There's a moment, about eleven minutes into a puzzle that has stopped moving, when you would press a hint button if there were one.
We didn't build one. That was deliberate, and it's the decision that shaped everything else about how the app helps you.
What a hint actually does
Think about what you're asking for in that moment. Nearly always it's not "teach me something" — it's "find the next square and fill it in."
So it does. And now the puzzle is one square smaller and you are exactly as stuck as you were, because nothing you learned in those eleven minutes changed. The next time the board goes quiet, you'll press it again. It's the same trade every time and it gets easier to make, which is the problem.
There's a sharper version of this. A hint doesn't just solve a square — it tells you the square was solvable. That's the part it can't give back. The whole pleasure of the thing is the moment you find a square that was solvable all along and you'd walked past it four times. A hint hands you that moment already spent.
You were never stuck on the square. You were stuck on the looking.
The line we drew: bookkeeping, not thinking
Refusing to build something is easy and cheap and usually wrong. What makes it a real decision is having a rule for what you do build, and ours is one sentence:
The app does the clerical work. You do the thinking.
Almost every job in a Sudoku puzzle falls cleanly on one side of that line, and it turns out the line is easy to draw once you look for it.
Rubbing out pencil marks is clerical. When you place a 5, every 5 you'd pencilled into that row, that column and that box is now wrong, and you have to go and erase them by hand. Miss one and it will lie to you twenty minutes later — and you won't suspect it, because you wrote it yourself. So the app does it. Place a number and it clears that square's marks and strips the digit from every square in the row, column and box. Nothing goes stale.
Counting is clerical. A digit that's mostly placed already is the one worth hunting, because every copy of it on the board rules squares out for you — and working out which digit that is means counting nine numbers across eighty-one squares. So the app does it. Every key on the number pad carries the count of that digit still unplaced, so the digit you want is the one showing the smallest number. Six 8s down means the 8 key reads 3, and that 3 is the signal.
Seeing which squares share a region is clerical. Working out that a square's row, column and box between them rule out eight numbers is thinking — but finding which twenty squares those are is just looking. So the app does it. Touch a square and its row, its column and its box light up together.
Every one of those removes work without removing a decision. You still have to notice what the shading is telling you. That's the part that's yours, and it's the part a hint button takes.
The honest version of the counterargument
Here's where I should be fair to the hint button, because there's a real case for it and pretending otherwise would be cheating.
Sometimes you are not stuck on the looking. You are stuck because the puzzle needs a technique you haven't met yet, and no amount of staring will produce it. That's a genuine wall, and "try harder" is useless advice in front of it.
A hint gets you past that wall. It really does. What it doesn't do is tell you which wall it was — and that's the difference between being unstuck once and being unstuck from now on. Nobody has ever learned what a pointing pair is by watching a square fill itself in.
So the objection is right about the problem and wrong about the fix. If the wall is a missing technique, the thing you need is the technique, and it takes about six minutes to learn. That is what this series is for: naked singles first, hidden singles when the board goes quiet, pointing pairs when both of those stop paying.
Those three will take you through almost every Easy puzzle in our app — 205 of the 207 that ship inside it — and 16 of the 57 Mediums. A hint button would have gotten you through one square of one of them.
What to do at minute eleven instead
Practical, since "learn a technique" isn't much use to you mid-puzzle.
Write your candidates down. Most of what feels like being stuck is re-deriving things you already worked out and didn't record. If you've been holding it in your head, that's probably the whole problem, and it isn't a character flaw — it's just an unwinnable amount to hold.
Change the question. If you've been asking "what goes in this square?" and getting nowhere, ask "where in this box can a 7 go?" instead. Same board, same rules, completely different set of things you notice. Most stuck boards aren't short of information, they're short of angles.
Look at the busiest region, not the emptiest. Singles hide where the board is crowded. The wide-open corner feels like it should be tractable and it's the worst place to spend a minute.
Put it down. Genuinely. Ten minutes away and the square you couldn't see is often the first thing you spot. This is not mysticism, it's just that you stop looking in the same place.
Worth being precise about that last one, because it sounds like "close the app," which is the one outcome we'd hate. It isn't the same thing. Putting a puzzle down mid-solve and picking it up later is the loop working — you left because you'd stopped seeing, and you came back because you wanted the square. Closing it because the wall never moved and there was nothing to do about it is the loop failing. Same gesture, opposite meaning, and the difference is entirely whether you came back.
Which we can't see from here, and I'd rather say so than pretend otherwise. At the moment you put the phone down, the advice I've just given you and the failure I'd most want to prevent look exactly alike. That's uncomfortable and it's the honest position: we built for the version where you come back, and the only evidence either way is whether you do.
The one we'd revisit
I want to name the limit of this, because a refusal you'd never reconsider isn't a decision, it's a slogan.
If it turned out that people were bouncing off Sudoku entirely — not putting a puzzle down and coming back to it, but closing the app on a wall and never opening it again — then "the satisfaction is yours" is a lovely principle that's protecting an experience nobody's having. The answer still wouldn't be a hint button. It'd be teaching the technique at the moment you need it, which is the thing a hint pretends to do and doesn't.
That's a harder feature to build and we haven't built it. But it's the direction we'd go, and it's why the line is "the app does the clerical work" rather than "the app does nothing."
Next in this series: what to do when you're stuck and there's no hint button — the longer answer to minute eleven. And if the answer turns out to be a technique you haven't met, Sudoku techniques, in the order worth learning them is the whole list.