The only two factors that matter in language-learning tech

Everything else is just marketing noise

In the last decade, many Language learning (LL) apps and courses have been thrown into the market; among the well-established apps, I have tried courses without a teacher like Duolingo, Rosetta Stone (more on these ones later), Babbel, Busuu; vocabulary flashcard apps like Memrise; video chat apps like Tandem and audio-only chat apps like SpeakNative, and Buddytalk. I have also tried teacher-based apps like iTalki as well as some of the apps that there are for every single online language school in the world (English Central, Open English, EF, etc.) My iPhone home screen has several pages of the LL apps I have tried. Apps for the smartphone have proliferated to the point that if you want to throw in a new one, it really needs to have something special (or great marketing) to stand out from the crowd.

What actually matters

Now, what makes a LL app “better” than another one? Of course, all of them claim that their app is “the best” out there because it is “made by experts” or “it’s stress-free”, and so on. But we need more specific criteria for finding the real objective qualities in the middle of so much marketing noise. So, can we objectively find out what really matters in LL apps?

Well, LL education research has shown that there is no such thing as a “breakthrough method for learning a language in 14 days”: the single thing that matters is the time you spend immersed in the language, developing the four basic skills of LL, which are the “passive” ones (listening comprehension and reading) and the “active” ones (talking and writing). That’s all.

Now, if all that matters is spending time in the target language, either reading, listening, speaking or writing, the question becomes this: how we make a person (ourselves, for instance) spend time honing your skills in the target language?

There are several very good answers to this. Polyglots all around the world, like Lydia Machova, Benny Lewis, Steve Kauffman, John Fotheringham, etc., propose useful principles for effective time-spending in LL, like making LL a habit, (I’m not an expert in making habits, but there are even books for that), making LL fun (more on this later), thinking directly in the target language instead of translating from your native language, and so on. You can get a long way by hacking your environment, doing simple things like watching movies in the intended language, eventually with subtitles in that language, using coffee-break time, hearing podcasts, and much, much more. FluentU offers a free immersion guide with this kind of advice.

But the topic I’m discussing here is more related to the LL apps themselves than to your life-hacking.

My take is that there are only two factors by which tech can support successfully spending time in LL:

1.- Convenience

2.- Engagement

Convenience is about friction reduction in the experience of LL. For instance, if you have to commute an hour to get to the Language School and another hour back, then there is friction because the commute is a drag that we don’t want. If an app is too difficult to use, that’s friction.

Another evident convenience aspect is cost: it a course is very expensive (like Rosetta Stone), frankly, it couldn’t be convenient except for the wealthiest.

Yet another convenience item is flexibility on when to do LL: Traditional language schools’ fixed schedules could be an obstacle for some people. Even having to schedule in advance could be friction.

Privacy is an underrated aspect of convenience. Video-based lessons, so popular in today’s internet language schools (OpenEnglish, English Central, etc.) force you to show not only your face but also your place, your hair, your clothes, the people around, etc., which could be less than desirable. You couldn’t show up in pajamas or in underwear, for instance. See below how this relates to the use of VR.

Engagement is the other part of the equation besides convenience. Engagement is that you feel motivated for learning languages. It means, in short, that the person is having fun while learning the language. I’m not inventing this: it has been repeatedly found by researchers and by the polyglots mentioned above. But to make the language learner feel engaged is easier said than done.

Of course, the content presented to the user could promote (or demote) engagement; old methods used to present lessons of the style “Mary goes to school”, while good current content contains compelling stories like one of those biological brothers who found each other many years after they were adopted by different families. But from the technological side, there are elements that reportedly tend to generate engagement, like instant feedback, immersion (like in VR), and gamification. Videogames have instant feedback, have immersion and have gamification, no wonder they are so damn addictive.

In short, the more convenient and engaging a given app is, the more useful it is to promote your LL time spending.

I made the exercise of organizing some available apps and methods in a two-dimensional chart, as follows:

I graded each category from 1 (low) to 3 (high), and for convenience calculated the sum of flexibility, privacy and the inverse of cost (affordability). Then I adjusted engagement (x6) to give more similar values to the other axis. The farther a point is from the origin, the “better” it is.

Online lessons appear on top of the vertical axis with VR rooms, but the latter fares a lot better on convenience, assuming we don’t consider the cost of the VR equipment itself, which in the case or VR on the smartphone is a valid assumption because all of us have one.

Disclaimer: I’m the CEO of Avalinguo startup, an app now in closed beta, where users appear represented as avatars in a virtual room, using just the cellphone.

Books, of course, get punished on the engagement side, even though they are very convenient (cheap, private, flexible about the time of use). Peer to peer video exchange is fun and cheap (even free), but it gets punished on privacy, which is part of convenience.

There you have it. This graph is only my evaluation, and evidently a sound study would have to gather the opinions of many users and use a solid research methodology.

What do you think? Please comment below.