Here’s what we’ve learnt in 8 years of developing an educational app, Levebee

The author of this article is a co-founder of Levebee and has been developing it since 2014. Levebee has already helped more than 300 000 children acquire essential literacy skills such as reading, writing, counting and foreign languages.

2021 was a turning point for Levebee. We succeeded in creating a maths section which was a unique opportunity for us to draw upon the knowledge we had been acquiring for 8 years while teaching reading. At the same time, we participated in the IMPACT EdTech accelerator programme which enabled us to discover other educational apps as well as familiarise ourselves with the structure of educational systems in other countries. I was quite surprised to find out that in comparison to other countries, very few apps are developed in Czechia. Especially since education is a common topic of conversation here and we have many fantastic teachers and developers! In this article, I would like to give you an inside look into the creation of an educational app and share a sneak peek at some interesting challenges one might encounter on such a journey.

What has worked for us

First, let’s talk about the positives. I’ll briefly explain what turned out to be crucial.

Dividing the curriculum into the smallest possible units

From the very beginning, we knew the importance of reducing cognitive load and the limits of working memory while learning. Of course, Levebee’s program of learning reading incorporates those theories. From the methodological perspective, we’ve covered everything. As a result, the app is extremely effective for children with reading difficulties as well.

What could have been done better in the reading section of Levebee was the user experience. Technically, it’s possible to learn in small steps but users have to customise the exercise settings themselves. That, in turn, has proved to be problematic as many teachers simply don’t notice the settings. We’ve tried highlighting the settings button, automatically opening the settings upon starting each exercise, and emphasising that information in the main menu. In the end, the only guaranteed solution was organising workshops for teachers and explaining it in person. That, however, significantly slowed down the adoption process.  In the maths section, we got rid of those issues by turning even the most basic mental skills into separate exercises. Hundreds of exercises in the app might seem like a high cognitive load (after all, nobody can remember hundreds of exercises) but in reality, it improves the usability of the app. If users go through exercises one by one (101, 102, 103, etc), they can’t go wrong and they won’t miss any content.

The interactivity enables instant feedback and saves teachers’ time

It’s not enough to just do something in order to learn it. Receiving constant feedback, ideally immediately after completing an activity, is essential. The shorter the feedback loop, the faster students improve. Digital technologies are superior to paper tools in this regard, provided they’re used properly.

Unfortunately, “digital textbooks” are often just glamorised PDF files comprising mere A/B/C exercises, which were created with the same mindset as paper books. There might as well be questions written by hand and answers marked right and wrong. The peak is a QR code which redirects the reader to a video.We approached Levebee completely differently. Exercises in the app are generated randomly (within the methodological parameters of a given exercise). We anticipate possible mistakes in advance and address them (for example, that number 13 consists of 10 and 3, not 1 and 3). This has several advantages:

  1. Children will never run out of exercises.
  2. Students receive feedback instantly and “discover” the correct answers themselves.
  3. Even if children with learning difficulties need a bit more time, they eventually master the topic just like the top students in their class and feel equally successful, thus reducing educational inequality.
  4. While students work independently, teachers can devote more time to helping other children.

It’s generally difficult to convince teachers that apps are better than paper books and those are very compelling arguments.

The entertainment aspect lies in the methodology and can be easily spoiled by poor accessibility

“Everything that’s on a screen is fun for children” is a fairly common misconception among parents, teachers and creators of educational applications. Defining ‘fun’ is a bit more complicated than that.

In the nutshell, the “children like what they’re good at” type of approach has proven to be the most effective. If kids are bad at maths, they won’t like it, nor will they enjoy practising it. If we help children improve their maths skills, they will start to enjoy it and have a positive attitude towards it. The key is to find the perfect difficulty and ensure that the given exercise is neither too easy nor too difficult (so-called concept of flow). This all depends on the methodology. Take the above-mentioned concept of dividing the curriculum into small units for example, or combining the objective and subjective evaluation (we simply ask the children whether the exercises are difficult or easy for them). If the methodology is not correct, no technical or graphic design will help. In Levebee, any potential oversight was nipped in the bud by our brilliant methodologist, PhDr. Renata Wolfova. As far as I can tell, however, it’s often a stumbling block for many creators of educational apps.

