How Many Pieces Can You Learn at Once Without Leaving Them All Half-Finished?

Building a Piano Piece · Season Two: From One Piece to Long-Term Progress · Article Seven

One piece can allow a single kind of problem to occupy every session. Seven or eight pieces can lead to a little daily contact with each and no completion after several weeks. The apparent question is “What is the maximum number?”, but count alone is not decisive.

Three large sonatas and three short dances are not equivalent. One piece being decoded, one prepared for performance and one maintained are not equivalent to three works beginning at zero. The useful management unit is not the number of titles. It is the role, current stage and next required result of each work.

Parallel repertoire can provide variation, delay and tests of transfer. Without completion rules, it can also create a life of perpetual novelty in which finishing is never faced.

Why interleaving deserves attention but not worship

Motor-learning research often compares blocked and interleaved practice. Blocked practice remains on one task before changing; interleaving alternates among tasks. Switching can require retrieval and reorganisation. It may feel less fluent during practice yet sometimes assist later retention.

Direct evidence in applied music remains limited. One study asked ten advanced clarinettists to practise concerto and technical excerpts. In one condition each item received a continuous twelve minutes; in the other, players changed every three minutes. Immediate and next-day performances were rated by three professionals. Some ratings favoured interleaving while many comparisons were not statistically significant. Participants also reported benefits for goal setting, concentration and error identification.

This small non-piano experiment is suggestive, not a rule that every pianist should change pieces at three minutes or that ten simultaneous works are better than two. A responsible application is narrower: once a problem has been identified and continuous repetition is producing little new information, leaving and later reconstructing the task can create a retrieval test.

Give every piece a role

A balanced portfolio can contain four roles.

Growth repertoire. The principal learning project, introducing one or two new capacities and receiving most diagnostic and integration time. One or two are often enough.

Fluency repertoire. Shorter or less demanding works used for timely completion, musical continuity and independent decisions. Easier must not mean careless.

Maintenance repertoire. Completed works retained through spaced cold starts or full runs rather than daily reconstruction.

Reading and exploration. Material read once or twice to broaden stylistic and pattern vocabulary, with no promise of completion.

A work can change role. Growth repertoire becomes maintenance; an excessive challenge becomes a study fragment; exploratory reading enters the fluency list because interest and suitability converge. Role is a current resource decision, not an identity.

Place a limit on work in progress

Large collections of half-finished pieces usually arise because no limit is placed on active acquisition. Every work simultaneously demands decoding, fingering, technical correction and structural learning, producing heavy switching costs.

As a starting rule, keep only one or two pieces in high-intensity growth, one in timely completion, several completed pieces in low-frequency maintenance, and exclude reading material from the completion queue. This is not a universal optimum. It prevents every piece from occupying the highest cognitive-load stage.

One growth piece may already be enough for someone with little daily time. A full-time student with shorter works and regular teaching can manage more. The functional test is whether every growth work receives at least one complete diagnosis–correction–delayed test–integration cycle each week. Without that cycle, the piece is nominally active but practically only visited.

Let stages, not mood, select today’s work

Assign each work a stage:

1. Map: edition, structure, principal demands and preliminary sound.
2. Build: notation, fingering, movement and voice relations.
3. Integrate: enlargement, seams and movement near performance tempo.
4. Perform: full runs, recordings, varied entries and contextual tests.
5. Maintain: delayed retrieval and necessary repair.
6. Dormant: no current resources, but records retained.

Each stage needs a passable gate. “Bars 34–41 can begin from three entries with stable pulse in both hands” is more useful than “continue the second movement”. Move the work when the gate is passed. If it remains closed for weeks, reduce the task, obtain teaching, lower difficulty or make the work dormant.

The system therefore permits ending an unsuitable project. Completion need not mean public performance. Evidence of a missing prerequisite, recorded and acted upon, is a successful management decision rather than a failure of character.

Designing a week

Suppose a learner practises five days for about forty minutes, with growth works A and B, fluency work C and two maintenance pieces.

Monday can cold-test A, repair it, leave for C, then return to A for delayed retrieval. Tuesday centres B and ends with one full maintenance performance. Wednesday integrates A and includes sight-reading. Thursday centres B and briefly tests C. Friday records a run of A or B, chosen according to the information most needed; the other receives only a cold start.

The calendar is not a prescription. Switching has a function: leaving supplies delay, another task changes demand, and return tests retrieval. Do not leave mechanically just as a problem becomes clear because three minutes have passed. Do not continue twenty more repetitions after a passage has become immediately fluent and call familiarity retention.

A repertoire board that makes completion visible

Use a compact table:

Update once a week. Record-keeping should not become a substitute activity. The board answers: Which piece is receiving resources? Which lacks a next action? Which should already have moved to performance, maintenance or dormancy?

Four rules against perpetual beginnings

When a new work enters, another changes state. Adding a growth project sends another to performance, maintenance or dormancy. Work-in-progress cannot expand indefinitely.

Give fluency repertoire a deadline. Two to four weeks for an acceptable recording can train decisions, proportion and closure without demanding perfection.

Require one single-take result each week. Perform a complete piece or current complete section once and record interruption and recovery. Exclusive local polishing conceals integration cost.

Permit intentional stopping. Record the reason and the future condition. “Retest after similar spans become comfortable” preserves more learning than silent disappearance.

The real benefit of several simultaneous works

Parallel repertoire can test the same capacity in different settings: establish voices in Bach and foreground–background balance in Schumann; adjust relocation in a scale and test it in Mozart; practise timely completion in an easier work and diagnosis in a challenge.

It can also vary attentional and physical demand. Moving from high-speed repetition to score study or slow voicing is not a medical prevention formula, but it makes cumulative load easier to observe than prolonged repetition of one action. Pain and persistent symptoms still require stopping the provoking activity and seeking professional assessment.

The right number is consequently dynamic. If every role is clear, growth projects receive feedback cycles and works move towards completion or dormancy, the portfolio may be manageable. When every piece remains at “played occasionally, no next gate”, there are already too many.

Repertoire management is not equal distribution of minutes. It is a limit on high-load work in progress and a requirement that every piece acquire a new, explicit state.

Primary sources and further reading

• Experiment comparing blocked and interleaved practice in advanced music learning
• Self-regulated practice strategies among tertiary piano students
• Temporal organisation and self-regulation in conservatoire exam preparation
• Music-practice microanalysis and self-regulation protocol

Continue reading: Building a Piano Piece series page

If you would like to bring these ideas about listening, understanding, and practice to the keyboard, you might try ScoreFlow, an app I developed to make score reading and daily practice flow more naturally together.