How to build a practice routine that survives a busy week

A practice routine survives a busy week when it has a short version. Build a session from four blocks, decide what each one is for, then write down a twenty minute version you can run on the worst day of the month. That version is the routine.

Why most practice does not stick

Two failure modes account for most abandoned routines, and neither is about willpower.

Repetition without a target

The common pattern: start the piece from the top, play until something goes wrong, stop, start again from the top. An hour later you have played the opening twenty times and the difficult bar four times, badly.

Repetition trains whatever you repeat. Playing a passage wrong ten times and right once teaches your hands the wrong version, because that is the version they rehearsed. The fix is to name the target before you play. Not "work on the tune", but "bars 17 to 20, clean, at 80".

That is the difference between practicing and playing. Both are worth doing. Only one of them changes what you can do next month, and Gerald Klickstein's book is the best single treatment of the distinction. We cover it in our review of The Musician's Way.

Playing what you can already play

The second failure mode feels better than the first. You sit down, play the things that sound good, enjoy an hour, and move nothing.

It is worth being clear-eyed about this, because it is the most common form of practice there is. The comfortable hour is not wasted, it keeps you playing, and it is not practice. If your session contains no passage you cannot yet do, you had a good time and you are the same player you were on Monday.

Kenny Werner's work is useful on why the discomfort of the second kind is hard to sit with, and we cover it in our review of Effortless Mastery.

Build the session before you build the habit

Do not start by deciding to practice every day. Start by deciding what a session contains. A session with a shape is easier to start than an open-ended hour.

The four blocks

Warm-up. Five to ten minutes. Low demand, physical, no new information. The point is to get your body working, not to accomplish anything.

Technique. Ten to fifteen minutes. Isolated mechanics: whatever the fundamentals are on your instrument or voice. This block is where the click does most of its work.

The one hard thing. Fifteen to twenty minutes. One passage, one problem, one target. This goes third, not last, because it needs the attention you still have. If it goes at the end of the session it gets the tired version of you, or it gets skipped when the session runs short.

Repertoire. Whatever time is left. Playing through material, working on the shape of a whole piece, and this is the block where playing for enjoyment is legitimate.

Four blocks, in that order. The hard thing sits in the middle where the attention is.

How long a block should be

A full session on a good day runs 45 to 60 minutes. Beyond about an hour, the returns drop off and the risk of practicing while inattentive goes up, which puts you back in failure mode one.

Now write the short version, and write it before you need it. Twenty minutes: three minutes of warm-up, five of technique, ten on the one hard thing, two of playing something you like so you leave the instrument in a good mood. That is the version that keeps the routine alive across a bad week.

Shorter and more frequent beats one long session, at least for the parts of practice that behave like learning. UC San Diego's psychology department summarizes the evidence plainly: spacing learning out over time lets you learn more and retain it longer than a single marathon session, across a body of work they describe as over 200 studies.

The honest caveat is that music is not only learning. A PLOS ONE study of piano learning found no spacing effect for accuracy, timing consistency or pressure consistency, and its authors concluded the effect may not always be demonstrable for complex motor skills. So treat frequency as the safer default rather than a guarantee. Four twenty minute sessions across a week are easier to actually do than one ninety minute session, and that is reason enough.

Practicing to a click

This is the section that does the work. Four drills, and the last two are the ones most players have never run.

Start under tempo

Under tempo means slow enough that the passage is clean the first time, not slow enough that it is nearly clean. That distinction is the whole drill.

Find the tempo where you play the passage correctly three times in a row. Not two. If you cannot get three, you are not under tempo yet, so drop it further. Players consistently start too fast, because the tempo where a passage is genuinely easy feels insultingly slow.

Then hold that tempo until three correct passes stop being an achievement. Only then do you move.

The speed trainer

Raise the tempo in increments small enough that the passage does not break. In practice that means a few beats per minute at a time, not ten.

A jump of ten beats is where the previous week goes. The passage stops being clean, you play it wrong a few times to get through it, and now you are training the wrong version at a tempo you cannot hold. Then you drop back and find the old tempo is worse than it was.

Small increments feel slow and are not. Two beats per minute a day is sixty beats in a month, which is most of the distance from a study tempo to a performance tempo on almost any passage.

The rule that makes it work: you go up when the passage is clean, and you go back down when it is not. Both directions, every session, without treating the drop as a failure.

Silent bars

The drill almost nobody runs, and the most useful one here.

Set the click and play a phrase. Then mute the click for one bar and keep playing, with the click returning on the next downbeat. If you are still with it when it comes back, your time held. If you are early or late, you now know exactly how much, and in which direction.

Extend it. Two bars silent, then four. A click on every downbeat only. Eventually a click on the first beat of every other bar.

What this exposes is whether you are keeping time or following it. A player who is following the click sounds locked in until the click stops. A player who is keeping time stays where they were, which is what playing with other people actually requires.

