Suasoria keeps a streak. It counts the days you have studied, it shows you the number, and one missed day a week is absorbed without comment by a grace day. What it will never do is turn that number into a threat.
No dying flame. No warning that your streak is at risk. No message the evening you have not opened the app. No screen, the morning after, mourning what you lost. The count is a record of something you did, offered because it is pleasant to see, and it is not a hostage.
This is worth explaining, because the other design is so common that its absence looks like an oversight.
What the threat actually does
The mechanism is not mysterious. A streak that can be broken converts a record into a possession, and losing a possession hurts more than gaining an equivalent one pleases. Build a number over forty days and the prospect of returning to zero acquires a weight out of all proportion to anything that number represents. That weight is the product. The counter is not there to inform you; it is there to make tonight feel expensive.
We could stop at "that is manipulative" and most people would nod. But the objection we actually hold is narrower and, we think, harder to argue with: it does not work for what we are trying to do.
Consider what happens on the day the streak breaks — and it breaks, because a person gets ill, or travels, or has a week where everything is difficult at once. Every reason to open the app that was made of the streak vanishes at that moment, all together. The learner who was studying to protect a number now has nothing to protect. This is the predictable end state of the design, and it arrives sooner for the learners who had the least reason of their own to be there.
Meanwhile, whatever genuine interest the material had generated has been quietly competing with the counter for weeks, and the counter is louder. A reason that shouts tends to crowd out a reason that does not. When it goes, it may take some of the quieter reason with it.
The learner here is often a child
That is the second half of the argument and the part we are least willing to negotiate.
A five-year-old in Tier I cannot evaluate what an app is doing to them. They do not have the distance to notice that the anxiety they feel about a number was manufactured, that it belongs to somebody else's retention target, and that it has nothing to do with logic. They will simply feel it, and will attach it to the thing that is in front of them — which is reasoning, which is the subject we would like them to associate with something else entirely.
A curriculum in argument that recruits anxiety to keep attendance up is teaching something it did not mean to teach, and it is teaching that lesson far more effectively than it teaches the syllogism. Every part of the app is an argument about what reasoning is for. A counter that pressures is an argument too, and not one we want to make.
What is left, and why we kept it
None of this means the streak is worthless. Seeing that you have studied nine days running is a real and quiet pleasure, and there is nothing wrong with a small piece of evidence that you have been doing the thing you meant to do. The pleasure lives in the record. The pressure comes from making the record fragile.
Removing the fragility keeps the first and drops the second. The grace day is the same principle applied one level down: a person who studies most days is a person who studies most days, and a counter that cannot represent that fact is measuring the wrong thing. Illness should not have to be reported to a spaced-repetition schedule.
The copy follows from all of it. Nothing on any screen frames a missed day as a failure, because it is not one, and because the app is not owed an explanation. When you come back, your work is where you left it and the lesson resumes at the screen you stopped on.
What we are relying on instead
The honest position is that we are taking on more work by refusing this. Loss aversion is a powerful lever, and giving it up means the material has to be genuinely worth returning to. Lessons have to end somewhere a learner wants to continue from. Explanations have to be worth reading when you got the answer right. Recitation has to feel like remembering rather than like a chore that has come due.
We would rather have that problem. It is the problem of making something good, and every hour spent on it improves the thing itself. The other problem — how to make leaving feel costly — improves nothing, and solving it well would leave us with a worse app that more people opened.
Come back tomorrow, or come back in a fortnight. The count will be waiting, and it will not have anything to say about it.