Using Action Planning to enhance Deliberate Practice

What each one is actually for

Action Planning

You know exactly what you would work on if you sat down, and most evenings you do not sit down. That specific failure — not confusion about the content of the session, but the twenty feet between the sofa and the instrument — is what Action Planning addresses, and it is worth being clear that it addresses nothing else. The tool is the implementation intention, a form of plan studied for decades by Peter Gollwitzer and colleagues, and what distinguishes it from an ordinary intention is purely its grammar. An ordinary intention names a goal: I want to practise more this month. An implementation intention names a situation and a response and welds them together into one sentence: when I put my coffee cup in the sink after breakfast, I will open the score to bar forty-one. The claim about why that grammar matters is mechanistic and testable rather than motivational. Naming a concrete situation makes that situation more likely to be noticed when it actually occurs, and pairing it with a single specified response means that noticing it is enough to set the action going, without a stage in between where you weigh whether now is a good time. Weighing is where intentions leak. You do not usually decide against practising; you arrive at the moment, find the decision still open, and an open decision loses to whatever is easier. Writing one is undramatic and takes about ninety seconds. You pick a cue that reliably happens and that you cannot miss, which is why a physical event in a particular place beats a time of day, and beats a feeling by a long way. You pick one response, and you make it small enough that it is not itself daunting — opening the score, not completing the session. You write it out as a full sentence rather than holding it in your head, and rehearse it once or twice so the pairing is actually laid down rather than merely agreed to. Then you stop thinking about it, which is the entire point: the design moves the behaviour off the list of things you have to remember to want. Note also what it does not do. It is entirely silent, by design, about what actually happens in the hour after bar forty-one is open. It will get you onto the bench with exactly the same reliability whether the hour that follows is structured work on your worst four bars or a pleasant play-through of the piece you can already perform.

Deliberate Practice

Deliberate practice describes a particular way of spending practice time, distinguished sharply from just doing the activity a lot. Three features define it. First, it targets a specific identified weakness rather than the whole skill, so a violinist does not play the piece through; they take the four bars where the shift goes wrong and work only on those. Second, it operates at the edge of current ability, hard enough that you fail a meaningful proportion of attempts, because within the range you already have you are rehearsing rather than developing. Third, it depends on feedback arriving immediately and specifically enough to correct the next attempt, which usually means either an external source, a teacher, a recording, a measurable result, or a task designed so the failure is unmistakable. The reason this is not obvious is that ordinary practice feels productive and is not. Playing through the whole piece is pleasant, produces a sense of progress, and mainly reinforces what you already do, including the mistakes. Ericsson's contribution was to look at what elite performers actually did in their practice hours and find that the hours themselves were not the story; the structure of the hours was. The design has a cost that the popular version of the idea tends to skip over: because you spend the session on the thing you are worst at, and because you are failing repeatedly by construction, deliberate practice is aversive. Not difficult in an inspiring way. Genuinely unpleasant, at the time, most days.

What you get out of pairing them

The gain from putting these together is concrete and shows up in about a week. Take the ordinary version of trying to get better at something. You know roughly what good practice looks like, and you have read enough to know you should work at the edge and seek feedback. The evening arrives and you are tired, and one of two things happens. Either you do not start, and the reason is never that you decided not to, it is that the evening dissolved and by nine it was too late. Or you start, and within four minutes you have drifted to the parts you can already do, because those parts are the ones that feel good and you did not arrive with anything more specific than practise. Both of those are failures of specification rather than failures of will, and both have a fix that costs ninety seconds on a Sunday. What the pairing buys you is that neither the starting nor the choosing consumes anything on the night. You arrive at the sink, or the door, or whatever your cue is, and the next action is already determined, and it is determined down to the bar number or the specific move, so there is no drift stage either. That leaves the whole of your available effort for the actual work, which is the only place effort produces skill. And there is a second gain that is less obvious and probably larger over months: when the plan names something specific enough to succeed or fail at, you get a readable signal about whether the difficulty was right. Practise the piano tells you nothing at the end of the evening. Play bars forty-one to forty-four at sixty beats per minute until five in a row are clean tells you either that you did it, or that sixty was too fast, and both of those are usable on Wednesday.

What Action Planning rests on

• Implementation intention closes intention action gap randomised trials
  Pre-committing to a specific situational cue paired with a specific response lets the behavior fire automatically when the cue occurs, without depending on in-the-moment willpower.
• If then planning offloads self control plausible mechanism
  Linking a response to a specific situational cue shifts behavior initiation from effortful self-control to automatic cue-triggered action.

