The Practice Tracker as a Memory Aid: How to Stop Forgetting Pieces You Learned

You don't forget the pieces you spent months learning because your memory is bad. You forget them because nothing is keeping score. The Chopin waltz you had by heart last spring, the jazz tune you could pull out at any party, the Bach you played for your recital — they didn't leave because you're incapable of holding them. They left quietly, while you weren't looking, because no one was watching them go.

That distinction matters more than it sounds, because it splits one problem into two, and they have completely different fixes. There's the forgetting itself — a piece fading from your hands over weeks of neglect — which is normal, predictable, and surprisingly cheap to reverse. And then there's the forgetting you never notice: the fact that you feel like you still have a piece right up until the moment you sit down and discover you don't. The first is a memory problem with well-known solutions. The second is a bookkeeping problem, and it's the one that actually costs you your repertoire. A practice tracker is, at heart, a fix for the second.

Forgetting is normal — and cheaper to reverse than you think

Start with the good news, because it reframes the whole panic. The decay of a learned skill is one of the oldest findings in psychology: in 1885, Hermann Ebbinghaus mapped what we now call the forgetting curve, showing that memory drops steeply soon after learning and then levels off. A 2015 replication by Jaap Murre and Joeri Dros, published in PLOS One, reproduced his results almost exactly more than a century later. Forgetting isn't a personal failing or a sign you learned the piece wrong. It's the default behaviour of every human memory, and a piece you haven't touched in two months sliding out of your fingers is simply the curve doing what it always does.

But Ebbinghaus measured something else that almost nobody quotes, and it's the part that should change how you feel about a "lost" piece. He studied relearning, using what he called the method of savings: even when material seemed completely forgotten, relearning it took far less time than learning it the first time. A list that took twenty minutes to memorise might take only twelve to bring back the next day — a 40% saving. The skill isn't erased when it fades. It goes dormant, and waking it is a fraction of the original cost.

This is the hinge of the whole problem. If maintenance were genuinely costly — if keeping ten pieces alive meant ten full relearns — then forgetting would be a resource problem, and the answer would be "practise more," which nobody has time for. But it isn't. Keeping a piece alive costs minutes, not hours, because of savings. So the real question was never how do I find the time to maintain everything? It's how do I know which pieces need a refresh before they slip past the point where the refresh is cheap? And that is not a memory question at all.

"A piece you think you've forgotten is almost never gone. It's dormant — and waking it costs a fraction of what learning it did."

The forgetting you never notice

Here's why pieces vanish without warning: you are a genuinely unreliable judge of what you still remember, and the error runs in one direction. Memory researchers call it the illusion of competence. When something feels familiar — and a piece you once played a hundred times feels deeply familiar — your brain reads that fluency as evidence that you still have it. But familiarity and retrievability are not the same thing, and studies of metamemory by researchers like Asher Koriat and Robert Bjork have shown again and again that people systematically overestimate what they'll be able to recall later. Your confidence about your own repertoire is not a measurement. It's a feeling, and it's biased toward "I've still got it."

So the decay happens in a blind spot. The piece is quietly sliding down the forgetting curve, but the only instrument you're using to monitor it — your sense of how well you know it — keeps reporting "fine, still there," because the cues that make it feel familiar are still present even as the ability to actually play it erodes. You don't schedule a refresh, because nothing tells you one is needed. Then comes the wedding, the lesson, the jam session, the friend who says "play that thing you learned," and the piece that felt totally secure five seconds ago is full of holes. The crash feels sudden. The decay wasn't.

This is the precise mechanism behind every "I used to be able to play that." Not a weak memory — an unmonitored one. And you cannot fix a monitoring failure by trying harder to monitor with the same broken instrument.

