Whatever you do, please don’t start another EdTech company. The L&D and education industries don’t need another software startup trying to solve old problems with fancy new technology. What organizations need instead is a more technological focus. That is, they need more coders on staff to collaborate with non-coders and to build custom solutions in-house, instead of outsourcing to generic, off-the-shelf third-party solutions. Some third-party solutions help organizations solve problems, save money, work more efficiently, increase revenues, or improve outcomes, but many do not.

Alien footwear: A solution in search of a problem

To understand why, imagine if a space alien invented some expensive but amazing new alien footwear, and then came to earth looking for species that have feet the footwear might fit. Odds are exceedingly slim these alien shoes would be designed for earth conditions and work on human feet, much less fit perfectly on any particular human’s foot. Any rational person, after recovering from the shock of encountering a space alien trying to sell them shoes, would realize the shoes don’t fit and pass on buying them. Similarly, EdTech software designed by people outside your company—and often outside your industry—won’t fit your company’s needs perfectly either.

Too often, programmers with limited industry experience rush to launch a company by building a software “solution” that uses the hottest emerging technology such as AI, AR, VR, IoT, or blockchain. They search for problems their software solution could theoretically solve, find a niche with 26 identical competitors with forgettable features, raise gobs of investment capital, quickly build a glitchy and poorly designed beta version, and launch … thus becoming forgettable competitor number 27. They heavily promote this dubiously useful software, vastly overselling its value. If they manage to keep raising money and stay afloat, they may somewhat improve the initial product over time, but the new versions introduce still more glitches and unforeseen problems and still don’t solve the initial problem entirely. Most potential users wisely ignore the software. A few buy it and are quickly disappointed, joining the ranks of cynics who ignore the next “transformative” EdTech solutions that come along.

Most products take years, decades even, before they’re refined enough to consistently create value and a positive experience for users. It’s hard to imagine creators of other mediums operating this way. Authors submit their books (and columns!) to editors for improvement before publishing, and musicians work with producers to refine their songs, but tech companies are forever releasing first drafts of software products.

I don’t blame programmers for wanting to start an EdTech software company. Programmers, like so many other people (including me), want to use their skills to build innovative and successful companies. It’s an admirable urge. However, when you have a hammer, everything starts to look like a nail. When you’re good at writing code, code becomes your hammer. “Solve it with software!” becomes the mantra, and every problem looks like a nail you can hit with your software “hammer.”

Don’t do it like this

I attended an AR event last week and met the founder of an EdTech startup. Her company creates QR codes that open up AR learning experiences for students. These QR codes can be included in textbooks and training manuals, with the eventual goal of replacing textbooks altogether. She has big clients, but I question whether this software creates any value for anyone. QR codes were invented in Japan in 1994–nearly a quarter century ago–but have you ever used a QR code? Do you know anyone who has? Do you even know how? Could you figure it out in less than five minutes? Would you assume it’ll work smoothly every time, especially when combined with something as complex as AR? Do you even attempt to use a QR code when you see one? My answer to all those questions, and the answer for most people, is no.

This same logic applies to virtually every new technology. The technology gets overhyped, a few companies in specialized industries put it to good use, and everyone else waits until the technology advances to the point where it’s seamless, reliable, and ubiquitous.

Do it like this instead

Fortunately, there’s a better way: build your own software. Rather than starting with a technology solution in search of a problem, a better approach is for an organization to start by identifying a problem, look for the best technologies to solve that problem, and then build a custom solution in-house, specifically tailored to the organization’s unique needs.

The architecture industry serves as a good case study. Plenty of uninspired third-party VR software exists in this space, mostly in the form of Unity plug-ins. However, a few forward-thinking architecture firms like Gensler are taking the build-it-yourself approach. They’re not looking for off-the-shelf VR and AR software to support their architectural design work, because no outside company can fully grasp the internal company workflows and industry nuances. Instead, they’re hiring experienced developers in-house to build exactly the immersive experiences they need. L&D departments and educational institutions would benefit from adopting a similar approach.

In order for larger organizations to create software in-house, they need to operate more like startups. Like Google does with its various divisions (Android, Google X, ARCore, Daydream, etc.), organizations need to be able to iterate, pivot, and adapt to rapidly changing needs without slowing down the product development process with approval required at every step from eight levels of bureaucracy. Talented technologists won't work for companies that stifle creativity, offer limited opportunities for personal growth, and don’t provide a dynamic work environment. Part of the allure of working for a startup is having the ability to make changes and receive immediate gratification from seeing the product updates. Assuming your organization can overcome these obstacles, you’ll be able to attract the talent you need to build valuable software solutions in-house.

In summary…

I’m no neo-Luddite. I do write a column about emerging technologies, after all. That said, I urge any aspiring EdTech entrepreneurs out there to think critically about your customers’ needs and whether your proposed solution solves any real problem. Conversely, if you and your organization do decide to build solutions in-house by leaning on your deep industry experience, you just might end up creating powerful software that many others would find valuable too. That kind of EdTech software would definitely be worth building.

Additional resources

I’m not the only grumpy old fogey shouting “Get off my lawn!” into the internet void. I’ve recently encountered a number of articles with a similarly cynical outlook toward software startups in general, not just those in EdTech. Here are a few articles I’ve enjoyed, plus another that I wrote.