15 minutes a day or 2 hours on the weekend — which works better?

15 minutes a day works much better than 2 hours on the weekend, even though the total time is the same. The reason is Ebbinghaus's forgetting curve: the brain loses up to 70% of what it learned in 24 hours if it doesn't review. Studying daily reactivates memory before forgetting happens. Studying once a week allows the brain to forget everything between sessions — you spend the first half of the session reviewing what you lost.

What is Ebbinghaus's forgetting curve?

Hermann Ebbinghaus, a German psychologist, discovered in 1885 that memory decays exponentially after learning. In 20 minutes, we lose 42% of what we just learned. In 1 hour, 56%. In 1 day, 67%. In 6 days, 75%. In 1 month, 79%. The curve flattens after that. To combat this, you need to review before total forgetting — hence spaced repetition (SRS) that moves the next review to 1d, 3d, 7d, 15d, 30d.

What does science say about distributed vs massed practice?

The most cited meta-analysis is Cepeda, Pashler, Vul, Wixted & Rohrer (2006), which reviewed 184 articles on the spacing effect in memory. Conclusion: distributed practice outperforms massed practice in all tested intervals, with a medium to large effect (Cohen's d between 0.4 and 0.9). The gain is greater when the interval between sessions is proportional to the desired retention time — which is why 15min/day beats 2h in one day.

Is there an ideal number of minutes per day?

There is no magic number, but there is a validated range: between 30 and 60 minutes per day is the sweet spot for working adults. Below 15 minutes, the brain barely enters a deep processing state. Above 90 minutes in a single session, fatigue reduces the gain per minute. The FSI (which trains US diplomats) uses 5h/day in immersion, but that's for 6-month goals; for a sustainable pace, 30-60 min/day is the best balance.

And what if I really only have 15 minutes a day?

If you have 15 minutes a day, you still make progress — it just takes longer. 15 min/day × 365 days = 91 hours a year. In 7 years, you reach the 600 hours that the FSI estimates for B2/C1 fluency in English. It's slow, but it works if consistency is total. The trap is that people who do 15 min/day tend to skip days and not make up for it — which nullifies the effect of consistency. 15 min/day 7x a week > 30 min/day 3x.

Why does 1 hour in one day seem equivalent to 1 hour divided into 4 sessions — but it's not?

Because the brain consolidates memory during the intervals between sessions, especially during sleep. Each 15-minute session is followed by hours of consolidation. 4 sessions = 4 cycles of consolidation, compared to 1 cycle in a single session. It's the same reason why athletes train every day instead of once a week — the body adapts during rest, not during brute effort.

Is it worth studying more on the weekend to compensate for the week?

No, it doesn't compensate. Studying 4 hours on Saturday after 5 days of nothing is less efficient than having done 30 minutes a day. The 4 concentrated hours generate cognitive fatigue, and the brain processes worse. Moreover, you spend the first hour just re-warming what you forgot. The ideal is to not skip days — even if it's 10 forced minutes.

How to maintain consistency of 30 min/day when life gets in the way?

Three validated tricks: (1) tie studying to an existing habit (during breakfast, on the way, before bed) — the old habit becomes the trigger for the new one; (2) reduce friction to almost zero (app on the phone, materials always ready, no decision of 'what to study'); (3) 2-minute rule: on bad days, do only 2 minutes — the important thing is not to break the streak. Consistency beats intensity, always.

15 Minutes a Day vs 2 Hours a Week

TL;DR

Even with the same total time, the results are completely different. 15 minutes a day (105 min/week) beats 2 hours on the weekend (120 min/week) — because the brain works in cycles. The Ebbinghaus forgetting curve shows that we lose 67% of what we learn in 24 hours without review. The meta-analysis by Cepeda et al. (2006), with 184 studies, confirms: distributed practice outperforms massed practice with a large effect. In this post, I show the 2 graphs (forgetting curve + direct comparison) and how to apply this.

Look, this is the oldest debate in language study. Some people prefer to accumulate study on the weekend — "I'll dedicate 4 hours on Saturday." Others prefer to break it up — "20 minutes every day during lunch." And the question is simple: if the total time is the same, is the result the same?

The answer is no. And it's not an opinion — it's one of the most solid discoveries in memory science, validated since 1885 and replicated hundreds of times. In this post, I'll show you where the rule comes from, the 2 graphs that explain it, and how to apply this to your language study.

The discovery of 1885 that still matters: Ebbinghaus

Hermann Ebbinghaus, a German psychologist, conducted an experiment in 1885 that changed the study of memory forever. He memorized lists of nonsense syllables (to eliminate semantic association) and measured, at increasing intervals, how many he could remember. The result: memory decays exponentially.

