A USMLE Step 1 study schedule that survives a bad week

What actually breaks a dated plan

A dated schedule makes a promise on your behalf. Tuesday is cardiology, Wednesday is renal, and the spreadsheet says so. It holds until the week you are on a ward, or ill, or simply flattened — and then Tuesday’s work is not merely unfinished. It is late.

What happens next is predictable enough to design around. You open the plan and the first thing it shows you is a list of things you failed to do. So you stop opening it. A week later you rebuild the whole schedule from scratch, which costs an evening and moves no material. The plan did not fail because the pace was wrong. It failed because it kept score.

A schedule should tell you where you will finish, not how late you are.

Setting it up is one screen

The course opens the first time with a plan wizard, and after that it stays out of the way. Four decisions, all of them reversible:

• Tick the systems you intend to cover. Organ system by organ system. Leave out the ones you have already worked through, or narrow it to the three that scare you.
• Drag them into order. The sequence you want to study in, not the sequence someone else thinks is canonical.
• Set a weekly pace in tasks per week. Seven by default. Tasks, not hours — you know what a task costs you after the first week, and you never really knew what an hour of studying bought.
• Set a weekly study goal in days. How many days a week you intend to touch the material at all. Five by default. This is a separate, smaller promise than the pace, and it is the one worth keeping on the bad weeks.

As you move the pace, the wizard shows the total number of tasks your ticked systems come to and how many weeks that implies. The arithmetic everyone does badly in their head, done in front of you before you commit to it. If the answer is uncomfortable, it is better to be uncomfortable now than in week nine.

A rolling week, not a calendar week

Pacing runs on a rolling seven-day quota window, anchored at the day your plan started. The course home shows the systems grid with a done count on each, this week’s progress as done out of quota, and an “up next” queue sized to whatever is left in the window. That is the whole dashboard.

There are no dates on tasks, and nothing is ever marked overdue, because nothing can be. Fall behind and exactly one number changes: the projected finish. The line reads on pace to finish and a point further out. Catch up and it comes back in. The plan reprojects; it does not accuse.

That distinction sounds cosmetic and is not. An overdue counter creates a debt you have to pay before the plan feels usable again, and debts get avoided. A projected finish creates a consequence you can look at calmly — a week of nothing costs you a week at the end, and you can decide whether to buy it back by lifting the pace or accept it and carry on.

Editing the plan halfway through a week — adding a system, dropping one, changing the pace — does not shift the window you are currently in. This week’s quota stays what it was on the day the week started. A Thursday reshuffle cannot hand you a fresh deficit for work you have already done.

Falling behind moves one number, and it is the date you expect to finish.

What a task actually is

A task is not a quiz and not a chapter. It is an ordered list of typed blocks, and the mix depends on the material: page assignments in a reference text, a list of videos with their titles and running times, links to supporting resources, a set of board-style questions, and notes written into the task itself.

Every block is its own checkbox. So is every individual video inside a video list. Tick the last outstanding one and the task completes itself — you never have to remember to close it.

The granularity exists for the same bad week as everything else here. Twenty minutes on a bus gets you two videos ticked, and the task remembers exactly where you stopped. Progress that only counts in whole chapters does not survive a real timetable.

The practice is attached, but it belongs to the bank

A task does not own a private set of questions. It carries question identifiers, and those are matched against the published question bank when the task loads. The practice button shows the result of that match in the open: a task whose ten questions include seven in the bank reads Practice here (7/10), and starting it gives you a session containing exactly those seven.

The three that did not match are left alone as external links, not quietly dropped and not padded out with substitutes. You can see how much of the practice lives in the app before you tap anything.

Questions you get wrong are where the assistant earns its place — ask why the answer was not the one you picked, and the explanation arrives with the page it came from, which you can open. That is the subject of how every answer is cited.

The streak, and what none of this does

Running alongside the plan is a study streak: consecutive days on which you did something real. Asked the assistant a question, answered a practice question, or completed a course task. It says nothing at all below two days, because one day is not a run. A streak that ended yesterday is still live — today is not over yet, and the app does not get to declare it lost before you have had a chance at it.

Now the limits. The course does not know your exam date, and it will not tell you whether your projected finish is early enough for it. That comparison is yours to make, and it is the one thing worth checking on a calendar. The material covers Step 1 only. The prep course is included in the Standard and Ultra tiers — the plans page has the detail.

And the obvious one: this is a study aid for exam preparation and revision, not medical advice. A completed task means you have covered the material, not that you have understood it. The questions are there to find out which.