The fun aspect of learning can also be marred by even the smallest issues with accessibility, especially among children with special needs. There’s no reason why digital materials can’t surpass paper books in this regard. In fact, they do. It’s not cost-efficient for publishers to print materials incorporating exercises that would address combinations of various special educational needs (that is why there are so few reading assignments dedicated to children with dyslexia).

Levebee allows you to change the font style and size (e.g. Dyslexic Font), the background colours, highlight syllables, listen to audio instructions and change the language of those instructions for children with a different mother tongue, etc. Taking everything into consideration, it was a worthwhile investment and it has certainly paid off. Those functionalities filled children, who are usually overlooked by the creators of other apps, with enthusiasm.

The term “gamification” has been gaining popularity. Rewarding users with points, badges, levels, etc. can help make spaced repetition more fun. From the pedagogical point of view, it’s sometimes helpful to revise the same topic over and over with some breaks in between. That’s when the gamification comes in. It makes students excited about otherwise tedious activities and so, it was implemented in Levebee.

Children collect points which they later use to buy animals in a ZOO, etc. It certainly works; children absolutely love it. It does, however, have its downsides which I’ll tell you about below.

What to look out for

In 2014, when we started our journey, there was a fairly popular notion in startup circles that tablets would completely replace paper textbooks within the next 10 years. As we know today, it hasn’t happened. I will briefly share my, purely subjective, assessment of why the technology hasn’t (yet) met those high expectations. If you have some ideas on how to solve those problems or if you know some international sources that could be an inspiration, I’d love to join the discussion!

Reliability (paper is reliable)

Our biggest competitor is, obviously, paper. Although digital technologies have many advantages, they will never surpass paper materials in one key aspect, namely reliability. They can only come close. Paper never runs out of charge, doesn’t get broken by the latest updates, doesn’t depend on the quality of your internet connection, and doesn’t require anyone to be trained in order to use it.

A big part of the problem is connected with the hardware which is something we at Levebee cannot influence. Schools either don’t have the hardware (in comparison with other neighbouring countries, the Czech Republic is actually doing relatively well,  although we’re nowhere near Scandinavian countries where there’s one computer per child), or they can’t keep it functional. Tablets are often uncharged, and the chargers are either lost, broken or left somewhere at teachers’ homes. All of this can be solved, but that means that the pros of an educational app must outweigh the above-mentioned issues. Otherwise, digital technologies can’t gain upper hand over paper. 

Software issues are already partially under our control. We know how to solve some but we still need a solution for others. For instance, it is difficult to give all students fast and secure access to an app in the classroom in order to save their progress. If the teacher wants to use multiple apps during a lesson, just logging in all children takes an excessive amount of time; especially if the students can’t read and write yet. In part, this problem can be solved by giving each child a personal computer as it is done in Denmark. Another solution could be a single sign-on provider (e.g., like the one directly provided by the Ministry of Education in Israel, or a private company Skolon, in Sweden). We are currently testing PINs, and we’re planning to try out QR codes, but so far, it hasn’t led us to a satisfactory result.What is, obviously, in our hands is the reliability of our application as such. We’ve learnt that teachers are very sensitive not only to any mistakes or errors in the app, but also to rapid changes. No wonder. Unlike an office application, Levebee is used publicly in front of many people. Any oversight looks like incompetence on the part of teachers and may undermine their authority. For that reason, we carefully test and explain everything to avoid any unpleasant surprises.

Mistaking fun for learning

The fact that children spend a lot of time and enjoy using an app, which at first glance looks educational, doesn’t mean anything. In order to learn something, one has to think it through and reflect on it (more on this can be found in e.g. Seven Myths About Education by Daisy Christodoulou). If a child spends 80% of the time in an app thinking about which outfit to buy as a reward, it’s not a very productive activity. On the other hand, it’s incredibly difficult to create an educational app which teaches absolutely nothing. Children naturally learn from everything they do. Making an app for learning English that wouldn’t teach a single word is borderline impossible. Almost every application can show a study proving that children have improved after using it. What’s important is how big the improvement is, whether it wouldn’t actually be more beneficial for children to spend time on something else (see Visible Learning by John Hattie) and if students can apply the knowledge to tasks outside the application. This concept is otherwise known as transfer learning. That is, however, difficult to measure and it’s hard to verify who is actually right.

