What the data actually shows
The clearest statement of the problem comes from work by Paschal Sheeran and Thomas Webb on what they call the intention-behaviour gap. Reviewing the evidence, they conclude that intentions account for only a modest share of the variation in whether people actually act, and that a large proportion of people who form a genuine intention — to exercise, to study, to follow through on a goal — nonetheless fail to carry it out. Intending is a weak predictor of doing.
One intervention stands out for closing part of the gap: implementation intentions, studied extensively by Peter Gollwitzer. These are specific 'if-then' plans that link a concrete situation to a concrete response — 'if it is 7am and I have finished breakfast, then I will put on my shoes and walk' — rather than a vague goal to be more active. A meta-analysis by Gollwitzer and Sheeran (2006) found that forming such plans had a medium-to-large effect on goal attainment across many studies, well beyond merely holding the intention.
A broader theme across this literature is that situation tends to beat willpower. People who appear to have strong self-control are often, on closer inspection, people who have arranged their lives so that fewer acts of self-control are required — temptations are out of reach, cues for the wanted behaviour are everywhere. The behaviour that happens is heavily shaped by the immediate environment, not just the strength of the resolve.
Why this feels different from how it actually is
It feels like intention should be enough because, from the inside, the decision feels like the hard part. Once you have sincerely resolved to do something, it seems as though the doing should follow automatically — so when it doesn't, it reads as a failure of character rather than a predictable gap that almost everyone experiences.
We also remember our intentions vividly and our inaction vaguely. The moment of deciding to change is emotionally charged and easy to recall; the hundred small moments where the plan quietly didn't happen are diffuse and forgettable. So the felt strength of the intention stays high in memory while the follow-through erodes unnoticed.
And the culture frames this almost entirely as willpower. Most advice tells people to want it more, to be more disciplined — which locates the problem in motivation. The research locates much of it in planning and environment, which is why trying harder to want something often changes so little.
The plan and the situation do much of the work that motivation is usually asked to do.
What the research says to do about it
The best-supported single tactic is to convert vague intentions into specific if-then plans. Decide in advance the exact cue — a time, a place, a preceding event — and the exact action that will follow it. This pre-commitment means that when the moment arrives you are executing a decision already made rather than deciding again under pressure, which is where intentions usually collapse.
Shaping the situation does much of the rest. Making the intended action easier and more obvious — laying out the gear, removing the friction, placing the cue in your path — and making the competing behaviour harder reliably shifts what actually happens. The research consistently favours changing the environment over relying on in-the-moment resolve.
Anticipating obstacles in advance also helps. A related technique pairs a clear picture of the desired outcome with an honest look at the specific obstacle likely to derail it, then attaches an if-then plan to that obstacle. Naming what usually goes wrong, before it goes wrong, lets you pre-load a response instead of being surprised by the same barrier each time.
What the research says does not help
Simply intending harder rarely closes the gap. Because intention is a weak predictor of action, strengthening the resolve without changing the plan or the situation tends to produce a stronger feeling and roughly the same behaviour. The repeated experience of 'this time I really mean it' followed by no change is the gap in action.
Relying on motivation or willpower to carry you through is fragile, because both fluctuate and deplete across a day. Systems that depend on feeling motivated at the right moment fail precisely when motivation is lowest — which is often exactly when the action is needed.
Vague goals — 'be healthier', 'study more', 'be more present' — are particularly weak, because they specify no cue and no concrete action, leaving every instance to be negotiated in the moment. The evidence strongly favours specificity over good intentions stated in general terms.
People who appear to have strong self-control are often just people who arranged their lives so fewer acts of self-control are required.
What this looks like in real life
'This time I really mean it'
Someone resolves, again, to start exercising — and feels the resolve strongly. Weeks later nothing has changed. That isn't weak character; it's the intention-behaviour gap in action. Strengthening the resolve without changing the plan or the surroundings tends to produce a stronger feeling and roughly the same behaviour.
An if-then plan instead of a vague goal
Rather than 'I'll be more active,' the plan becomes 'if it's 7am and I've finished breakfast, then I'll put on my shoes and walk.' When the cue arrives, the decision is already made — you're executing, not negotiating under pressure. Laying the shoes by the door the night before lets the situation carry part of the load that willpower usually gets asked to bear alone.
Real numbers in context
In Sheeran and Webb's reviews, intentions explain only a modest fraction of the variance in actual behaviour — and a substantial share of people who form a genuine intention do not act on it. The exact figures vary by behaviour and study and should be read as a consistent pattern rather than fixed constants: intending is real but is a weak guide to doing.
On the other side, the meta-analysis of implementation intentions by Gollwitzer and Sheeran (2006) found a medium-to-large average effect on goal attainment across a large number of studies. That is a sizeable improvement for a technique that costs nothing but a moment of specific planning — though, as with all averages, individual results vary and the plans help most when the goal is something you genuinely want to do.