“I’d like to learn spreadsheets”, “I should improve my English”, “I’d love to understand data analysis”. Sentences that stay identical for months.
The cause is almost never time or talent. In most cases it’s two far more mundane things: there’s no deadline, and the study method is the least effective of the ones available. On the second point, unusually, there are systematic reviews that rank techniques by measured effectiveness. And the result is uncomfortable, because the most-used techniques sit near the bottom.
Study techniques, ranked by measured effectiveness
In 2013 Dunlosky, Rawson, Marsh, Nathan and Willingham published a review in Psychological Science in the Public Interest assessing ten common study techniques against the available experimental evidence, rating each as high, moderate or low utility. The summary:
| Technique | Utility in the review | Why |
|---|---|---|
| Practice testing | High | Retrieving information from memory consolidates it more than re-reading it |
| Distributed practice | High | Spaced review beats the same total hours crammed |
| Elaborative interrogation (“why is that so?”) | Moderate | Forces connection rather than recognition |
| Self-explanation | Moderate | Surfaces the gaps the text was papering over |
| Interleaved practice | Moderate | Improves strategy selection, not just execution |
| Summarisation | Low | Effective only if you’re already good at summarising |
| Highlighting and underlining | Low | Produces familiarity, which gets mistaken for mastery |
| Re-reading | Low | Same reason: the text gets easy, you don’t get good |
The two entries at the top rest on a large literature: the testing effect is documented directly by Roediger and Karpicke (Psychological Science, 2006), where people who tested themselves retained substantially more over time than people who re-read, while being convinced of the opposite immediately after studying. That’s the core point: the feeling of learning is a poor indicator of learning.
On spacing, the standard meta-analysis is Cepeda, Pashler, Vul, Wixted and Rohrer (Psychological Bulletin, 2006).
Two myths worth clearing out
“I’m a visual learner.” The idea that you learn better when material is matched to your style (visual, auditory, kinaesthetic) is extremely popular and doesn’t survive testing. Pashler, McDaniel, Rohrer and Bjork (Psychological Science in the Public Interest, 2008) searched the literature for studies with the experimental design needed to demonstrate the hypothesis and found very few, with results that didn’t support it. The practical conclusion: the right modality depends on the content, not on you. You learn a procedure by doing it and a pronunciation by hearing it.
“It takes 10,000 hours.” The popular formulation is a simplification of Ericsson’s research on deliberate practice. The meta-analysis by Macnamara, Hambrick and Oswald (Psychological Science, 2014) quantified how much deliberate practice explains performance differences: a lot in games and music, less in sports, and little in education and professions. That doesn’t mean practice is useless: it means the number of hours is a weak predictor, while how the hours are spent matters far more.
What “in 30 days” honestly means
A month doesn’t produce competence. It produces something else, worth more on the labour market than it looks:
In 30 days you don’t get good. In 30 days you produce verifiable evidence that you can do one specific thing, and you move from the “says they can” pile to the “has done it” pile.
That’s exactly the difference between the two columns of a CV, and the reason this is the only version of the “30-day method” that makes sense. On how to write that evidence up once you have it: how to write an effective CV.
The constraint that makes the difference: output first, input second
The structural mistake is starting from the material (“I’ll buy the course, then we’ll see”). The correct order is the reverse: define the final artefact first, and pick the material to serve it.
One-line formula: by [day 30] I will have produced [concrete object] that [someone] can look at.
Examples that work, because they’re verifiable:
- a dashboard that refreshes your department’s monthly figures on its own;
- a procedure rewritten and actually adopted by your team;
- a ten-page analysis of public data, with sources;
- twenty minutes of technical conversation in a second language without switching back.
Examples that don’t work: “learn Python”, “improve my English”, “understand AI”. They have no completion criterion, so they don’t complete.
The plan, four weeks
| Week | Goal | Operating rule |
|---|---|---|
| 1 | Narrow the field | Learn only what the artefact needs. Everything else is later |
| 2 | Produce the first bad version | Something wrong has to exist before it can be corrected |
| 3 | Hunt for the error | Self-testing and external feedback, not more material |
| 4 | Refine and ship | Delivery is what makes the deadline real |
Four rules that apply across all four weeks.
