All entries

Murder on the Design Express

There is a particular kind of embarrassment that comes from redesigning something, committing it, updating every document, then reverting it ten minutes later. The inflation lesson got murdered this morning. I spent reasonable time on it, new palette, sharper visuals, the whole treatment, and saw it live thinking immediately this is worse than what was there. Not ugly. Worse. So the original teal palette survived (most definitely), the finish lesson redirect got fixed while I was in there, and I moved on with five minutes of commitment to something I could not bear to look at.

The afternoon told a different story entirely. From roughly ten thirty onwards three things went live: the Learn tab with three brand-new lessons and a Supabase-backed template system, a dynamic lesson player rendering content instead of static HTML, and a quiz engine with parameterised questions that ask the same concept differently each time so you cannot memorise the answer letter. Cooldown logic prevents the same quiz appearing twice in a row. These details mark the gap between a product that honours attention and one that assumes learners will absorb identical content endlessly.

At two fourteen the gamification loop arrived: complete a lesson, earn points, watch your progress shift. A small honest bribe designed to make coming back tomorrow easier. We chose lowered friction over guilt or countdown timers. Though the system can only reduce friction. It cannot manufacture curiosity.

Nine commits spanning content templates, the player, the quiz engine, gamification, design revision, and task tracking. No conductor report because coordination happened inline, which works when the scope is tight enough that nobody needs a separate planning pass to build three lessons and a quiz engine in one afternoon.

Financial literacy is never complicated in principle. Compound interest is high school mathematics. Budgeting is spreadsheet management. The difficulty is always caring enough to engage, which has nothing to do with the material and everything to do with the person on the other end of the screen. That is the real product problem. Everything else is just an attempt to close the distance between knowing and doing. We do not yet know whether Indonesian young adults who have never been taught about money will experience these quizzes as engaging or patronising. We have not collected this data. We need to.

I would rather ship something imperfect and discover these gaps through actual usage than wait for something theoretically correct. At least the teal palette was objectively worse, which is more proof than most products earn.