Fallow
Day Setup 0h budget 0.00 h planned 4h mind 6h body 15min per switch
· 0.00 h planned + 0.00 h switching · fully committed
· Intense mental work you sustain per day (~4h)
· Intense physical work you sustain per day (~6h)
· Context-switching overhead per transition
Plan
Does the day fit?
Is the time well spent?
What is it costing?
Can I keep doing this?
A to-do app that does calculus so you don't have to
Every productive day poses the same question: you have more tasks than hours — how should you split your time to get the most out of the day? Fallow doesn't guess and doesn't split evenly. It models each task's productivity curve and solves for the time allocation with the highest average productivity per task, weighted by the importance you give it and summed across your day — within the hours you have and the mental and physical energy you can actually sustain.
How it works
Productivity over time isn't flat. On any task you start at some initial level, ramp up as you get into the zone, peak at flow state, then decay as you tire. Fallow describes that curve with three things you feel intuitively: how demanding the task is (effort), how much you like doing it (enjoyment), and how long it takes you to get going (time to flow).
p(t) = (a·k·t + p₀) · e^(−kt)
Enjoyable, low-effort tasks start productive right away; hard ones start slow but peak higher. Every task also has a mathematically optimal stopping point beyond which your average productivity on it falls. The better Fallow knows your flow times, the closer the point it plans for sits to 1.52–1.79 times the time-to-flow; the less sure it is, the earlier it stops. Fallow plans in 15-minute blocks and never adds one that would pull a task's average down.
What makes Fallow different
Fallow doesn't guess the allocation: greedy marginal analysis over 15-minute block values, with an exhaustive search over which tasks deserve any time at all once context-switch costs are charged. When only your hours have to be respected, that search returns the exact optimum for up to 12 tasks, and past that whenever your day, list and switch cost together keep it bounded; once the cognitive and physical pools bind as well, it reaches the optimum on about 94% of days and lands within a few per cent on the rest. It then reports the Fallow Gain — how much better the optimized plan is than an equal split.
A dual energy-pool model tracks cognitive and physical capacity separately, so a day can't book more hard thinking than you have head for — and the suggested run order alternates the two systems instead of burning one down. The model also personalizes itself: log how long it took you to reach flow, and a least-squares fit tunes its constants to you. The Energy Lab takes it further — breaks and quitting time emerge from the math.
Frequently asked questions
Is Fallow free?
Yes. Fallow is free and open source — the full model, including every formula and derivation, is documented publicly. If it helps you, you can support development and buy me a coffee.
Where is my data stored?
Entirely on your device, in your browser's local database (IndexedDB). Your tasks never leave your machine and no account is required.
What is the Zenith Gradient Algorithm?
An approach to daily time allocation that models each task's productivity as a curve that rises to flow state and then decays, and allocates your day by marginal value — every next block of time goes to the task where it currently counts most. It originated in the article "How to Over-Engineer a To-Do App"; Fallow extends it with corrected math, energy pools, and context-switch costs.