If you’re putting in time but not getting better, you may be repeating “low-value reps”: actions that feel productive because they’re familiar, measurable, or urgent, but don’t create learning or meaningful output. This post will help you identify these traps, audit your own sessions, and redesign your workflow for actual results.

• You’ll repeat them anyway (familiarity): it’s hard to shake the feeling of work when you’re regurgitating the same moves.
• They become the norm (habit): yet another practice window, another set done, another follow-up email.

When encountering these low-value repetitions, we glorify busywork. We settle for something that feels hard therefore must be valuable, when we’d maximize improvement by rearranging our practice. But who wants to admit they just spent all day being busy without improving?

10 minute audit

To spot and flip low-value repetitions, one simple method is the following:

In a spare ten minutes, try this “10 minute audit” of your most recent practice sessions: list out the last ten practice episodes you had, name the micro-skill, the source of your feedback, and most importantly write down what changed in the actual practice skill from last session to this session.

Doing this will help you identify and replace low-value repetitions with higher value repetitions by adding one of 4 levers:

• Clearer target (focus on learning outcome).
• Tighter constraint (make it more of a stretch).
• Faster feedback (try filming, put them under pressure).
• More variability (interleave multiple challenges).

The clear-upside about paying attention to low-value repetitions is how easy it is to track progress. You can look back and find evidence in the form of increased quality, speed, reduced error rate, more sordid transfers, rather than just by looking at the number of minutes logged at the violin.

• They’re “measurable” in the shallow sense (time, streaks, checklist) even if your results don’t, thanks to bonus points, get better.
• They reduce anxiety: It feels psychologically safer to repeat something you kind-of can, than to practice something you definitely can’t.

In knowledge work, this can become a culture problem (and also a self-induced problems): “busyness” in such cases is your culture’s stand-in for productivity (“pseudo-productivity”) when it becomes difficult to discern if the quality of what is produced changes day to day. (ft.com)

4 signs your practice is busy but ineffective

1) The reps are below your “learning edge”

If you can do it on autopilot, you are mostly just reinforcing what you already know. This might be useful for maintenance (staying warm), but it is weak at generating growth. Growth practice usually involves a tolerable struggle, where you are attempting something you can’t yet do reliably.

2) There’s no fast feedback loop

If you can’t quickly distinguish what went wrong, and what “better” looks like, then you’ll just repeat mistakes with confidence. That’s why deliberate practice research pushes structured goals, error diagnosis, and informative feedback—not just time-on-task. (frontiersin.org)

3) You’re repeating the same context, so nothing “transfers”

Blocked repetition (same type of rep, same recency, same conditions, etc.) can create the illusion of improvement that then vanishes when in a real situation. Introducing variability, like interleaving related problem types, has been shown to improve both memory and problem solving on novel tasks, even when it feels like it is harder in practice. (nature.com)

4) The session is “fragmented” from switching and interruptions

Even if you have a good practice plan, if you are constantly switching then you ruin it “quietly.” Research on “attention residue” shows that when you switch tasks, part of your attention can remain stuck on the previous task, hurting what you do next. (researchgate.net).

Related research on workplace interruptions also finds that interruptions can amplify stress and frustration, even when people attempt to compensate by working faster. (researchgate.net).

A 10-minute audit to find your low-value reps (do this today)

1. List your last 10 practice sessions (or work blocks). Use real dates and rough durations.
2. For each session, write one sentence: “I was trying to improve ______.” If you can’t name it, flag it.
3. Write the feedback source: coach review, tests, code review, timer, recording, customer response, etc. If feedback was “vibes,” flag it.
4. Write what changed from the prior session (harder constraint, new problem type, different tempo, new audience). If nothing changed, flag it.
5. Circle the sessions that produced a visible delta within 24–72 hours (fewer errors, faster time, better outcome, clearer decisions). Those are your high-signal sessions.

Score your repetitions: the “Rep Value” checklist

Apply this quick scoring rubric to any rep you’re about to do. “High-value” means it scores well on at least 3 of 5 dimensions. A rep that scores an 8-10 is usually “growth practice.” A 3-5 is maintenance or busywork. Maintenance isn’t bad—just don’t mistake it for improvement.

8 common low-value repetition traps (and how to remedy each)

Trap 1: Autopilot reps (you’re practicing what you already can do)

What it looks like: running the full song, same set/pace, rewriting the same type of code, re-reading notes, etc.
Why it happens: it feels fluent, and fluency feels like mastery.
Fix: choose one micro-skill and tighten the constraint (tempo, threshold for accuracy, timer, smaller detail). Stabilize the rest.

