We Built an App, It Fell Short, and Scrapping It Was the Best Decision We Made

I want to tell you about a version of our platform that no longer exists.

Early on, we built what most people would expect. An LMS for the boys with XP points, levels, streaks, badges — the full gamification playbook that almost every education product reaches for because it looks like what kids are supposed to want. We had curriculum blocks. We had a progress dashboard. On the surface it looked like a real product.

It wasn’t right.

And the hardest part was not admitting that. It was figuring out why, because everything on the checklist had been done correctly.

The problem was not the build. It was the order we built in.

We designed the technology before we had properly understood the teaching. We knew what an LMS was supposed to look like because we had seen hundreds of them. What we did not yet know, in tested detail, was what actually moved these two specific children from confusion to mastery.

So the app was polished and empty at the same time.

Streaks reward showing up, not understanding.

XP measures activity, not depth.

None of it was dishonest. It just was not true to how learning actually happens for a child who needs three days of quiet processing, or a child who needs several different entry points into the same idea before it sticks.

Here is the principle we landed on, and we now treat it as almost sacred:

Compulsion mechanics — streaks, points, badges — retain adults. They backfire on children.

An adult chasing a streak is disciplining themselves toward a goal they chose.

A child chasing a streak is being trained to perform compliance. And compliance is the opposite of the curiosity we were trying to protect.

What we did instead

We stopped building the app and went back to the kitchen table. Back to watching what actually happened when two real children sat down with real material.

What we found became the real architecture of everything that came after. We call it the Daily Five. It is not a feature or a worksheet template. It is a simple, repeating structure of five question types that, after a lot of trial and error, turned out to be the only set that reliably told us whether understanding had actually happened.

We use it every day, on every topic. The five questions are:

Define — say what the concept is, in your own words, with nothing in front of you.
Explain — say why it works or why it matters.
Solve — use it on a real problem.
Research — go and find one real-world connection.
Life Principle — connect what you learned to how you want to live or treat people. This one is never marked for correctness.

Two small rules make the whole thing work:

Say the answer out loud before you write it down.
Mark it yourself the same day, while the thinking is still warm.

That is the entire engine. Five questions. Two rules. No streaks. No points. No badges.

Then we rebuilt — but only after the teaching was clear

Only once this structure had been tested, broken, fixed, and tested again did we return to the technology. What came out the other side looked nothing like the gamified LMS we had scrapped. It became smaller and sharper: a simple four-page case-study format built around the Daily Five, plus two focused tools that each do one job well.

Good teaching architecture does not need more features.

It needs fewer, truer ones.

If you are building anything for your own children right now

Whether it is an app, a spreadsheet, a routine, or just the shape of your week — resist the pull toward features before you have done the unglamorous work of actually watching your child learn.

Write down what you observe.

Test it against what really happens, not what you assumed would happen.

Be willing to scrap the thing that looked impressive on the outside if it is not true on the inside.

The child is the constant.

Everything else — including the system you were proud of building — is the variable. It should bend. If it will not bend, it is not finished yet, no matter how finished it looks.

You are allowed to throw away the version that only looked good.

Many of us have had to. The ones who end up with something that actually works are usually the ones who were willing to start again from the kitchen table.

Frequently Asked Questions

Why did you scrap the app? It had XP points, levels, streaks and badges, and on the surface it looked like a real product, but it was not right. We had built the technology before we understood the teaching. Streaks reward showing up, not understanding, and XP measures activity, not depth.

Why don't streaks and points work for children? Streaks, points and badges keep adults coming back, but they backfire on children. An adult chasing a streak is working toward a goal they chose. A child chasing a streak is being trained to perform compliance, which is the opposite of curiosity.

What is the Daily Five? It is a simple, repeating structure of five question types that we use every day, on every topic: Define, Explain, Solve, Research and Life Principle. Two rules go with it: say the answer out loud before you write it down, and mark it yourself the same day.