How long should a practice session be, and how often should you practice?

Three hours every Sunday, and by Wednesday it is gone

Picture a cello in the corner of a small living room. Its owner is 52 and took lessons as a girl. Last fall she started again.

Her weekdays are full, so Sunday is cello day. She practices for three hours straight. By late afternoon the new scale runs clean and her bow arm feels loose.

By Wednesday, when she picks it up for five minutes before bed, the scale is a mess again. Her fingers hover over the strings, unsure.

Nothing is wrong with her hands or her memory. The problem is the shape of her week. Three hours on one day and nothing on the other six is one of the weakest ways to spend the same time.

The typists who trained one hour a day

Popular advice is very precise about session length. You will see "practice for 20 minutes" or "45 minutes" as if a study proved it. The book looked, and could not find a study that supports any exact number of minutes for adult skill practice.

What the research has is a ceiling. In 1978, Baddeley and Longman trained four groups of postal workers to type code on a typewriter keyboard. The groups trained in sessions of one or two hours, once or twice a day.

Learning per hour of training was best in the group that did one one-hour session a day. It was worst in the group that did two two-hour sessions a day. Months later, typing speed had dropped by about 30 percent, and the heaviest group again did worst.

There was a cost. The one-hour group took 12 weeks to learn what the heaviest group learned in 3, and they resented it. Spreading out costs calendar time. It buys more learning per hour.

Other sources land in a similar range. Ericsson's 1993 violin study found practice sessions averaged 80 minutes, and cited work suggesting effective practice "may be closer to 1 hr per day." A 2025 surgical review recommended one session a day, at most 2 hours long.

So the best-supported unit is about an hour or less, about once a day. Anything shorter is a reasonable guess.

Spread the sessions, not the reps

How often matters even more than how long, and here the evidence is strong. Two meta-analyses found that spreading practice beats massing it. In 1999, Donovan and Radosevich pooled 63 studies and found an average effect of 0.46 in favor of spacing, a moderate size. The book notes it came out larger in less careful studies.

Trials with real skills agree. In 2006, Moulton and colleagues randomly assigned 38 junior surgical residents to learn to join tiny blood vessels in one day or across weekly sessions. Both groups got the same total time. One month later, the spread-out group did significantly better on most measures.

A 2025 review by Jørgensen and colleagues covered 28 studies of surgical training. Of the 19 that directly compared spread-out and massed training, 15 favored spreading. None favored massing.

Two limits keep this honest. First, it is about spreading sessions, not resting between every rep. In a 1989 study, longer rests helped a continuous task but not a quick one-shot move. Second, nobody knows the perfect gap. The surgical reviews lean toward short ones, and a 2023 review found back-to-back days seemed best. For a busy adult, that means most days rather than once a week.

A week plan you can start Monday

The Practice Lab turns this into two drills in Short Sessions, Spread Out.

The Practice Week Map is one of the book's five SOLID drills. It takes ten minutes once a week:

1. Write the total practice time you have in a normal week. Be honest.
2. Split it into short sessions on separate days. Four to six is the book's planning default, not a research number.
3. Put each one on a day and time, attached to something you already do, like after coffee.
4. Keep the gaps short. Back-to-back days beat one session a week.
5. At week's end, mark which sessions happened. Move a missed one to a spare day. Do not stack it.

The Short Block (PROMISING, with a STORY ONLY starting length) sets each sitting. Start with about 20 minutes and one target. Stop when the timer goes, even if it is going well. Keep the day to about an hour, and adjust the length after a week.

For our cellist, three Sunday hours might become five weekday blocks of 20 minutes and one shorter Sunday split into pieces. That is a little less time. It is spread so no gap is longer than a day.

The Practice Plans section at the back builds four month-long plans this way, for an instrument, a sport, a spoken language and drawing. Each uses 20 to 30 minute blocks, four to six times a week. The book says plainly that nobody has tested those exact plans. It also repeats its one safety rule: pain is not part of the method. Stop and ask a professional.

If your weekday gaps come from being unable to play at all, mental practice can fill some of them.