What Deliberate Practice rests on

• Small wins compound randomised trials
  Repeating a small consistent action accumulates over time into a result too large to produce in a single step.
• Desirable difficulty drives skill growth observational studies
  Practicing at a level just beyond current ability rather than within already-mastered territory drives skill improvement.
• Immediate feedback accelerates learning randomised trials
  Feedback delivered immediately after an attempt rather than delayed allows faster correction and more effective learning.

Where they actually meet

The same move appears in both, aimed at two different moments. If-then planning takes the question should I start now, which is answered badly under fatigue, and removes it by answering it in advance under better conditions. Deliberate practice takes the question what should I work on, which is answered badly once you are already in the room and drawn toward what you can do, and removes it by answering it in advance too, using an analysis of where your skill actually breaks. Both are, at root, the same manoeuvre: identify a decision that will be made badly at the moment it arises, and shift it earlier, to a moment when the person making it is not under the pressure that would distort it. That is the connection at the level of mechanism, and it explains why they slot together rather than merely coexist. It also explains something that is otherwise puzzling about each of them in isolation. If-then plans reliably get people started and the literature is oddly quiet about what happens next, which makes sense once you see the plan as a device for removing one decision and only one. Deliberate practice has an enormous literature about the structure of the session and almost nothing about attendance, which makes sense once you see that its authors were studying people, conservatory students and elite athletes, whose attendance was already guaranteed by an institution. Each method has a blind spot exactly where the other was looking, and the blind spots are not accidental; they are what each one was not built for.

The principle underneath both

Named plainly: deciding under pressure produces worse decisions than deciding the same question in advance, so the useful work is moving the decision rather than trying harder at it. Neither of these traditions owns that. It is the same principle behind a shopping list written after a meal rather than before, behind a chess player's opening preparation, behind a doctor's standing orders, and behind the checklist a pilot runs rather than reasoning from first principles at four hundred feet. In each case somebody noticed that a particular decision would recur under conditions that reliably distort it, and moved the decision to a calmer occasion where it could be made once and well. The evidence for the general version is unusually good in one specific form. Gollwitzer and Sheeran's meta-analysis, pooling ninety-four independent tests, found that forming an if-then plan produced a medium-to-large improvement in goal attainment over holding the same goal without one, and the effect held across health behaviours, academic tasks and interpersonal goals. It is worth being exact about what that shows and does not: it shows that specifying situation and response in advance changes whether an intended behaviour happens. It says nothing about whether the behaviour was any good, which is precisely why the second half of this pairing has to come from somewhere else.

What that makes the real goal

If moving decisions out of the pressured moment is what does the work, the goal separates from both traditions and can be stated in one line: arrange it so that on the night, the hard specific thing is what happens by default, and neither showing up nor choosing what to do costs you anything you need for the work itself. That framing is more useful than either method's own framing because it tells you what to do when the standard advice does not apply. If you are already showing up reliably and the sessions are unfocused, the whole of your effort goes into specification and none into planning cues. If you are focused but sporadic, the reverse. If you cannot yet tell what you are worst at, which is the ordinary situation for a genuine beginner, then the first target is not skill at all; it is getting enough attempts in front of someone who can see what you cannot, so that a weakness can be identified in the first place. The goal also makes clear what is not required. It does not require an hour, or a teacher, or the word deliberate. It requires a cue you cannot miss, a target narrow enough to fail at, and some source of feedback that arrives while you can still act on it.

Other practices that serve the same goal

Several other practices head for the same goal, with different evidence behind them. Mental contrasting with implementation intentions, packaged by Gabriele Oettingen as WOOP, for wish, outcome, obstacle, plan, adds a step before the if-then sentence: you picture the outcome you want, then deliberately identify the concrete internal obstacle that has actually stopped you before, and only then write the if-then plan aimed at that obstacle rather than at the activity. It has randomised support across health and academic domains, and it is the better choice when you already know your problem is not forgetting but something more specific, like dread when you open the instrument case. Habit stacking, attaching a new behaviour to the end of an established routine, is the popular-book cousin of if-then planning; it uses the same mechanism with a rule about cue selection and rests on field reports rather than trials. On the practice-quality side, the most effective route is embarrassingly old and rarely taken: get a teacher. Ericsson was consistent that a coach who can see errors you cannot is the ordinary way the feedback requirement gets met, and self-supplied feedback is the weakest component of solo practice by a wide margin. A cheap substitute with real value is recording yourself, because a recording delivers the information your ears filter out live. And spaced retrieval practice, testing yourself at increasing intervals rather than reviewing, is worth naming for anything with a knowledge component; it has one of the strongest experimental bases in learning research and it is a different lever from either method here.

