Why managing diabetes feels like a second job
People with Type 2 diabetes mostly know what to do. They stop anyway. I studied why, and designed a daily assistant that keeps people going without asking for discipline.
Same person, same day, two apps
Watch a day play out, hour by hour. Then switch to Sugar Crush.
A typical tracker hands people a to-do list for their own body.
Two things to do next. That's it.
Pick a screen, or let it play.
just after lunch. no plan, no energy to make one.
Most know exactly what to do. Most still can't keep doing it.
The problem isn't knowledge. It's burnout.
A vague brief, and a long way down to the real problem
Managing diabetes is basically eating right and exercising. So why can't people keep doing it?
Here's how we got from that brief to a product. Tap a tier.
Explore
We picked healthcare from three domains, narrowed it to Type 2 diabetes, and went to patients' homes to watch how management really happens.
We picked the wrong problem first
54% said they had no time. So we kept asking why.
A strong start, then the sharpest decline.
They didn't know how to adapt Indian meals.
Variety kept people going for longer.
Logging is one more task on a long list.
Six apps that track behaviour. None that sustain it.
My read of the apps I audited. Tap a dot.
Sugar Crush
The white space: guidance first, with clinical depth.
they track behaviour. they don't sustain it.
From tracking better, to keeping going
Then three things I built, and removed. Each one quietly brought the obligation back.
Removed
KeptEarly versions let people tick off Up Next actions as complete.
In testing, unticked boxes at the end of the day produced guilt. The feature meant to motivate did the opposite.
I removed the checkboxes. Up Next became suggestions, not a to-do list. Nothing is recorded, and I cut back the content so the actions lead.
Version 1 did everything. That was the problem.
A textbook process picked an all-in-one app with rewards, leaderboards and check-offs.
we ranked motivation last, at zero. research later showed it was the whole problem.
“Cheat meals!!! New reward unlocked”
Compete to make healthier choices
Track today's tasks and goals
Meal plans after each consultation
Learn about diabetes daily
Connect when demotivated
Discounts on medicine and supplies
Everything in one place
Five tempting features. Five no's.
Why it was temptingA strong pattern that encourages consistency.
Why I said noPunishes inconsistency. Missing one day often leads to quitting, not recovering.
Why it was temptingRemoves decision fatigue entirely.
Why I said noCreates dependency, not capability. Logistics-heavy, thin margins. The goal is food literacy.
Why it was temptingClinically important, and in most apps.
Why I said noMandatory logging before a habit forms increases drop-off. Deferred until trust is built.
Why it was temptingStandard in wellness apps, and precise.
Why I said noAssumes motivation already exists. High effort, low long-term retention.
Why it was temptingDirectly addresses the emotional burden.
Why I said noLow adoption here, and doesn't fit everyday life. Reassurance is built into the guidance instead.
What I validated, and what needs shipping to know
Academic project · not shipped
- Across 35+ interviews, nobody said they needed more information.
- The daily-plan idea brought relief, not excitement.
- Estimation over perfection landed with people who had quit trackers.
- Whether autonomy-first design keeps more people on day 30.
Behaviour lasts when it feels chosen, achievable and supported.
People who quit tell you exactly where the system failed.
Drop-off here isn't churn. It's someone's body.