❓ Problem

Children are surrounded by technology and understand almost none of it. US tweens now clock 5 hours 33 minutes of entertainment screen time a day, and 50.4% of teens aged 12 to 17 are on a device four or more hours daily. Every minute of that is spent on the consumption side of a glass rectangle nobody has ever opened. The device is magic, and magic is the opposite of understanding.

Parents have noticed, and they are voting with their wallets. Tonies did $731m in revenue in 2025 selling a screen-free audio box for kids, and Yoto did $128m in 2024. Both are pouring their growth into North America. The willingness to pay for "hardware instead of screens" is proven and enormous.

But here is the part the STEM aisle keeps getting wrong. The kits demonstrate a concept and then die. You build the volcano, you build the catapult, you build the little solar buggy, and by Tuesday it is under the bed with the other $40 of injection-moulded regret. Nothing produced by a STEM kit ever earns a place in the house. It is a lesson dressed up as a toy, and children can smell the difference instantly.

The alternative is worse. A parent who actually wants to build something real with their kid has to pick components, cross-reference three tutorials, order from four suppliers, and then debug a stranger's abandoned GitHub repo from 2019 while a seven-year-old asks every ninety seconds whether it works yet. Saturday afternoon becomes unpaid IT support, and the whole thing ends in a drawer of loose jumper wires.

Here's the idea.

✅ Solution

A monthly subscription that ships a family everything needed to build one genuinely useful piece of technology together, and then keep using it.

• Every box makes something with a job in the house. A screen-free WhatsApp messenger so the kids can talk to their grandparents. A bedroom weather station. A plant that texts when it needs water. A door sensor that tells you the dog got out. A secret-code radio between two bedrooms. Not a demo, a device.
• The build teaches as it goes. Each component is explained at the moment it goes in, so the child learns what a sensor, a microcontroller and a power rail actually do rather than reading about them. Three build paths ship in every box: one for younger kids, one for older kids, and one for the parent who will inevitably start "improving" everything at 11pm.
• Hardware and support are paid, code is free and open source. Every project lives in a public repo. Families remix builds, reuse components from previous months, and publish their own modifications back. The community becomes the content engine and the reason nobody cancels.
• The parts compound. Month three reuses the brain from month one. That is cheaper for you, cheaper for them, and it quietly teaches the most important lesson in engineering, which is that this stuff is made of reusable pieces rather than magic.

The wedge is deliberately narrow: one flagship build that a family will still be using thirty days later. Everything else is a distribution and retention problem.

📊 Key Numbers

Market size

• Analyst estimates for the global STEM toys market range from $4.67bn in 2026 to $13.5bn, depending on how tightly the category is drawn. Treat the number as directional, not gospel.
• The wider educational toys market sits at roughly $78bn in 2026, with STEM the fastest-growing product segment at a 9.92% CAGR.
• The subscription box market reached $42.5bn in 2025 and is forecast to hit $124bn by 2034.
• Around 58% of parents say they prefer educational toys to conventional ones, and subscription learning kits are the fastest-growing sub-channel at roughly 31% recurring adoption growth.

Unit economics