As creators of educational applications, we must not become complacent after receiving positive responses from the pilot projects, and must continuously work on improving the educational impact of our apps.

The conflict between educational and business goals

Just like everywhere else, even among educational applications various questionable practices, which are not in line with the interests of students, are used.
Over the years, I have observed a typical recurring pattern in many educational applications: creators supported by powerful investors find out after a few years that the education system can’t be changed as fast as they first anticipated, so they start looking for shortcuts to get funds (unethical pressure to purchase something, tracking ads, automatic subscriptions which customers forgot about) or at least try to show a growth in user activity that would justify another round of financing.

An example of such practices on the less harmful side, is overgamification, as in e.g. Duolingo. Collecting points and badges for many people becomes the main motivation for using the app after it’s no longer sufficiently educational.

From a pedagogical point of view, at a certain point, it would be more beneficial for a student of French to focus on conversation or some other skill outside of Duolingo, but the app still “makes” the student practise vocabulary, offering minimal improvement (see UX exit points). Taking millions of users into consideration, it translates to countless wasted hours.

The worst cases include blackmailing children emotionally. (One app, which shall remain nameless, allows children to collect points in order to feed their digital pet. If parents do not pay after a trial period, the pet dies.) Unfortunately, especially in the US, there are numerous cases of aggressive collection and monetisation of children’s data. In the EU, the pressure to protect data is much greater, and rightly so; however, it causes many problems which are beyond the scope of this article. 

Levebee is financed solely through a fair sale of licences and doesn’t contain any ads or tracking cookies. If a student stops using the app for three months, the subscription is automatically cancelled. We believe that this approach is actually more profitable in the long term on account of the trust we build.

Excessive emphasis on the graphic design

Graphics have become extremely overrated. Children perceive aesthetics differently from adults and like things that would seem off to those with a keen sense of aesthetics. Consequently, an overly complex design can defeat the purpose. The more sophisticated it is, the more distracting it can become (while children need a distraction-free environment). It also becomes obsolete sooner. The pursuit of graphics at the same level as the ones in current top games has become the death of many educational applications. It’s prohibitively expensive and it significantly slows down app development. “Pretty” educational games which were created in 2014, seem quite ridiculous today. Therefore, spending too much money on graphic elements is not a good investment. We’ve never had to change our simplistic artwork and it has never limited us in any way. 

Educational inequalities

I have to admit that this is the topic I’m currently dealing with. Even though technologies should theoretically expand access to education, it’s not always the case in practice. We could observe that, unfortunately, during the COVID-19 pandemic when many students from disadvantaged backgrounds were falling behind due to the lack of proper equipment or internet connection. 

Levebee was created precisely to reduce the inequalities in education. We can help children who struggle with basic literacy skills for various non-financial reasons: specific learning disabilities, different mother tongues, frequently changing schools, distance learning, etc. We are, however, aware of the fact that even $8 per month may be too much for many families. The last thing we would want is to create more inequalities by trying to address one (although I genuinely believe that problems which can be solved with money are among the simplest ones).

For that reason, we offer a free licence for children from socially disadvantaged backgrounds; however, identifying those children is somewhat problematic. So far, we’ve established cooperation with People in Need, META organisation (Society for Young Migrants), Help the Children fundraising campaign, Salvation Army, and Diaconia organisation. All hospitals, orphanages and shelters also automatically receive Levebee for free. Many of these organisations tutor children using their own devices but it doesn’t cover all their needs in the slightest.

What’s next?

The success of the first pilot projects has motivated us to expand internationally. Pilot testing is already underway in Germany, Poland, Sweden, Finland, Israel, Slovakia, Romania and soon in Great Britain and US. We’ll see which countries will be the most promising. From our experience, we know that if we get teachers to test Levebee actively with their students, they become very enthusiastic about the app. Therefore, the biggest challenge for the next year is to find out how to do this (with our limited budget). If you are a teacher or know teachers who might want to cooperate, we would like to get in touch with you.

Michal Hudeček
LinkedIn | Twitter

Leave a Comment

Your email address will not be published. Required fields are marked *