Methodology
How Pomop decides what you should do next
Pomop scores every open task on five weighted factors — urgency (35%), importance (25%), time fit (15%), task age (15%), and workload impact (10%) — and recommends whichever scores highest.
A task list can tell you what exists. It doesn't necessarily tell you what deserves attention right now. Pomop's decision engine is designed to turn a list of tasks into one deterministic recommendation, based on the current task data — not a mood, not a guess, and not an AI model.
Written by Independent Product Builder · Last updated August 28, 2026
The five factors
Every open task is scored on five factors. None of them decides the outcome alone.
Urgency
Based on how close a task's deadline is. Pomop's implementation uses fixed bands: overdue is the most urgent, followed by due within 24 hours, due in 1–3 days, due in 3–7 days, and due in 7 or more days. A task with no deadline is treated as the least urgent — an undated task never outranks a dated one on urgency alone.
Importance
The priority level you set on a task — low, medium, high, or critical — mapped to an evenly spaced scale. If a task's importance is missing or invalid, it defaults to medium, never to critical, so incomplete task data can't accidentally dominate the ranking.
Time fit
How closely a task's estimated remaining time matches the time you actually have available right now. A task that needs more time than you have isn't excluded for that — it's still eligible, just recommended as a partial session. A task with no time estimate at all isn't penalized for the missing data.
Workload impact
How much of a task's remaining work one session would actually clear. A short task you could finish outright scores differently here than a large task a session would barely dent — this is separate from the day-level Workload Summary shown elsewhere in the app, which looks at your whole list, not one task at a time.
Task age
How long a task has been sitting on your list, on a scale that's capped after 30 days. This is deliberately the gentlest factor — an old task gets a modest nudge so it isn't buried forever, not a takeover that lets it jump ahead of genuinely urgent work.
The scoring model
The current implementation weighs the five factors as follows:
| Factor | Weight |
|---|---|
| Urgency | 35% |
| Importance | 25% |
| Time fit | 15% |
| Task age | 15% |
| Workload impact | 10% |
These are Pomop's current implementation weights — a product decision, not a universal productivity rule or a scientifically proven optimum. The current implementation combines these factors so the recommendation isn't based on a single task attribute.
Why not just priority?
A task can be marked high importance and still not be the best thing to work on right now. It can become more relevant for reasons that have nothing to do with how important it was originally marked: its deadline is approaching, it fits the time you actually have available, it's been waiting a long time, or working on it now would meaningfully reduce your workload. Pomop's decision engine combines all of these rather than sorting on any single one.
Deterministic, not AI
The recommendation is produced through a fixed scoring calculation, not an AI model. Given the same tasks, the same available time, and the same moment, Pomop always returns the same recommendation — predictable, inspectable, and repeatable. That's a tradeoff, not an automatic win: deterministic doesn't mean perfect, it means the same inputs always produce the same, explainable output, rather than a black-box judgment call.
"Why this task?"
Every recommendation comes with plain-language reasons, like "Due today" or "High importance." These reasons are built directly from the same factors used to calculate the score — not a separate explanation layered on afterward. If a reason doesn't apply to a task (for example, a task with no deadline has no urgency-based reason), it simply doesn't appear. This isn't AI-generated reasoning; it's a direct readout of the scoring inputs.
What the recommendation doesn't know
The decision engine only ever sees the task data you enter. It has no way to account for:
- your personal energy or focus level at a given moment
- context or importance that isn't reflected in a task's priority or deadline
- unexpected events that come up outside the app
- dependencies between tasks — it doesn't know if one task is blocked on another
- other people's priorities, deadlines, or work — Pomop only sees your own task list
When Pomop may not be the best fit
Pomop is built around one person's own task list, not shared work. It's likely a weaker fit for:
- team or collaborative project management, where priorities depend on other people's work
- heavily meeting-driven schedules, where a calendar — not a task list — is the primary source of truth
- tasks with complex dependencies that need to be sequenced, not just ranked
- environments where priorities are set and change outside of what's recorded in the app
Privacy
Your task data — titles, deadlines, priorities, and progress — is stored only in your browser's localStorage, with no account and no backend that receives or processes it. This is separate from website analytics: Pomop uses Google Analytics, gated behind the cookie-consent banner, to understand aggregate site usage. That's a different, third-party service from your task data, and it never receives your tasks. See the full Privacy Policy for details.
Related
- What is decision fatigue?
- How to reduce decision fatigue
- What to do when you have too many tasks
- To-do list vs. calendar
- Eisenhower Matrix vs. time blocking
See it in practice: add your tasks and get one recommendation onWhat Should I Do Next.