A tradition that already does this

The conservatory got here first, by about a century, and it fused both halves without naming either. The instruction given to serious music students, in the tradition that runs through Leopold Auer and Ivan Galamian and into most conservatory studios today, is that you do not enter the practice room without a written practice plan, and the plan specifies the passage, the tempo, the number of correct repetitions, and the order. That document is an if-then plan whose second half is a deliberate practice prescription: when I sit down at ten, I will play bars forty-one to forty-four at sixty until five consecutive attempts are clean. Galamian's writing on practice is explicit that the value lies in the specificity, and that a student who practises for three hours without a plan will get less from it than one who practises for forty minutes with one. The Suzuki tradition arrived at the same structure from a different direction with the one-point lesson, where a single narrow correction is named at the end of a lesson and is the only thing worked on until the next. What makes this a genuine precedent rather than a nice parallel is that the studios were solving both problems at once and knew they were: the plan existed because students drifted to what they could already play, and it was written the day before because a tired student at ten in the morning would not write a good one.

Making the if-then sentence name the weakness, not the activity

The direct move here is to rewrite your plan so the second half is a deliberate practice prescription rather than an activity. Most people write when I get home from work, I will practise for an hour, and that plan does its job perfectly and then delivers an hour of playing what you already play. The upgrade is to make the response as specific as a target you could fail to hit. Compare: when I sit down after dinner, I will practise Spanish for thirty minutes, against, when I sit down after dinner, I will run twenty flashcards of the verbs I got wrong last session, out loud, checking each one before moving on. The second is a plan and a session design at once. Three things make the difference. Name the specific weakness, which means you need a record of what you got wrong, so the plan has to include how the weakness gets identified as well as what to do about it. Name the difficulty level explicitly, a tempo, a weight, a time limit, a number of items, because the edge of your ability is a moving target and unnamed difficulty drifts downward toward comfort within about four minutes. And name where the feedback comes from before you start, since the most common way a well-planned session degrades is that you do the hard thing and have no way of knowing whether you did it right, which turns effortful practice into effortful guessing. There is a second, less obvious use of if-then planning inside the session rather than around it, and it addresses the actual reason people drift. Write a plan for the drift itself: when I notice I have started playing the passage I like, I will stop and go back to bar forty-one. That is the same grammar pointed at an internal event rather than an external one, which is the form Oettingen's work tends to use, and it is the one that survives contact with an actual Tuesday evening.

Using practice design to fix the plan that keeps not being kept

Run it the other direction and deliberate practice supplies a diagnosis that if-then planning does not have. When a plan fails repeatedly, the standard advice is to make the cue more concrete or the action smaller. Sometimes that is right. But there is a different cause that the planning literature has no vocabulary for: the plan is failing because the thing you have planned is the wrong difficulty, and your nervous system has correctly noticed. A plan pointed at something far above your current level generates dread, and dread reliably beats a well-specified cue. A plan pointed at something you can already do generates boredom, which is a slower and more insidious killer because it lets you keep the plan while quietly hollowing out the session. So the practice-design question, is this at the edge of what I can just about do, is a diagnostic to run on any plan you keep breaking, before you go about redesigning the cue. Concretely: for the next week, at the end of each attempted session, write one number, the proportion of attempts you got right. If it is near one, the plan is too easy and the fix is to raise the tempo, the weight, the complexity or the speed until you are failing a meaningful fraction. If it is near zero, break the target smaller, since two bars is a legitimate target and so is one. The range where skill moves is somewhere in between, and it is uncomfortable by design, which is worth knowing in advance so that the discomfort does not read as evidence you have chosen wrong. The second thing deliberate practice lends is the review cadence. An if-then plan is written once and then run until it stops working, and because nothing in the method looks back, it usually keeps running for weeks after the target has been outgrown. Borrow the practice tradition's habit instead: at a fixed interval, once a week is enough, look at what you got wrong and rewrite the second half of the sentence to point at that. The cue stays. Only the target moves.

