Yes, business outcomes should not be the primary measure of new educational models, but not everything can be funded by the state or charity and without monetization, we can not truly sustain education innovation).
A Brief History of Educational Gaming
Every new computing platform brings a rush of new game releases, educational or otherwise. This was certainly seen with the Apple II and later the IIGS (1986), which inspired updates to one of the best educational games ever, Oregon Trail (originally built for mainframes by Minnesota school teachers in 1971) and, in so doing, spawned the world’s first EdTech Unicorn, the Minnesota Educational Computing Consortium (later, Corp). Indeed, MECC became one of the largest software producers of the 1980’s and realized a $370 million exit in 1995 to The Learning Company (which in turn sold for $3.8 billion to Mattel – still the largest edtech exit on record as well as an Anti-Case Study in horrific corporate deal-making, though one of its key architects went on to to become an investing “Shark”).
Another historical EdTech (paper) unicorn, Lightspan, was the first to monetize educational gaming for the Sony Playstation (and the PSP), until washing out in the Bubble’s burst and eventually selling to Plato Learning.
While Lightspan and MECC may be forgotten as corporate entities, titles like Oregon Trail or Number Munchers live on today via Facebook and iOS. You can even get a Chrome plug-in to play the original Oregon Trail online (or just go to the Internet Archive), and, frankly, despite all the possibilities of mobile, the original version still stands out for its educational layout.
The Current Market’s Rising (Red) Tide
The launch of the iPad in 2010 brought immediate impressive sales figures (with schools alone representing 3.5 million units by 2013) and overnight viral hits (Flappy Birds, anyone?), which inspired hundreds of developers to explore educational gaming ideas. Education titles sprouted up like mushrooms after the rain (doubling from 50,000 to 100,000 just over 2012) and investors soon followed looking to harvest. Nearly $250 million of venture capital has flooded just 9 consumer-focused and freemium educational game publishers (not including district sales led models like Dreambox and TTL or Kickstarter-funded plays like Reading Rainbow), with another 550 earlier stage start-ups fundraising on Angel.co (though some of these are mislabeled) and countless basement-dwelling developers cranking out code to teach their kid fractions or phonetics (here’s the Apps in My Pocket story).
If that sounds like a crowded market, it is. Indeed, nearly half of the 9 venture-funded start-ups above have already failed with two of them even pivoting and failing a second time (see DimensionU and Gazillion Entertainment). But these corporate failures are just a symptom of the underlying problem, consumer focused educational gaming apps do not monetize. Indeed, over this past winter, we watched firsthand as a client moved into the top five of the App Store’s Top Grossing Education apps, all off a daily intake of just two hundred dollars (yes, $200 — or $75,000 annually).
Osman Rashid (@osmanrashid) knows about the challenges of scaling in EdTech. After co-founding Chegg (one of just two EdTech IPOs during the current cycle), Rashid attempted to further disrupt the textbook industry with the Andreessen Horowitz and Intel backed Kno. With current start-up, Galxyz, he may be taking on his greatest challenge yet, direct to consumer science games that also bring extremely high quality and narrative to educational games. Rashid knows that the mobile educational game market is still very early in terms of consumer adoption / monetization, but has built his company to bridge this long curve, stating to us:
“The odds of building the next viral app are infinitesimal: regardless of the industry, scaling any company is hard. The games industry tends to more brutal since it is “hits” driven and requires most of the investment upfront. Education gaming is different in that is does not launch or monetize like consumer games: research in QA and testing of initial users over a longer period (including educational outcomes) are make or break. The revenue curve for educational games is less like consumer gaming hits and more like eCommerce, building out over time. We believe that Galxyz is a couple of years ahead of its time, which is exactly where we want to be. Revenue is less the driving concern at this moment than understanding what parents and teachers are willing to pay for – then we’ll scale.”
Running the Gauntlet of the App Store
For any investors considering co-investing in one of these consumer educational gaming apps, allow us to reiterate a point made in our earlier Educelerate post on “Why Silicon Valley Sucks at EdTech” (republished for its one year anniversary at Medium.com/@EdTech): parents do not spend on their children’s educational content. Leapfrog is a micro-cap stock and Crickets a nano-cap. And for anyone that points to the massive US markets for tutoring ($10 billion) or homeschooling ($1 billion), consumer-sold games will no more eat into this proven willingness to pay than they will public school tuition.
