Break It Through
A sixty-day guided recovery programme. One small task a day, and the morning it ends.
People fail to heal alone because they have no plan and no accountability. This gives them both.
Break It Through
A sixty-day guided recovery programme. One small task a day, and the morning it ends.
Read the full case study on Behance↗EIGHTEEN SECTIONS · THE COMPLETE DECKTHE BETPeople fail to heal alone because they have no plan and no accountability. This gives them both.
Three reasons people stop healing alone.
No plan, and no accountability
Grief has no structure. People wake up knowing they should do something and have no idea what. Motivation is the wrong thing to depend on at the worst moment of someone’s life.
Streaks punish the people who need them
Every habit app ships the same mechanic. Miss a day, lose the streak, see red. To someone grieving, a broken streak is not a nudge. It reads as proof they are failing at healing too.
A chatbot is not a clinician
The category filled with AI therapists overnight. For a person still deciding whether to ask for help at all, being handed an algorithm is the fastest way to lose them entirely.
None of this is a technology problem. All three are decided in the interface.
01
Six rules, agreed before a single screen.
The clinical team wrote six rules into the brief. I turned them into a list of patterns the product was not allowed to use, and designed those out before drawing anything. Every one of them is a habit-app default, which is exactly why they had to go on the wall first.
Even the emergency button refuses to be red.
02
Nine screens before Day 1.
A plan a person does not trust is a plan they abandon on Day 3. Intake asks what brought them, how long it has been, how the days feel and what would move them, then shows the finished plan back for review before a single day begins.
03
Tokens before screens.
Forty-six primitives resolved into thirty-four semantic tokens across two modes, and nothing in 325 screens holds a raw hex. No screen was redrawn for dark. The mobile app took six days; the other 236 screens took eight, because by then the system was doing the work.
One collection, two modes. Flip one property and three hundred and twenty-five screens follow.
The same system, three times over.
The member never sees the console. The operator never sees Day 12. Both are the same system, speaking to different people.
If it works, nobody will notice the design.
The sixty-day programme was authored by clinicians. The design only had to stay out of their way.