Am I actually stuck?

This is a sorting tool, not an explanation. It makes no claim about why a task stalls. It only takes the moment where something has not moved for two days and separates it into cases, because once you know which case you are in, the next step is usually obvious and sometimes trivial.

Take one specific task. Not the module, not the project — the actual thing that has not moved. Read the seven and stop at the first that fits.

1. You are waiting on someone else

   How to tell. Write the very next step. If the sentence has another person's name in it, this is the case.

   What to do. Send one message today with a date in it. Not any update? — instead: I need the figures to finish this by Thursday. Can you send them by Wednesday morning, or tell me if that is not possible?

   Then move it off your own list onto a waiting list with the date on it. It is not yours at the moment, and every hour you spend feeling bad about it is an hour the delay stays disguised as your problem rather than theirs.

   What not to do. Do not start rebuilding the missing piece yourself as a way of feeling active. That is often more work than the whole task.

2. You do not know how

   How to tell. You could describe the finished thing, but not one step of getting there. Opening the file produces nothing.

   What to do. Convert it into a question and ask a person. Write the question first, in one sentence, because half the time writing it answers it. If nobody is available, spend a fixed twenty minutes finding one worked example of the same kind of thing — not reading around the topic, but finding one finished instance you can look at directly.

   What not to do. Do not sit with the blank document. Time in front of it does not accumulate into knowing how, and after an hour you will have added a belief about yourself to the original problem.

3. You know how, but not what to physically do first

   How to tell. The task is a noun. The report. The application. The kitchen. You know how each part works, and there is no obvious place to put your hands.

   What to do. Write the first physical action, at the level of a body movement. Not start the report but open last year's report and copy the headings into a new file. Not tidy the kitchen but take everything off the left half of the counter.

   If the first action still takes more than about a minute to imagine, it is not small enough. Halve it again.

   What not to do. Do not plan the whole thing. Planning is enjoyable and this is the case where it substitutes most convincingly for starting.

4. You are dreading one specific part

   How to tell. Something in you moves away from a particular section, sentence, phone call or name. Everything either side of it is fine.

   What to do. Name the part, out loud or on paper, as narrowly as you can. Usually it turns out to be small — a single email, a number you will have to admit, a paragraph you know is wrong.

   Then choose one of two: do that part first, alone, with nothing else attached to it, or do everything else and leave a clearly marked gap where it goes. Both work. What does not work is starting at the beginning and hoping to arrive at it with momentum.

   What not to do. Do not treat the whole task as difficult. It is not. One piece of it is, and mislabelling it makes the other nine parts inaccessible too.

5. You do not believe it matters

   How to tell. You can do the work. You just cannot find a reason that survives contact with your actual opinion of the task.

   What to do. Say what it is actually for, in one honest sentence — including so that a person I do not respect stops asking me about it, if that is the truth. A real small reason holds better than an invented large one.

   Then settle what standard it needs. A great deal of stalled work is held to a standard nobody asked for. Ask what is required, and if the answer is that it needs to exist rather than be good, do the version that exists.

   What not to do. Do not try to generate enthusiasm. If it were available it would already be here.

6. You are too tired

   How to tell. You have read the same line several times. Decisions that are usually automatic have become effortful. This has been true since before you sat down.

   What to do. Stop, and stop properly — away from the desk, without a screen, for a real length of time. A tired hour produces work you will redo, so the hour is spent either way.

   If stopping is genuinely not available, pick the least demanding piece of the task. Formatting, references, filing, tidying inputs. Low-demand work done tired is fine. Decisions made tired are the expensive ones.

   What not to do. Do not sit at the desk in a state of neither working nor resting. That is the one option that costs the time and returns nothing.

7. The task is wrong

   How to tell. It has stalled repeatedly, across different weeks and different moods, and each time you look at it something about it does not make sense. Perhaps it duplicates something. Perhaps the person who asked for it has moved on. Perhaps you agreed to it in a week when you agreed to too much.

   What to do. Say so. I do not think this is worth doing, here is why is a legitimate sentence, and if it turns out to be worth doing you will get a reason back, which resolves case 5 as a side effect.

   If it is genuinely yours to decide, delete it. Write one line somewhere recording that you decided not to do it and when, so it does not return in three weeks as a vague feeling of failure.

   What not to do. Do not keep moving it forward a week at a time. That costs more attention over a term than doing it would have.

What the pattern says about the setting

Run this a few times and note which number you land on. The distribution is more informative than any single result.

If you land on 1 repeatedly, the problem is a handover, not your persistence. Work that routinely waits on someone with no agreed response time is a queue that nobody has been made responsible for. That is raisable with a supervisor, a course rep or a team lead, and the request is concrete: an agreed turnaround for requests, or a named second person to ask.

If you land on 2 repeatedly, something is being assumed as taught that was not taught. This is very common in the first term of a course and the first months of a job, and it is rarely individual — if you did not know, most of the cohort did not either. One worked example on the module page or in the shared drive settles it, and asking for that example helps everyone behind you.

If you land on 7 repeatedly, work is being generated that nobody has audited. That is worth naming out loud in the room where the work is assigned.