Maximum 1-3 active implementation intentions at a time — add more only after existing ones have compiled into automaticity

Why This Is a Rule

Implementation intentions work by heightening cue accessibility — making your brain more sensitive to detecting the trigger condition. "When I see my running shoes by the door → put them on and run" works because the intention increases your perceptual sensitivity to the running shoes. But this heightened sensitivity consumes attentional resources. Each active implementation intention claims a portion of your cue-detection bandwidth.

With 1-3 active intentions, each gets enough bandwidth for reliable cue detection. With 7+ active intentions, the bandwidth is spread so thin that cue detection becomes unreliable for all of them — none fire consistently because each is barely above the detection threshold. The result is worse than having no intentions at all: you've committed to behaviors that don't activate, producing the frustration of failure without the learning of having tried.

The progression rule — add new intentions only after existing ones have compiled into automatic execution (Test habit automaticity by effort level, not frequency — willpower-maintained consistency is not genuine delegation to the habit system) — ensures bandwidth recovery. Once an implementation intention has become automatic (the cue-response fires without conscious monitoring), it no longer consumes active cue-detection bandwidth. That freed bandwidth can then support a new intention.

This is Automate each habit link before adding the next — stacking unautomatic behaviors creates an effortful pile, not a self-sustaining chain (automate each link before adding the next) applied specifically to implementation intentions: sequential installation rather than simultaneous overload.

When This Fires

• When designing implementation intentions for multiple new behaviors simultaneously
• When existing intentions aren't firing reliably — check if you've exceeded the 1-3 capacity
• When you want to install many new behaviors at once ("new year's resolutions" syndrome)
• Complements Automate each habit link before adding the next — stacking unautomatic behaviors creates an effortful pile, not a self-sustaining chain (sequential habit stacking) with the implementation-intention-specific capacity limit

Common Failure Mode

New Year's resolution overload: "I'll exercise daily AND meditate AND journal AND eat clean AND read AND study." Six simultaneous implementation intentions, each consuming cue-detection bandwidth, none firing reliably. By January 15, all six have failed because none had sufficient attentional support. Three at most; ideally one until it's automatic, then add the next.

The Protocol

(1) Count your currently active implementation intentions (novel if-then plans that haven't yet become automatic). (2) If ≤3 → you're within capacity. Each has enough cue-detection bandwidth to fire reliably. (3) If >3 → you're over capacity. Prioritize: which 1-3 are most important? Pause the others until the priority intentions compile into automaticity. (4) Progression: when an active intention becomes automatic (Test habit automaticity by effort level, not frequency — willpower-maintained consistency is not genuine delegation to the habit system — fires without deliberation) → it no longer counts toward the active limit. Add a new intention to fill the freed bandwidth. (5) The tempo: install 1-3 intentions → compile over 2-4 weeks → verify automaticity → install next batch. This is slower than simultaneous installation but produces reliable habits rather than a graveyard of abandoned intentions.