Every kind of work Point does carries a level of its own. At the bottom, suggest-only: Point stays out of that work and leaves it with you. In the middle, Point prepares the work and waits for you. At the top, Point handles that kind of work outright, without asking. A new account arrives with everything sitting in the middle.
That’s the fourth of the six questions, and it’s the only one you answer yourself. What gets read, what can leave and what gets written down are facts about the product. They’re true before you touch a thing. This one is your decision, taken one kind of work at a time. You can take it again whenever you like.
Careful people do go wrong here, and rarely in the way they brace for. Raising a setting too early is seldom the problem. The common one is reading the whole arrangement as a single question about nerve. You answer it once, at the bottom, and leave it there for years.
Not a slider on your nerve
Picture what almost everybody pictures. One control, running from cautious to reckless. Moving it says how far you’ve come to trust the software in general. On that reading there’s one thing left to do, which is wait until you feel braver. Evidence rarely makes anyone feel braver. Two years pass.
Here’s the one correction worth making, because this one costs you something. The dial isn’t a single setting on your nerve. It’s a level per kind of work. Filing has a level, replying has a level, scheduling has a level, and each moves on its own. A single sentence you type can straddle two. Ask for a thread to be dealt with and answered, with the filing raised and the answering left in the middle. The filing happens. The reply is prepared and sits there waiting for you. One request, two levels, and each action carries its own.
That changes the question being asked of you, which is the useful part. It’s always about one narrow kind of work. For that kind, you decide: do it yourself, check it, or hear about it afterwards. That question has an answer.
Three places to put your check
The three levels are three places to stand in relation to an event.
At the bottom, the work stays yours. In the middle, your attention sits in front of the action. Point prepares the thing, you look at it, and you say whether it goes. At the top, your attention sits behind the action. It happened, and you find out by reading what was done.
The last chapter drew the line those three sit on. A check in front of an action can stop it. A check behind one only tells you about it. Both are worth having, for different jobs. The level is simply which of the two you’ve chosen for a given kind of work.
Put that way, a good deal of the fear drains out. Raising a setting moves a safeguard, from the costly side of the event to the cheap side. Whether that’s a good trade depends entirely on the work, which is the next thing.
The middle position is not free
Review looks like the position with the risk taken out, and a new account is handed to you there for sound reasons. It has a price, and the price is easy to miss. Missing it is how people end up worse off than they’d be at either setting beside it.
Something prepared and waiting is still an item. It sits somewhere. It wants a decision, and the decision falls to you at a moment you didn’t pick. For work that comes up twice a week, that’s fine. For work that comes up ninety times a day, it’s a second inbox, and everything in it is something you already agreed ought to happen.
You know what becomes of that queue. You’ve watched it happen to every approval queue anybody has ever built. It gets cleared rather than read. By the fourth week you’re going through it in blocks, at speed, on the sender and the shape of the thing rather than the contents. That’s exactly the reflex the first chapter named as the failure already running quietly in your building.
That lands you worse off than either honest position, for a reason worth being blunt about. A queue you clear at speed leaves everyone exposed, and it adds one thing a raised setting leaves out. It produces a record saying you approved. You take on everything approving means, and the reading is the part that got skipped. When something goes wrong, the record says you saw it.
So the safe position is the honest one. Picture a bad week in November, with a filing deadline on Thursday. The right setting is the one you’d still be keeping properly by then. A setting you follow sloppily costs you more than the honest setting on either side of it. That test alone sorts most of the rows, whatever you feel about AI in general.
You are raising a family, not a task
Levels are held per kind of action, and the kinds sit in families. Raise a family and everything beneath it inherits. Set the sorting and filing side of things to run on its own, and all of it runs on its own. That includes the members you’d forgotten were in there.
Worth slowing down for, because what you raised is wider than what you had in mind. You were thinking about the receipts. The family also holds whatever else counts as sorting, and that list lives in the product rather than in your head.
Which gives the two weeks after any change a job to do. The record is where a family becomes visible, one row at a time. Read it closely for two weeks, and what you’re really learning is what you agreed to. Most of what goes past is dull. Dull is the finding. The row that stands out is the one telling you the family was wider than the intention.
There’s a second condition, and it’s easy to miss. Asking Point for something in plain words is itself a kind of action, and it carries a level like everything else. So what you hand over that way runs on its own once that level is raised too. Both have to be open. A raised setting somewhere in the tree stays a setting on that one branch, and it sits alongside the strangers held at the door rather than replacing them.
What one notch too high costs
Now the thing the last chapter promised: what a setting costs when it sits one notch above where it belonged.
Inside your own account, it costs less than you fear, and the shape of it is specific. Something was done that you’d have done differently. You find it on a Tuesday, four weeks in, with a time against it and the reading it was made on. It has a location. The filing goes back and the level comes down. The rest of the arrangement stands where it was, because the levels are independent of each other. Turning one down is as cheap as turning it up. A setting is something you live with, and you can change it on the day.
Outside your account the shape changes, and that’s why the row covering what you say to people earns the slowest movement of any of them. A message that went out under a raised setting is sitting on hardware belonging to somebody else, outside your reach, and nothing described in this series fetches it home. What you get is the record, quickly. That’s worth a great deal on the day, and it arrives after the fact.
The other direction costs you too, and it stays quiet while it does. A setting left too low produces no event, no row, no complaint. It buys a year of approving things you’d already decided were fine, with the benefit still out of reach and the quiet holding all the way through. That’s the same asymmetry that makes a blind spot expensive: the error that announces itself is the cheaper of the two.
The rows that stay where they are
The top is one place to stop among three, and for some rows the answer is to stay put. Some kinds of work are finished where they are. Leaving one on review permanently is a decision reached, rather than a decision postponed.
The reason sits in what the mailbox holds. Every judgment Point makes comes out of the correspondence. Correspondence is a partial account of a relationship: the call, the ten minutes in the corridor, the thing about that family you’ve known since 2014 and would never put in an email. Where the right answer depends on the part nobody wrote down, the work stays with you. That’s a fact about the document rather than about the model, and a better version of the software leaves it exactly where it is.
In a practice, that ceiling turns from a preference into something written down. Who may commit the firm to what has always been a question the firm settles. A firm’s own policy is where it belongs. Point holds the level per kind of action across everything a mailbox can be handed. What each level does, task by task, is set out on the dial’s own page.
A setting that cannot see where it is
One thing this chapter has quietly assumed all the way through. It’s the assumption that breaks first for the people most likely to be reading.
Every level here attaches to a kind of work. Filing. Replying. Scheduling. That holds while a kind of work means one thing to you, and for most of a week it does. Some weeks you wear two hats. If the practice is one of two things you run, replying to a client splits in two. It’s a careful thing on one side of a line and an ordinary thing on the other. One setting covers both only when something underneath already knows which side you’re standing on.
That’s a question about the wall between the two, and it’s the fifth of the six.