That long revenue curve is more like a flatlining EKG and the death of this business model is predicated upon four fundamental fallacies:
- Direct to Consumer: The fundamental challenge in consumer education sales is the unique multi-tiered sales dynamic where the publisher is selling both to the parent who pays and the child who plays. The broader video game industry only has to target the young adult gamer as they also pay. But in education, the additional parent validation, not to mention the general call for educational outcomes assessment, slows down the sale process. This unique market dynamic begins to get really complicated with elementary school aged children – beyond age 10, kids do not play anything their parents buy (and are not going to be clamoring for algebra games on their birthday), leaving the addressable market for consumer education games to just younger children. And yet, the most interesting pedagogical gamification value is with older kids. Younger children’s games are basically automated repetition around basic facts and the world is already awash in free versions of such (see CoolMath, MathGameTime, etc.). The only way to monetize the larger end of this market is through school district sales.
- Ineffective Search and Discovery: Osman Rashid further points out that one of the key challenges for any game is that “search” is as limited on the iOS App Store as with browsers fifteen years ago. Like prehistoric Google, iTunes search parameters limit an app’s description to just one page. Moreover, these descriptions are prone to fake keywords to manipulate traffic. While YouTube is pulling out text from the video, which becomes searchable (thanks cielo24!), there is no metadata layer of game dynamics or educational assessment. Even the iTunes star ratings have little objective basis. Inevitably, parents and teachers resort to the App Store’s download rankings, but, as Balefire Labs notes, this is vulnerable to incent mobile marketing or inter-publisher juice from having another successful app (see Toca Boca or Movile). Balefire Labs has built an open and objective, evidence-based evaluation covering 4,100 educational apps, but this is not native to the App Store and requires parents be sufficiently motivated in their children’s education to seek it out. (And concerns over the educational outcomes of such games is nothing new as seen in this great 2004 article).
- CAC Exceeds LTV: Despite the limitations of App Store search, Apple can at least offer new game developers 800,000 users with credit cards; a tantalizing proposition, though one that comes at a steep cost. Yes, the App Store’s commission is steep at 30%, however, the true costs arise in that (1) in education, traffic from Apple features or external PR does not convert to consumer payments or generate sufficient viral spread, (2) paid traffic (i.e. Facebook mobile ads) is too costly at around $4 CPI and even at that price, its hard to generate a material number of installs (without driving it even higher), and (3) it can be very hard to get a meaningful k-Factor in Free to Play games even from more qualified users (Facebook has continued to reduce free social posts along with other legacy Farmville and Candy Crush strategies). Ultimately, many gaming start-ups find that Customer Acquisition Costs (CAC) exceed the Lifetime Value (LTV) of those customers, an untenable situation.
A July 2014 blog post cites the story of Pixowl, which considered entering the school district sales markets with its Sandbox EDU, but then begged off, content to live off its $100,000 in consumer App Store revenue (that doesn’t even pay rent in SF!). If our post does not lead them to re-evaluate just how sustainable these consumer sales are, they might want to look at another example like DragonBox, a previously high ranking (in downloads terms) algebra game from Scandinavia. In revenue terms, they were stuck in the low six figures (still heroic in education), so they slapped on some school district sales staff; but without pursuing this as a primary focus, institutional revenue was even more limited (and with diverted resources, they have now slipped out of the App Store’s top 100 rankings). Even with sufficient venture funding to try and grow out of an CAC-LTV imbalance, any scaled publisher would immediately balk at paying the 30% commission to Apple. And so, successful players like JumpStart drive their players to browser games and subscriptions, paying a far smaller commission to PayPal. For school district driven models, students can download a free .edu version of an app and authenticate via a code from the publisher’s site, which eliminates the commission, but also removes all such sales from the iTunes charts.
- Technology Limitations: Addressing the educational needs of STEAM and 21st Century Learning and making a child an active participant in such is very different from most free games today (i.e., the CoolMath-Games model). But, this also ratchets up the level of complexity and development and cost – when LTVs for education games are already anemic. MMOG has proven success in consumer games (think Clash of Clans) and could uniquely power P2P learning, but apps are particularly power-hungry and MMOG-based models consequently untenable for non-monetizing educational gaming apps.
What Sells (at least sort of)?
Like a walk down a grocery store aisle, a quick review of the App Store’s Top Grossing education apps (also see on App Annie) helps to illuminate what kinds of models are selling (though, not necessarily in great volume). While the individual rankings vary, the Top 10 can generally be classified into four categories:
- The Big Cheese – launched in 2007 by Age of Learning, led by the founder of Neopets ($160 million exit to Viacom), this Southern California early education company launched the ABCmouse app in Q4 of 2014, but draws upon its broad consumer reach in terms of instructional books, songs, activities, puzzles, as well as games through the browser-based, subscription site ABCmouse.com (with direct response, day-time television ads figuring into their marketing strategy).