1. An hour a day, spread out, beats eight hours on Saturday. It isn’t a willpower matter, it’s the spacing effect documented in the Cepeda meta-analysis. If you’re short of time, the better move isn’t to study more: it’s to space it better.
2. Close the material before you check. At the end of each session, book closed, write down what you understood and what you’d do tomorrow. If nothing comes out, you haven’t learned it, and you found out today rather than at delivery.
3. Fix when and where, not how much. “I study after dinner, in the kitchen, from 21:15” works better than “I study an hour a day”. It’s the implementation intentions mechanism described in time management: what actually works.
4. One goal at a time. Two skills in parallel over thirty days produce two unfinished things, which are worth zero on the market.
The most expensive mistakes
Picking the wrong skill. A month spent well on something nobody wants is still a month lost. Before starting, it’s worth looking at the demand data rather than the hype: what skills employers actually ask for.
Confusing consumption with learning. Watching twenty hours of video is a pleasant activity with an excellent sense of progress and a far smaller effect than two hours of exercises done wrong and corrected. It’s precisely the gap between “familiarity” and “mastery” the reviews describe.
Postponing the hard part. The section you’re avoiding is the one that contains the skill. Push it to day 28 and you won’t do it.
Showing nobody anything. External feedback is the only instrument that finds the mistakes you don’t know you’re making.
My take: thirty days only work if you give something up
I think the most underrated part of the method isn’t discipline but renunciation: thirty days only work if you accept learning 20% of the subject and deliberately ignoring the rest. Almost everyone fails here, because studying in order, from the start of the manual, feels safer than studying towards a goal.
And something less diplomatic: the feeling of learning is inversely correlated with learning more often than we’d like. Re-reading is pleasant, trying to recall is unpleasant, and the research is clear about which one works. If you feel permanently competent throughout the month, you’re probably using the wrong method.
Frequently asked questions
Are thirty days really enough? For a bounded, operational skill yes, if the goal is an artefact rather than mastery. For broad skills (a language, programming) thirty days get you past the threshold where progress becomes self-sustaining, not to the finish.
How many hours a day? Fewer than you’d think, as long as they’re spread out: an hour a day for thirty days beats the same total crammed into weekends, because of the spacing effect documented in the literature.
Paid course or free material? Structure and the presence of corrected exercises matter more than price. A course mainly saves you from wasting time choosing a path; it doesn’t replace producing the artefact.
How do I prove the skill to a recruiter? With the object produced and the number attached to it, not with the line in the CV. Same logic as transversal skills: how to prove them.
I skipped a day. Have I ruined it? No. The plan runs to a deadline, not an unbroken streak. The useful rule is not to skip twice in a row, because the second skip is what turns an exception into abandonment.
Sources
- J. Dunlosky, K. A. Rawson, E. J. Marsh, M. J. Nathan and D. T. Willingham, Improving Students’ Learning With Effective Learning Techniques, Psychological Science in the Public Interest, 2013. Ten study techniques rated for utility: practice testing and distributed practice rated high, re-reading and highlighting rated low.
- H. L. Roediger III and J. D. Karpicke, Test-Enhanced Learning, Psychological Science, 2006. Active retrieval produces better retention than re-reading, despite students predicting the opposite.
- N. J. Cepeda, H. Pashler, E. Vul, J. T. Wixted and D. Rohrer, Distributed Practice in Verbal Recall Tasks: A Review and Quantitative Synthesis, Psychological Bulletin, 2006. Effect of spacing review over time.
- H. Pashler, M. McDaniel, D. Rohrer and R. Bjork, Learning Styles: Concepts and Evidence, Psychological Science in the Public Interest, 2008. Lack of adequate evidence supporting the learning styles hypothesis.
- B. N. Macnamara, D. Z. Hambrick and F. L. Oswald, Deliberate Practice and Performance in Music, Games, Sports, Education, and Professions: A Meta-Analysis, Psychological Science, 2014. Deliberate practice explains a limited share of performance differences, much reduced in education and professions.
- P. M. Gollwitzer and P. Sheeran, Implementation Intentions and Goal Achievement: A Meta-Analysis of Effects and Processes, Advances in Experimental Social Psychology, 2006. Effect of planning when and where on goal execution.
- Cedefop. European analyses of skills demand and mismatch.
- ESCO. European classification of skills and occupations, useful for bounding the goal in recognised terms.



