The vast mass of entrepreneurial energy at last week’s Web Summit in Lisbon was distraction embodied: Robots, keynote speakers and spotlit stalls clamoured for the attention of the thousands of attendees who were otherwise glued to their screens by the all-knowing conference app. Despite the competition, chance happened to bring our small Berlin crew into contact with a very interesting education tech company.
On the final day, as we sought to reach the afterparty for a well-earned wine, we ran into two gentlemen based in Hong Kong, Max and John.
Our colossal shared wait in line for metro tickets gave us the chance to learn about Zapzapmath, the gaming ecosystem they founded to help boost kids’ ailing maths comprehension worldwide.
Max Teh — a venture capitalist and Zapzapmath cofounder— later gave me a glimpse behind the scenes.
What was the moment that made you and your co-founders stop and say, “There’s a problem, but hey, we might have a solution”?
John was a math teacher, and he stumbled across a problem about ten years ago: In schools, all kids, teachers and parents want to see improvements in math. It requires very little memorization, just practice, but kids generally veer away because there are more interesting things to do. So, about four years ago John came to me with this pain point, as an ex-maths teacher wanting to make apps for maths games.
My pain point was a little different — I’d been reading lot about the declining standards of math in the world over the last 20 years, with the analysis saying that kids in the K6 levels — 4 to 11 years-old — are exposed to a lot more distractions now that they have access to things like computers, smartphones and tablets. Parents don’t know what to do with this, and there’s a lack of educational games. So we have a decline in maths, and maths is very important, no matter if you’re a venture capitalist, in tech, a banker or even a lawyer. So I said this a pain point we need to solve, we need something to tackle this decline in math standards. I was thinking along the lines of something to engage kids to solve this social problem, he was thinking about solving boredom.
And then, bam! Gamification of math lessons!
So how can gamification really be used in the education sphere?
Gamification was the right thing to do because kids love games and we wanted to put learning back into games. But I said we can’t just create apps with games, we also have to keep the parents involved, and the back end dashboard we created allows this, with the database telling parents every time their kids spend an hour on Zapzapmath practising a particular topic. It then brings in the maths teachers by giving them the database so they know exactly what each kid has done and what they get wrong or right, so that they can help the students.
To make it really useful for kids, parents and teachers, we decided the math lessons we gamified must correlate to what schools teach the kids. So we designed and developed a planetary system comprising more than 180 maths games to provide comprehensive math content which is compliant with many international math standards in the world. It can match with a school syllabus very quickly, compared to other games, which are mostly focused on certain selected subtopics or grades.
And this only began back in 2014?
Together with our other co-founder, Adam, who is behind the UI/UX design, we got together in 2014, the developers basically went into hiding to develop the MVP (Minimum Viable Product) and almost ten months later, the first beta product came out. What really shocked us and surprised us in a nice way, is that in the first few months we had a few hundred thousand downloads.
We had a close look and saw that 80% came from the West — mainly from the US — and then it hit me like a bomb, “Bam!”. It hit me that they need this and they like what we have created. We went to the US, started attending serious edtech trade shows and started engaging with schools and teachers, and we never looked back. We’ve now gone past 2,500,000 downloads, across 150 countries across the world. The US and China account for around 60%, and all this was done without any real marketing.
There is something interesting going on behind the scenes while kids are playing the games — how can this use of AI and machine learning help children learn?
So this is partly machine learning, and it’s partly adaptive learning: The machine learning uses computers to learn about each student’s behaviours and practice — and then tries to use and evaluate this to create better learning material for other students. On the other hand, adaptive learning is when we say, okay, you are playing on grade 3 and you’re having a lot of difficulties. The AI, Elf, will then quietly bring you down to the A2 level without telling you, or similarly, adaptively nudge you up to play at a higher level. It works both ways, meaning that kids avoid getting frustrated from boredom or from difficulty.
You’ve now got a huge reach; can you already see that this is having an impact?
So we launched our full product in August 2016, so it’s 15 months old, and we’re still collecting a lot of data and verifying it, so from what teachers are saying to us, the kids are actually really enjoying it. What we cannot say at this time is that the kids have improved by 20% or anything like this, but over time, we hope to be able to show something like this with empirical evidence to go with the positive commentary.
What are the next steps from here on in for Zapzapmath?
We want to continue to enhance the product to fit the requirements of as many countries as possible in terms of what schools demand — we have 70% of what US schools and teachers want already, and we want to continue to make the product even better as that’s our competitive edge. Secondly, we want to reach out to as many people as possible, through the web and other mobile platforms, to reach as many kids as possible, as there as 500 million kids in the K-6 category in the world. Only through this can we realise our vision of arresting the decline in math skills and making math fun and engaging for kids. Sadly, our third priority is to convert some of these guys into paid customers.
What have the biggest challenges been in getting to where you are now?
I think the constant feedback from teachers all over the world, in terms of them wanting things in the ecosystem and then having to deploy and tweak and pivot, which takes time to adjust. Also, three years ago there was almost no one doing what we are doing, but nowadays there are three to five other guys doing what we are doing in terms of high gamification and having comprehensive and standard compliant content. At the same time, we have to make money to survive.
Finally, many people are worried about what technologies like AI and machine learning will do to our societies — but do you think that they can instead be used as a tool to educate and prepare people for a highly-automated world?
I am in favour of AI in secondary uses, which is different to primary uses like using a robot at home, letting the robot learn from us and having to understand us fully. If you fast forward, this robot can be smarter than you and me. But on the other hand — in secondary uses — it’s meant to help the system to work in a better way for the users, so I don’t see any problem and fully support the use of AI to help us become better and faster in whatever we are doing.