- Brain Training – Lumosity and Elevate (which grew out of Mindsnacks) are app-only and like their Puzzler peers, tap into adults’ passion for Suduko, while focusing on higher order skills, like critical thinking and problem solving (also look at Serious Games). In my humble estimation, these are probably the only VC-backed education apps that will generate much return for investors. That other adult-centric app Duolingo (language learning) also ranks highly (though on the list free education apps) and I suppose once could bring up the Craftsy app here (but I won’t). In all of these apps, its important to note they represent informal learning, rather than more formal education, and are targeted to and paid for by the same person (and an adult at that).
- Content Curation – Playkids (from Movile of Brazil) and Reading Rainbow (its back!) are also app-only (aside from some old Levar Burton videos), but like a pre-House of Cards Netflix, they largely curate, rather than produce, their kids content.
- Start-up of the Month – typically, positions 5 and below are filled with fresh-eyed game app developer with brand new budgets to pump up Facebook mobile ad buys trying to make this whole App Store thing work.
It is also telling which educational games are not featured on this list. Yes, there are the various popular free education games and apps like Duolingo (mentioned above) or the latest free games from Toca Boca, but the above list also misses the separately classified and monetizing models of leading educational gaming publishers BrainPop and JumpStart.
By way of background, JumpStart, aka Knowledge Adventure, dates back to 1991 through Davidson & Associates (publisher of Math Blaster) and its eventual agglomeration into Vivendi. Spun back out in 2004, they partnered with DreamWorks Animation in 2012 to build educational gaming apps off its film characters (also taking over a failing Neopets from Viacom in 2014). And while their School of Dragons is billed as the top selling STEM-focused MMOG, it does not appear anywhere above because (1) they do not classify their titles under “Education” and (2) like ABC Mouse, their free apps drive traffic to their browser and its subscription accounts (see School of Dragons / Memberships or MathBlaster / Memberships). Moreover, MathBlaster and their eponymous JumpStart games primarily monetize via school district sales (which merits a separate post unto itself).
Moving Beyond the App Store
So does this mean the plucky Pixowl or Duck Duck Moose should just fold up? Of course not, but I would advocate considering the following simple strategies to balance the above challenges of consumer-led educational gaming apps:
- School Districts-Led Sales: We greatly appreciated speaking with Lee Wilson, CEO of Filament Games, in drafting this post. Filament has built its business on school district sales and, perhaps not coincidentally, is a bootstrapped, profitable Midwestern company (On Wisconsin!). They have seen particular success in packaging their games with companion curriculum materials, such that they can actually sell supplemental course materials (just with an accompanying app).
- Browser-based Games: B2B doesn’t just refer to institutional sales, it can also mean “Back to Browser”! The browser uniquely enables subscription based models (see ABCmouse, JumpStart) as well as ad-supported models (see the notoriously fascinating case study of CoolMath, which largely acts as a surreptitious funnel to its ad-based entertainment gaming site). To be fair, the browser does introduce compatibility and support issues with three to four different browsers as well as multiple versions of those browsers (IE 8 still has 16% market share!). Moreover, Chrome recently dropped support for Unity Web Player and the latest version of Unity not requiring a plug-in is still not ready. However, developers need to take Design Principles into account. In addition to the revenue benefits cited above (subscriptions and ad), the browser also enables another seemingly archaic move to the keyboard and mouse! This is both relevant in terms of pedagogy as well as market reality – iPad school district sales have slowed and Chromebooks are now outselling tablets amongst schools. And besides, Tynker manages to teach coding via its browser-based model and Open Web standards well enough to win this year’s ASU GSV Summit Return on Education award.
- Other Markets: While this post was intentionally iOS-centric, many game producers are also on Google Play, but we have found traction to be even more limited there (though Google is preparing a Play for Education store with a team of teachers helping to vet apps for educational quality). Of course, Fuhu is “Leapfrogging” legacy devices with parents (they must be if they’ve already created an Investor Relations webpage!), but again, whats really getting traction in schools is Chromebooks. Interestingly, Filament Games shared with us that they have actually had the most success in putting their games on STEAM which has 60 million users as well as room and need for more educational content. And finally, developers should never limit their thinking to just platforms – there are also geographic markets to consider. Indeed, Locomotive Labs recently raised $4 million to expand in Asia (China already accounts for 330,000 of its 1.1 million downloads). East Asia is the only market in the world where parents spend more on their children’s education than the state – indeed this may be why it is rapidly emerging as the world’s EdTech laboratory.