Trap 2: “Input binges” (consuming instead of producing)

• What it looks like: more tutorials, more articles, more videos—little shipping, little testing.
• Why it happens: input is “safe”; output exposes some gaps.
• Fix: swap to retrieval/output: summarize from memory, do a cold attempt, then check. Set an output “tax” per input hour (eg: learn 30 mins, now build 60 mins).

Trap 3: Repetition without error diagnosis

You can repeat till the cows come home, but without a diagnosis it isn’t deliberate. The classic details are that the coach (or whatever system) helps identify the problem, what would be meaningful information allowing improvement, and sequences/steps tasks such that it gets harder as you get better. (frontiersin.org)

1. Call out (to yourself) the name of the category of error after each rep (don’t just say “messed up”). Eg: timing drift, unclear claim, missed edge case, weak objection handling.
2. Pick a cue relevant for correcting that genre of the error for the next rep (one).
3. Re-run immediately, and compare. Is your error different? If not, you picked the wrong cue, or it is too vague.

Trap 4: Blocked practice when you need discrimination

Blocked practice (doing a type problem for a long strip) looks good at a glance: but isn’t always as good as interleaved (mixed truly related types) when later performance on something completely new (and tougher) is what’s being tested. (nature.com)

Trap 5: Tool-tweaking disguised as practice

• What it looks like: tweaks on the app, rewrite gameplan, conglomerate the template, deep in “system” work.
  Why it happens: it offers fast feedback and guards against feeling out of control.
• Fix: cap it. Example: 20 minutes/week for tooling. Does this specific tool change get rid of Specific Bottleneck you can name? If not, it’s procrastination in nice clothes.

Trap 6: Context switching that turns practice into confetti

Research on attention residue suggests that switching feels poorer after context-switching, particularly when the switched-from task is unfinished. (researchgate.net)

• Fix: create a “closure minute” before switching. Write: what’s the next action?, what’s the open question?, what point do I restart at if I switch here?”
• Fix: batch “shallow work” (messages, scheduling) into fixed windows so they do not leak into high-cognitive work.
• Fix: if you have to interrupt yourself, interrupt at a clean boundary (either end of set, end of rep, end of paragraph), not mid-rep!

Trap 7 “Visible work” that never reaches an outcome

“Status updates,” meetings that become habit around “keepin’ it moving”, and always seeming “on the ball” can create feeling of momentum while making it hard to do the cognitively demanding work that creates value.
This dynamic is often discussed around product management, pseudo-productivity, and “shallow work.” (ft.com)

• Fix for teams: Define “output moments” (decisions made, artifacts shipped, tests passed) and protect time for them.
• Fix for individuals: For every coordination task, ask “What outcome does this enable and when will it be done by?” If it goes away, good, now you’ve eliminated waste!

Trap 8: Perfection loops (polishing instead of learning)

• What it looks like: Rewriting one section over and over; tuning one feature for years; doing take after take no matter what change in flow you make.
  Why it happens: perfect practice makes perfect, right? Except when it isn’t, and instead of doing it’s avoidance of doing.
• Fix: separate out “exploration reps” from “performance reps.” Exploration reps are where you optimize for maximum learning gain per attempt (many attempts in various conditions). “Performance reps” are where you care about your final output (find a polish, a fresh take). Put them both on your Calendars as different “modes.”

How to redesign practice so it actually works (a simple 4-lever model)

If your practice is enough, don’t add time. Added minutes alone aren’t that valuable. Add leverage. Don’t worry about drastically changing your routine, just try to upgrade every session by merely pulling one of these levers at a time:

The levers line up with deliberate practice map very closely: structured tasks, clear goals, feedback, challenge.

___You could spend an hour reading about copywriting and that’s probably a low-value rep. You could spend an hour writing 10 headlines for the same offer, testing them against, say, a checklist of criteria around specificity and benefit clarity and audience match, and that could be a higher-value rep that you might get feedback from a peer or a rubrics from your own style and maybe A/B test or even just a forced “cold consider of multiple hours away” in your gut.

___Low value: “Let’s play this piece start to finish.

___Higher value: “You know what? The hardest sit in the whole thing, just those toughest two measures, play it with a metronome, but really, really slow on purpose, 20% slower maybe at least to start, and loop and drench that and video and then change only committed technique cue on that one aspect and only learn it a few ticks maybe smaller parts faster and gradually speeding it up but not until it’s unlocked at that lower speed.

