Herbert Simon
The best one was never going to arrive
Buyers come one at a time, never all at once
"A Behavioral Model of Rational Choice", Quarterly Journal of Economics, 1955; "Rational Choice and the Structure of the Environment", Psychological Review, 1956
Both options are fine, I have compared them for weeks, and I am still waiting for a better one. When do you stop looking?
What actually happened?
In 1955 Simon used the sale of a house to make his point. The rational man of the textbooks lines up every possible buyer and every possible offer, then takes the highest. A real seller cannot wait for every buyer: they arrive one at a time, and the one turned away today may be gone tomorrow. So the seller fixes a price he will accept and takes the first offer that reaches it; if none does for a while, he lowers the price a little. That is not laziness. It is the only kind of rationality a limited mind can finish computing.
Optimising needs three things you never have
To pick the best you must know every alternative, every consequence of each, and how to rank those consequences. Simon's point was that real people have none of the three in full. He called it bounded rationality: people are not unreasonable, reasoning just costs more than they have.
You have seen twenty flats and want to see one more. The first good one has already been let.
He gave the habit a name
In the 1956 paper he modelled a small animal hunting for food and able to see only a short way ahead, and showed that a very simple rule kept it alive without any optimising. He called the rule satisficing: clear the line, then stop.
Write down three hard requirements for the hire and make an offer to the first candidate who meets all three.
The line moves by itself
The line is not fixed. When satisfactory options turn up easily, aspiration rises; when they will not turn up at all, it falls. Good enough is not a number you bring to the search. It is the number the search and the world agree on.
Thirty applications and not a single reply means the line should move, not that you should send thirty more.
A pair of scissors
In 1990 he put it another way: human rational behaviour is shaped by "a scissors whose two blades are the structure of task environments and the computational capabilities of the actor." When you cannot choose, stop sharpening your own blade and work on the other.
Eight plans are too many to compare. Set a rule that cuts five, then compare the three left.
How do I use it today?
Where you are: two good options, two weeks of comparing, and a quiet hope that a third will turn up.
Ask first: which few things do I actually need - written down, with the ones I cannot do without marked - and has one of the options in front of me already cleared all of them.
Where it goes wrong: reading it as settle for whatever. He never says the line should be low. It should be set first, written down, and moved with how hard the search proves to be - and the less reversible the choice, the higher it sits and the longer you look.
Lines to keep
Draw the line first. Then see who clears it.
Easy to find, raise the line. Hard to find, lower it.
When you cannot compare them all, change the problem.
Same situation, other people are asking
Every option makes sense and I can't choose.They're all good ideas. That's exactly the problem.I'm climbing and it isn't the thing I wanted.What I want to do and what I should do are two different things.Too many paths · all 10 questions →If this one named what you are going through,
send it to someone who needs it, or keep it somewhere you will find it again.