How many behavior standards should a personal principle have?

Two or three, and the reason has nothing to do with tidiness. Below that number a principle stays unfalsifiable — there is nothing specific enough to point at and say it held or it did not. Above that number the daily check quietly turns into a small exam with several questions on it, and exams get gamed, rushed, or eventually skipped. The useful range is narrow, and it is worth understanding why, because the number itself is less important than what is failing on either side of it.

What each additional standard is actually buying you

A behavior standard exists to make an abstract rule checkable on an ordinary day — the thing a stranger following you around could confirm or deny without asking you what you meant. The first standard usually does most of that work: it turns "be honest with my team" into something like "I say the hard part of an update out loud, not just the parts that land well." The second standard often catches a different situation the first one misses — honesty in writing as well as in meetings, say. By the third, you are typically covering the last obvious gap. Past that point, each new standard tends to restate one you already have, or reach for an edge case rare enough that it does not deserve permanent real estate in a daily check.

The floor: one is barely better than none

A single standard can work for a genuinely narrow principle, but for most principles one is not enough coverage — it lets you satisfy the letter of the standard while missing the point of the principle on every axis the standard does not touch. "Be present with people I love" reduced to one standard — "phone in another room at dinner" — can be fully true on a night you were physically there and completely checked out. One standard is a start, not a finished principle; if it is all you have after a week of using it, that is a sign to look for the situation it is not catching, not a sign to stop.

The ceiling: why three tends to be where it breaks

At check-in you are holding one day in your head for a few seconds, not reconstructing it from notes. Two or three yes-or-no questions fit inside that window; five or six start pushing you into estimating rather than recalling, and estimating is where standards quietly stop being checked and start being assumed.

The averaging problem

The more standards a single principle carries, the easier it becomes to bury a real miss under several easy passes. Five standards where you clear four and miss one still reads, at a glance, like a good day — and it may even be recorded as one. That is the exact failure mode that makes streak-based habit trackers unreliable, imported inside a single principle instead of happening between a principle and its tracker: a proxy that averages away the one data point that actually mattered.

What overload looks like in practice

• Reviewing the standards for one principle starts to feel like a small chore, distinct from how the check-in feels for everything else.
• You catch yourself marking a standard "yes" without genuinely checking, because checking all of them properly would take longer than you are willing to give it.
• A standard got added not because it clarifies the principle, but to cover for one specific exception that happened once — a patch that never gets removed.

If a principle keeps needing a fourth standard, it is probably two principles

When a principle resists staying at two or three, the usual cause is not that the principle is unusually demanding — it is that it is covering more than one situation and has been written as though it were one. "Be present with people I love" quietly spans attention at home, patience with strangers who are not owed it, and following through on plans made weeks in advance. Trying to write one standard for each of those under a single heading produces exactly the pile-up described above. Splitting it into two narrower principles — each with its own two or three standards — almost always produces a system that is easier to keep than one broad principle straining under five.

A worked minimum-viable version

Principle: I do not answer a request at the moment I am asked, if answering costs me more than an hour.

• I say "let me get back to you" instead of an immediate yes to anything above that threshold.
• I give myself until the next morning before sending the actual answer.

Two standards, both checkable by end of day, both catching a different part of the same failure — answering too fast, and caving before the delay is up. A third would likely just restate one of these in different words.

When one is genuinely enough

Some principles are narrow enough by design that a single standard is not a compromise — it is the whole job. "I do not check work messages after 8pm" needs exactly one checkable fact, and adding a second standard to a principle that specific usually just pads it. One clean, honestly-checked standard beats three vague ones every time; the target is not a number, it is a principle you can look at for five seconds at the end of the day and actually know the truth about.

CreedOS's built-in templates ship with their behavior standards already written, generally two or three per creed, for the same reason described above: a daily check-in only holds up if answering it takes less time than deciding whether to bother.