___Kind of interestingly, just besides scrolling to read, one variable might be interleaving some other tech “drills after”. When working on articulation, rhythm, or some facets of the dynamics or just try doing any component for 30 minutes but at least interleave things with a little more diversity within a 30 minutes session can help with transfer and discrimination.

___Instead of just watching another tutorial while following along, drop 20% of the time and try implementing the main idea from scratch.

___Feedback can be more about unit tests and linting and (smaller) peer reviews that drill into just one dimension of code: readable or performant? Best of all are the built in checks along with limited or targeted code review. One thing I’ve seen is cycling into smaller pieces you do with that thing but really, how important now is lining out all the code to fix an error on performance that we’ll misbuild with the computer clock.

___Low-value: Workout every first same schedule, just broadly tracking time spent and so on.

___Higher-value: just pick one of our outcome metrics (maybe it’s a pacer we want, rep at a given load approaching something) “you know, make a brief, row a tiny speck space up in the gym or feel a certain frame or run 400m fast forward.”

___Kind of feedback, we answer that and video it and while we’re coaching ourselves pick a strange two pace gauges, 10 with this cannon at these weight, bang looking for these strains and matches.,

How to protect high-value practice from interruptions (without becoming unreachable)

You don’t need a monastic lifestyle. You need fewer mid-rep interruptions and fewer half-finished tasks. Attention residue research suggests that unfinished work can “stick” in your mind and degrade your performance on what follows. (researchgate.net) So:

• Set a small, explicit “availability contract”: e.g., I’ll check messages at 11:30 a.m. and 4:30 p.m. (or whatever cadence suits your role).
• Do 45–90 minute practice blocks on a single target with a defined end condition (a test, run-through, or draft).
• End each block with a 3-line debrief: what improved, what failed, what’s the next rep? Create closure to reduce cognitive carryover.
• If you manage others: replace “always on” with predictable response windows and clear escalation path for true emergencies.
• Reducing interruption isn’t just about speed. Research on interruptions suggests that people may work faster to compensate for interruptions, but with higher stress and frustration. Protecting your focus is also protecting your nervous system. (researchgate.net).

What to track instead of “hours practiced”

Time is an input. Improvement needs evidence. Pick 1–2 metrics that reflect the skill you care about and review weekly.

Examples:

• Quality: rubric score, fewer critical errors, clearer readable, better form.
• Speed: time-to-solve, time-to-draft, cycle time.
• Consistency: performance variance across days/conditions.
• Transfer: can you do it with a new prompt, new opponent, new dataset, new audience?
• Decision quality: fewer reversals, clearer next actions, better forecasting accuracy (where relevant).

A one-rep-away daily practice: 3 sessions that beat 7 scattered ones

If you’re busy you don’t need more sessions, you need fewer, higher-quality sessions. Try this structure for two weeks:

This prevents you from trying to fix everything that’s wrong at once, and burn out or lose signal.

A common mistakes when we failed to fix low-value repetitions: trying to fix everything at once (you’ll burn out / lose signal). Change 1 lever per day, and so on.

• Seeking novelty instead of addressing a bottleneck (variability matters, but only around a specific skill you’re trying to master).
• Planning instead of practicing (your plan should fit on a sticky note during the session).
• Motive instead of design (design the environment and feedback probe so you can practice even when you don’t feel like it).
• Never testing transfer (if you don’t test outside the practice context, you’ll never know what’s real).

FAQ

References

• Ericsson (2019) on why the original definition of deliberate practice applies to apparently unintentional undertakings (Frontiers in Psychology)
• Ericsson, Krampe, Tesch-Römer (1993) The Role of Deliberate Practice in the Acquisition of Expert Performance (Psycholog
• Leroy (2009) The attentional residue paper (Organizational Behavior and Human Decision Processes)
• Strategy+Business summary of Leroy’s attention residue findings
• Mark, Gudith, Klocke (2008) The cost of interrupted work: More speed and stress (CHI)
• Samani & Pan (2021) Interleaved practice enhances memory and problem solving in undergraduate physics (npj Science of
• Cambridge Handbook of Cognition and Education (2019) chapter on interleaving practice
• Financial Times review discussing Cal Newport’s take on “slow productivity” (Slow Productivity)
• Wall Street Journal piece on “slow productivity” and “how to cultivate meaningful output—not mere busyness”