Rollo

When I became a product manager, the volume of decisions, follow-ups, and context I carried grew faster than my personal system. I used a custom Notion template for daily notes and manually moved unfinished items from one day to the next. Tasks were easy to leave behind, and the reasoning behind them became detached from the work itself.

I spoke with roughly ten colleagues, friends, and technically minded people about their workflows. Everyone had assembled a different mix of notes, task managers, calendars, local language models, and sync services. A meaningful minority had abandoned productivity systems because maintaining several tools created more work than clarity.

What if the system carried unfinished work forward automatically—and kept the context beside the task?

The first web version was deliberately narrow. Today held immediate work, Someday held the backlog, and Timeline held future commitments. At midnight, unfinished tasks rolled forward. A manual End Day action provided closure and an undo path.

I shared early builds with a few colleagues for usability feedback and bug discovery. The response was positive, but requests expanded almost immediately. Mobile access was the loudest need, so I turned Rollo into a progressive web app installable on desktop and mobile.

The defining constraint was that every task should remain connected to knowledge. A note block could become a task in one action. A task made inside the planner became a real checkbox inside a daily Markdown note. There was only one task model, regardless of where work began.

The native application worked, but I postponed distribution. Android had only been tested through an emulator, and desktop distribution introduced developer-account, signing, and trust requirements before I knew whether anyone would use the product. I decided to polish first.

That decision created the wrong loop. I kept improving UX and adding capabilities from competitive research while avoiding the question that mattered: did the product solve a common problem better enough to earn a place in someone else’s workflow?

At the same time, Claude, Codex, and other agentic tools gained direct access to many file formats. Users could assemble personal workflows, automations, and interfaces around their existing habits instead of adopting Rollo’s standardized system. I concluded that the opportunity no longer justified continued investment and stopped development.