Twenty past nine on a Tuesday, and you are playing the easy bit again

It is twenty past nine and you have been at the keyboard for eleven minutes. You are playing the second page, which you like, which you can already play, and which you have now played four times tonight. The plan worked: you wrote it on Sunday, when I put the last plate away, I will sit down at the keyboard, and the plate went away at nine and here you are, so the hard part supposedly went fine. And yet what is actually happening is that you are practising something you finished learning three weeks ago, and you will finish at ten feeling like you did an hour, and next Tuesday the shift in bar forty-three will still be wrong, as it has been since March. Here is the moment worth catching. You did not choose this. There was no point at which you thought, I will avoid bar forty-three tonight. You sat down, the plan had specified sitting down and nothing else, the decision about what to play was still open, and it went the way open decisions go at nine in the evening. That is not a character finding. It is the predictable output of a plan that stopped one word too early. So on Sunday you write a different sentence. When I put the last plate away, I will play bars forty-one to forty-four at sixty on the metronome until five in a row are clean, and only then anything else. And you put the metronome on the music stand on Sunday night, so the setup is not a decision either. Next Tuesday it is unpleasant. You get three clean, then a bad one, and start again, and it takes nineteen minutes to get five, and you do not enjoy any of it. But at nine forty you have a passage that is measurably different from the one you had in March, and you have a number, nineteen minutes, that will tell you something on the following Tuesday.

When this is worth reaching for

You'll know it's the week for this when the recording from tonight sounds like the recording from March — the same shift wrong in the same four bars — despite showing up on schedule for a year or more. A flat curve under reliable attendance is almost always a specification problem rather than an effort problem, and it only pays off once you can name the specific thing you're worst at and have some way — a teacher, a recording, a number you write down — of telling whether an attempt actually landed.

Where this does not apply

There are conditions where tightening the specification is the wrong move. The first is genuine beginnerhood. If you have been doing something for three weeks, you do not yet know what your weaknesses are, and a plan pointed at an assumed weakness will make you efficient at practising something that was never the constraint. Early on, breadth and volume and enjoyment do more, and the honest advice is to do a lot of the thing badly and get in front of someone who can see what you cannot. The second is when you have no feedback source at all. A difficult target with no way of knowing whether you hit it correctly is a good way to install an error deeply, and for physical skills in particular that is expensive to undo later. The third is exhaustion, and it deserves more weight than it usually gets. Deliberate practice is demanding by construction and the studies of elite performers consistently report a ceiling of a few hours a day even among people whose whole life is arranged around it. If you are running on four hours of sleep with a young child in the house, a plan that specifies the hardest available thing will simply not be executed, and the failure will be read as a character problem when it is a capacity problem. A less ambitious plan you actually keep is worth more than a correct one you abandon. Fourth, and against the grain of both traditions: some pursuits are not skill-acquisition projects, and treating them as such removes the reason you were doing them. If you play in the evening because it is the only hour of the day nobody wants anything from you, converting it into structured drilling may improve your playing and cost you the hour.

What the evidence does and does not support

The two halves of this page have quite different evidential standing, and the difference runs opposite to their popular reputations. If-then planning is the better-established of the two for the thing we are using it for. Gollwitzer and Sheeran's meta-analysis across ninety-four independent tests found a medium-to-large effect on goal attainment, the effect has been replicated across health, academic and social domains, and the mechanism, that specifying situation and response in advance removes an in-the-moment decision, has direct experimental support. Deliberate practice is well supported for a narrower claim than the one it is famous for. That structured practice with a specific target and immediate feedback beats unstructured repetition within a domain is not seriously disputed. That accumulated practice explains most of the variance in expert performance is disputed, sharply: a 2014 meta-analysis by Macnamara, Hambrick and Oswald found deliberate practice accounted for roughly a quarter of the variance in games, a fifth in music, less in sport, and very little in professions, which is a long way from the popular ten-thousand-hours reading and prompted an extended exchange in the literature that has not fully settled. So when this page recommends the practice design, it is leaning on the part that holds. The combination itself is untested. The specific comparison worth running would be simple to describe and awkward to execute: give two groups the same if-then cue, and have one group's plan name an activity while the other's names a specific weakness at a specified difficulty with a named feedback source, then measure skill on a blind assessment after eight weeks rather than measuring how much they practised. Almost every existing study measures the second thing, because it is easy, and it is the wrong thing.