You can't run a maintenance system from memory

If you've read much practice advice you already know the technique that beats the forgetting curve: spaced review. Refresh a piece, and each time you do, the next drop is shallower, so the gaps between refreshes can grow — a day, then a few days, then a week, then a month. It genuinely works, and the techniques for doing it well — small cold starts, fixing the weak spot before it spreads — are worth reading on their own; we cover them in how to stop forgetting pieces you already learned. But knowing the technique is not the same as running it, and here's the catch that defeats most people.

To run spaced review on your repertoire, you have to hold three things in mind: every piece you can currently play, when you last played each one, and which ones are therefore due. For a single piece over a week, fine. For twelve pieces across several months, that is exactly the kind of dated, multi-item record that human memory does not keep — and worse, you'd be asking the failing faculty to schedule its own repairs. The reason players who fully understand spaced review still lose pieces isn't that they reject the method. It's that nothing ever tells them "the Bach hasn't been touched in five weeks," so the refresh that would have cost three minutes never gets triggered until it's a relearn. The missing ingredient was never the technique. It was the record.

The tracker as external memory

The fix is almost boringly simple: move your repertoire memory out of your head and onto something that doesn't forget. A plain list of every piece you can play, each with the date you last played it and a rough status, turns the whole problem inside out. "I think I still know that" — a guess your confidence will get wrong — becomes "last played 34 days ago, due a refresh," which is just a fact you can act on. The tracker isn't doing anything clever. It's holding the three facts your memory won't, so that the spaced-review system you already believe in can finally run on autopilot. You stop trying to remember what to maintain and simply respond to what the list surfaces.

It helps to sort the list into a few honest categories, because not every piece deserves the same attention, and pretending otherwise is how people burn out and quit maintaining anything at all.

For the maintaining set, the schedule is the whole point, and it's designed to get lighter over time. Each refresh resets the curve, so each successive gap can be longer than the last — which is what makes maintaining a real repertoire feasible for a busy person rather than a second full-time job.

Mechanically, any external store works: a notebook, a spreadsheet, the back of a folder. The advantage of a practice app is only that it can date the entries for you and surface what's overdue without you having to scan a list — which is precisely what we built Practis to do, so weigh that accordingly: a repertoire view where each piece carries its last-played date and quietly flags the ones slipping toward "at risk." The tool earns its place not by adding work but by being the thing that remembers, so the responsibility stops living in your already-overloaded head. If your repertoire has grown past what you can hold at a glance, the related habit of keeping several pieces moving at once is worth a look too: tracking multiple pieces without losing focus.

"The pieces that survive in your hands aren't the ones you learned best. They're the ones something was keeping track of."

Setting it up this week

You can build the whole system in fifteen minutes. Write down every piece you can currently play start to finish — the act of listing is itself revealing, because most people find the list shorter than their self-image suggested, which is the illusion of competence showing up in real time. Tag each piece active, maintaining, or dormant, and be ruthless: you can only truly maintain a handful at once, so a deliberately small maintaining set beats an ambitious one you'll abandon. For each maintaining piece, note today's date as the last-played date, then start the widening refresh schedule — a quick cold start, not a full practice, just enough to confirm it's still there and patch any soft spot.

Then let dormant pieces genuinely sleep without guilt. This is the part that frees you: you are not obligated to keep everything alive at performance standard, and trying to is the surest way to keep nothing alive. Because of savings, a dormant piece is a small deposit you can withdraw later for far less than you put in. The list isn't there to shame you about everything you've let lapse — it's there so that when you want a piece back, you know it exists, you know roughly how far it's slipped, and you know the way back is short. If you're new to choosing what's even worth keeping, what a repertoire is and how pianists build one is a good companion.

Common questions

The pieces you've poured months into don't have to be disposable, quietly expiring one by one until "things I used to be able to play" becomes the largest part of your musical life. They slip away not because you're incapable of holding them but because nothing is watching them slip. Give your repertoire a list and a last-played date, and you change the default: instead of losing pieces by neglect and rediscovering the loss at the worst possible moment, you catch them while the way back is still short. The memory aid was never going to be a better memory. It was going to be a record.