You decided. Properly, not vaguely. You were going to start running, or finish the application, or have the conversation.
Three weeks later it has not happened, and not because you changed your mind. If asked, you would still say you intend to do it.
That gap between deciding and doing is one of the most studied phenomena in behavioural psychology, and the useful finding is that the fix is not more resolve.
The gap, and how large it is
Intentions do predict behaviour. They predict it considerably more weakly than people assume.
Reviews of the literature have found that a substantial proportion of people who form a genuine intention do not follow through, and that the relationship between intending and doing, while real, leaves a great deal unexplained. Estimates vary by domain and by how intentions are measured, and the consistent finding across them is that knowing what someone intends tells you much less about what they will do than intuition suggests.
The important detail is who the non-actors are. They are not people who secretly did not want it. Studies distinguishing between people who never intended to act and people who intended and failed find that the second group is large, and that their intention was sincere.
Almost everyone who fails to do the thing still wants to do the thing. The problem is not located where it feels like it is located.
Why intentions are weaker predictors than they feel
Three reasons.
An intention is formed in one state and executed in another. You decide on Sunday morning, rested and reflective. The execution happens on Wednesday evening, tired, with competing demands. These are different conditions and the version of you present at each is differently equipped.
An intention specifies an outcome, not an action. "Get fit" or "finish the application" names a destination. Nothing in it says what happens at any particular moment, so every moment requires a fresh decision about whether now is the time.
Competing demands arrive continuously. The intention has to win repeatedly against things that are urgent, present and concrete, while it remains abstract and postponable.
None of these are character problems. They are structural features of how intentions are usually formed.
What actually closes the gap
The most robust finding here concerns implementation intentions, a technique developed by the psychologist Peter Gollwitzer.
The format is simple: instead of stating what you intend to achieve, state when, where and how you will act. "If situation X arises, I will do Y."
Not "I will exercise more" but "on Monday, Wednesday and Friday, when I finish work, I will change into running clothes immediately and go out of the front door."
A meta-analysis by Gollwitzer and Paschal Sheeran, covering a large number of studies, found that forming implementation intentions had a medium-to-large effect on goal attainment across a wide range of domains. That is unusually strong for a behavioural technique, and it has held up better than many findings from the same era.
Two caveats worth stating. Effects are larger for goals people already want than for goals they are ambivalent about. And the technique works best for behaviours that are within your control, so it helps with going to the gym and does less about outcomes that depend on other people.
A worked example: the same intention, two formulations
Two people want to finish a professional qualification. Same course, same deadline, same genuine motivation.
Person one forms a goal intention. "I am going to study for this properly and get it done before June."
What this leaves undetermined: which days, what time, where, for how long, what happens first. Every single study session requires deciding, in the moment, that now is the time. Each of those decisions competes with whatever else is happening.
Week one goes well, because novelty supplies the energy. Week three has a busy Tuesday and nothing happens. Week five is mostly not happening. By April there is a sense of being behind, which makes starting harder, because starting now means confronting the backlog.
Person two forms an implementation intention. "On Tuesday and Thursday, when I get home and put my bag down, I will sit at the kitchen table and do one chapter before I do anything else."
Note what is specified: two named days, a physical trigger, a location, a defined unit of work, and an ordering relative to other activities.
Tuesday arrives. The bag goes down. The behaviour is already decided, so there is no deliberation. On several Tuesdays it happens more or less automatically.
Person two also misses sessions, because nothing works perfectly. But the misses are single events rather than the collapse of a system, because the following Tuesday the cue arrives again unchanged.
The difference is not motivation. Both wanted it equally. One handed the decision to a cue and one left it to be made repeatedly under pressure.
The most common error. People write implementation intentions that are still goals wearing a plan's clothing. "I will study more this week" is a goal. "I will study on Tuesday" is closer but still leaves when open. "On Tuesday, when I put my bag down, I will do one chapter at the kitchen table" is an implementation intention. If your plan does not specify a trigger you will physically encounter, it is not specific enough to work.
Why specificity works
The proposed mechanism is delegation.
A vague intention leaves the decision to be made deliberately, each time, by the effortful part of your mind, in whatever condition it happens to be in.
A specific if-then plan attaches the action to a situational cue. When the cue occurs, the action is initiated with less deliberation, because the deciding was done in advance under better conditions.
This connects directly to the fast and slow processing described elsewhere in this cluster. Implementation intentions work by moving a behaviour from something the slow system has to authorise each time into something the fast system can trigger on recognition. The strategic decision is made once, deliberately. The execution becomes closer to automatic.
Which explains why the specificity matters so much. A cue you will actually physically encounter, at a moment you will actually notice, is what makes the delegation possible. "When I have time" is not a cue. "When I put my bag down" is.
How this is taught inside Astra Trainer
This sits in the Subconscious patterns direction of the Psychology & Mind world, alongside a sequence on habit loops that includes a lesson called Why You Break Promises to Yourself.