Without review, we lose:

20 minutes later — 42% forgotten
1 hour later — 56% forgotten
1 day later — 67% forgotten
6 days later — 75% forgotten
1 month later — 79% forgotten

The curve flattens after that (the little that remains is more resistant), but the important point is: in 24 hours, the brain throws away 2/3 of what it learned if nothing happens. Those who study once a week come back the next week and find they have forgotten almost everything.

The graph of forgetting (with vs without consistency)

Ebbinghaus's Forgetting Curve · with and without daily reinforcement

Memory retention over time — red: without review · purple: with daily review

The retention difference is 4x — just by the review pattern.

Looking at the graph, the story becomes clear: those who review every day maintain memory between 85% and 100%. Those who do not review drop to 21% in a month. The "intensity" of 2 hours on the weekend helps at the moment — but if you don't review in the following days, you will lose most of it.

The meta-analysis that confirms: Cepeda et al. (2006)

In 2006, Cepeda, Pashler, Vul, Wixted & Rohrer published the most influential meta-analysis on this effect — known as the spacing effect or distributed practice. They aggregated 184 studies from 1900 to 2005 and found a consistent pattern: distributed practice outperforms massed practice in all tested conditions, with a medium to large effect (Cohen's d between 0.4 and 0.9).

The gain is greater when the interval between sessions is proportional to the desired retention time. To retain for 1 month, intervals of 1-2 days between sessions are ideal. To retain for 1 year, intervals of 1 week work. To retain for a lifetime (the goal of language learning), the ideal is increasing cycles — exactly how the SM-2 algorithm by Wozniak works in SRS.

The direct comparison: 15min × 5 vs 2h × 1

▸ Consistency

15 min × 7 days

= 105 min/week

✓ Review before forgetting
✓ 7 cycles of consolidation in sleep
✓ Automatic habit in 30 days
✓ Retention of 85-90%

▸ Intensity

2h × 1 day

= 120 min/week

✗ Forgetting between sessions
✗ Only 1 cycle of consolidation
✗ No habit — relies on willpower
✗ Retention of 25-30%

The person who studies "2 hours on the weekend" studies 15 minutes more per week and still retains 3x less. It's counterintuitive, but it's the math of memory.

Why 30-60 min/day is the sweet spot

15 min/day works, but it's the minimum. The ideal point for working adults is 30 to 60 minutes per day, for 4 reasons:

Enough time for deep processing — below 15 min, the brain hardly enters "learning mode." Between 30-60 min, there's time for input, active practice, and review.
Reasonable volume to reach 600 hours in ~2 years — validated goal from FSI for B2/C1 fluency.
Sustainable in the long term — 60 min/day fits into the routine without becoming torture. 2 hours/day wears you out, and most abandon it in 3 months.
Allows distribution among skills — 20 min listening + 20 min speaking + 20 min vocabulary = balanced coverage.

4 tricks to maintain consistency

1. Tie to an existing habit

"Whenever I have breakfast, I open the app." The old habit becomes the trigger for the new one.

2. Zero friction

App on the phone, no "deciding what to study." The less friction, the more consistency.

3. 2-minute rule

On bad days, just do 2 minutes. The important thing is not the quantity — it's not breaking the streak.

4. Visible streak

Seeing the accumulated days (X consecutive days) activates the reward circuit. Breaking it hurts — and that's good.

What to do when "there's no time"

Those who say "I don't have 30 minutes a day" usually have 30 minutes spread over 6 moments of 5 minutes. The solution is not to find more time — it's to take advantage of the micro-moments:

5 min in the Uber → 5 flashcards
5 min in the coffee line → 1 short podcast episode
10 min at lunch → 1 voice conversation with AI
5 min before bed → review words of the day
5 min brushing teeth → shadowing audio

Total: 30 min/day, without changing anything in your routine. This is what differentiates those who say "I've been studying for 5 years" and never reached B2 from those who reached B2 in 18 months — it wasn't talent, it was smart distribution.

Frequently Asked Questions

Does 15 min a day work for fluency?

It works, but slowly. In ~7 years, you reach the 600h of the FSI. 30-60 min/day is ideal.

Can I make up for lost days on the weekend?

No, it doesn't compensate — you lose due to deconsolidation. Better not to skip.

What is the most important study on the subject?

Cepeda, Pashler, Vul, Wixted & Rohrer (2006). Meta-analysis of 184 studies.

Is it worth studying before bed?

It's very worth it — sleep consolidates memory. Passive listening before bed has proven effects.

How much rest time between daily sessions?

At least 4 hours. The brain needs the interval to consolidate.

Start your streak today

If you do just one thing after reading this post: set a fixed time of 30 minutes a day, start tomorrow, and don’t break for 30 days. Use the 2-minute rule on bad days. In 30 days, the habit will be automated, and the results will start to show. Nothing accelerates progress more than this.

Automated Consistency

Lanna tracks your daily streak, sends reminders at the ideal time, and adapts the session to 5, 15, or 30 minutes as you have time. Consistency becomes easy.

Try Lanna for free