Subdivisions, accents and odd time

Move the click off the obvious beats and different problems appear.

Put the click on beats two and four instead of all four. Suddenly you are responsible for one and three, and any tendency to rush shows up immediately.

Put it on the offbeats only. This is uncomfortable at first and the discomfort is the information: it means your internal subdivision was leaning on the downbeat.

For odd meters, set the click to the grouping rather than the bar. A bar of seven counted as 3+2+2 needs a click that agrees with the grouping, or you are fighting the tool.

And set the click slower than the music. One click per bar, or per two bars, at a moderate tempo. It gives you a checkpoint rather than a scaffold, and it is the closest thing to playing with a drummer who has good time and does not talk.

Practicing with other people

A shared click and the rehearsal that stays honest

Band rehearsal is a different discipline from individual practice. Individual practice is for fixing things. Rehearsal is for finding out what happens when four people play at once, and it is the wrong place to learn your part.

The argument about who is rushing is the standard rehearsal problem and it is usually unresolvable by discussion, because everyone hears themselves as steady. A click resolves it in about thirty seconds. Run the section to a click through the monitors and the answer becomes a fact rather than an opinion, which takes the personal charge out of it.

Two rules make this work. Use it to diagnose, then turn it off and play the section again, because a band that can only hold together over a click has not fixed the problem. And agree beforehand that the click is checking everyone, not the person somebody suspects.

Attention during a session is its own skill, and Barry Green's book is the standard treatment. See our review of The Inner Game of Music.

Tracking whether any of it is working

The tempo log

Write down three things at the end of each session: the passage, the tempo, the date. Nothing else.

Six weeks of that tells you something memory cannot. Memory says the hard passage is not improving, because it still feels hard. The log says you were at 72 in March and you are at 104 now. It also catches the opposite case, where a passage has sat at the same tempo for a month and needs a different approach rather than more repetitions.

Keep it in whatever you already use. A note on your phone works. The value is in the dates, not the format.

Learning a record at its real tempo

If you are learning something off a recording, get the actual tempo before you practice it. Guessing puts you at the wrong number, and a passage rehearsed at 118 does not arrive ready for a record that sits at 126.

The key is worth having too, for transposition and for playing along. Our guide to finding the key and BPM of a record covers the by-ear method and the faster ones.

Then practice to that number rather than to the recording. Playing along with a record hides your timing inside the mix. A click does not.

When to stop practicing and start performing

There is a point where more repetition stops paying. It arrives when the passage is reliable at tempo across several sessions, on different days, without a warm-up run.

After that, the remaining problems are not technical. They are about playing the thing in a room with people in it, which is a different skill that practice alone does not build. At that point the work moves to run-throughs, to playing it for someone, and to performance preparation, which is its own subject. We cover that in nailing your performances.

Frequently asked questions

How long should I practice each day?

Long enough to run the four blocks with attention, which is usually 45 to 60 minutes. On a bad day, run the twenty minute version instead of skipping. Consistency across a week matters more than length on any one day.

Is it better to practice every day or in longer sessions?

Shorter and more frequent is the safer default. Research on spaced practice supports it for learning generally, though a piano study found no spacing benefit for the motor-skill side, so the evidence is not uniform. The practical argument holds either way: four short sessions are easier to actually complete than one long one.

Should I always use a metronome?

No. Use it for isolating problems, for the speed trainer, and for silent-bar work. Turn it off for repertoire playing and for anything with rubato or a deliberately loose feel, because a click will flatten phrasing that is supposed to move.

Why do I speed up when I play alone?

Usually because a passage is slightly beyond your comfortable tempo, and rushing is what tension does. Silent bars will show you where it starts. So will setting the click to one beat per bar and hearing yourself arrive early.

How do I practice without an instrument to hand?

Work on the parts that are not physical. Count a passage out loud against a click. Read through the material away from the instrument and find the sections you cannot hear in your head. Rhythm work on a table top with a click is a real practice block, not a substitute for one.

What tempo should I start a new piece at?

The tempo where you play it correctly three times in a row, which is usually slower than you expect. If three correct passes are not happening, you have not gone slow enough yet.

Does practicing slowly really make you faster?

Slow practice is a method for getting a passage correct before it becomes automatic, which is the reason it is standard advice. It is not the only route, and some teachers argue for practicing closer to tempo in short bursts. Both work in different situations, and the reliable rule is that you should be able to play the thing correctly at whatever tempo you are using.

How do I restart after months off?

Rebuild the session shape first, not the hours. Run the twenty minute version for a week before you try for a full session. Expect the tempo log to start below where it ended, which is normal and not worth reacting to.

The routine is what gets the work done. Getting the work heard is a separate problem, and Unchained delivers to 150+ platforms when you have something ready. If reading is more your thing this week, our reading list for musicians covers the books behind most of the methods above, and the distribution plans are there when the recording is done.