That title names the experience this article is about, and the direction's framing is the useful one: fourteen courses on naming the automatic patterns you repeat and running small experiments that change them without trying to become a different person. An implementation intention is exactly that shape of intervention, a specific change to how a decision gets made rather than an attempt to become more disciplined. Lessons take about five minutes and a guide walks you through anything strange.
The second half: what to do when it goes wrong
Most people who use implementation intentions stop at the first half, and the second half is where a good deal of the remaining benefit sits.
The addition is a coping plan: an if-then for the most likely obstacle.
"If it is raining when I finish work, I will do the indoor session instead." "If I get home after eight, I will do half a chapter rather than skipping."
Why this matters: most plans fail at a small number of predictable obstacles, and when one arrives without a prepared response, the default is abandonment. Having decided in advance means the obstacle triggers an alternative rather than a collapse.
The practical method is to ask what is most likely to stop this, then write one if-then for it. Not every possible obstacle, which becomes unusable. The one or two that will actually occur.
There is a related finding worth knowing, from research on self-control failure: people who treat a single lapse as evidence that the whole attempt has failed tend to abandon, while people who treat it as one missed instance tend to continue. A coping plan helps partly by making the lapse feel anticipated rather than fatal.
Why deciding feels like progress
Worth naming, because it explains why people repeat this cycle.
Forming an intention produces a genuine sense of having done something. The decision feels like the hard part. There is relief, a sense of resolution, sometimes even a small satisfaction.
That feeling is not proportionate to what has changed, which is nothing. But it is rewarding enough to be self-reinforcing, which is why people can form the same resolution repeatedly across years, each time experiencing the formation as meaningful.
Some research on goal-setting suggests that announcing intentions publicly can reduce follow-through in certain conditions, apparently because the social acknowledgement supplies some of the satisfaction the achievement would have provided. The finding is debated and not universal, and it is consistent with the general point: the feeling of having decided is not evidence that anything will happen.
The practical implication is to treat the sense of resolution as a warning rather than as progress. If deciding felt good and nothing was specified, nothing has been done.
Why this compounds with the daily structure
An implementation intention works by attaching a behaviour to a reliable cue. Anything that supplies a reliable daily cue makes that easier.
That is one of the things the Psychology & Mind world's structure provides. Daily quests, points and a streak create a recurring prompt, with a streak shield for the day that goes wrong, which functions as a built-in coping plan for a lapse. Quizzes follow each topic and each course ends with a ten-question final exam, and the daily Connections round and 10x10 crossword are generated from this world's own lessons, so the review has a fixed slot rather than depending on remembering.
A Circle gives you a small group with a shared weekly goal, which adds a different kind of cue, since a group expecting something at a particular point is one of the more reliable triggers available.
What this does not solve
Three limits, stated plainly.
Ambivalence. Implementation intentions work much better for things you genuinely want than for things you think you should want. If part of you does not want the outcome, a more specific plan will not resolve that, and the unaddressed ambivalence will keep producing failures that look like planning problems.
Avoidance with a function. If the thing you are not doing is being avoided because approaching it produces anxiety, a plan specifying when to do it does not address the anxiety. The plan will be made and not executed, and the failure will be attributed to discipline.
Circumstances outside your control. Plans work for behaviours you can initiate. They do not work for outcomes that depend on other people, and framing those as personal follow-through failures is both inaccurate and demoralising.
And a boundary worth naming: where an inability to act is persistent, distressing, and accompanied by low mood, exhaustion or a sense of everything being too much, that may not be a planning problem at all. Depression in particular produces exactly this pattern and does not respond to better if-then statements. That is a matter for a doctor or a qualified professional rather than for a technique.
What to take from this
Intentions predict behaviour weakly, and most people who fail to act still want the outcome.
The gap is structural rather than motivational: intentions are formed in one state, executed in another, and specify outcomes rather than actions.
Implementation intentions close a meaningful part of it. When, where and how, attached to a cue you will physically encounter.
Add one coping plan for the most likely obstacle, because most failures happen at a small number of predictable points.
And treat the satisfaction of having decided with suspicion. It arrives before anything has changed, and it is the reason the same resolution can be made for years.
How specific does a plan need to be?
Specific enough that it names a cue you will physically encounter and an action you will take immediately on encountering it. If you cannot picture the exact moment, it is not specific enough yet.
Does this work for avoiding things as well as doing them?
Yes, and the same format applies: if a particular situation arises, I will do this alternative. Specifying a replacement behaviour tends to work better than specifying an absence, because an absence gives the moment nothing to trigger.
What if I keep missing my own plans?
Usually one of three things: the cue is not reliable enough, the action is too large for the moment, or there is ambivalence about the goal itself. Checking those in order is more productive than concluding you lack discipline.
Should I tell people my goals?
The evidence is mixed and the strong claims in both directions are overstated. What is clearer is that public announcement can supply some of the satisfaction of achievement, so it is worth noticing whether telling people feels like progress in itself.
Where can I learn this systematically?
The Subconscious patterns direction inside the Psychology & Mind world of Astra Trainer runs to fourteen courses on the patterns you repeat, including a sequence on habit loops and why promises to yourself break. Lessons take about five minutes and the first needs no card. You can see what is inside the world here.
