Why the daily load multiplies
The ladder used here is 1, 3, 7 and 14 days: a word learnt today comes back tomorrow, then three days out, then a week out, then a fortnight out. Once the schedule has been running for fifteen days, five sets overlap on the same day — today's new words plus the four review sets — so a plan of 30 new words a day is a plan of about 150 words a day.
That is the number to test against your patience before you start, not the intake figure. The first week feels manageable because the stack is not yet full, and the second week is where the real load arrives. Planning against the peak rather than the intake is the difference between a schedule that survives and one that gets quietly abandoned around day twelve.
Once the new-word phase ends, the load runs the other way: nothing new goes in, the ladder drains, and the last review lands fourteen days after the last new word.
What is and is not settled about spacing
That spreading study over time beats packing it into one session is one of the better-supported findings in the study of memory, and it holds across ages, materials and test formats. What is not settled is the schedule. The 1-3-7-14 ladder used here is a common practical choice, not a derived optimum; other tools use 1-4-7-14, or a doubling ladder like 1-2-4-8-16, and there is no consensus that any fixed set is best.
Adaptive schedulers behave differently again. Anki, SuperMemo and similar systems adjust each word's next interval based on how you answered it, so their load does not stack the way a fixed ladder does — easy words disappear into long intervals while difficult ones come back quickly. If you are learning more than a few hundred words, that behaviour is worth having, and this page is best used before you start, to work out what daily intake is realistic.
No retention figure appears anywhere on this page, because a schedule cannot produce one. How much you keep depends on the words, the language, what you already know and how you review, and any percentage attached to an interval ladder is decoration.
Making the reviews cheaper
- Retrieve, do not reread. Covering the meaning and trying to produce it is a different activity from reading the pair again, and it is the one that most of the evidence on testing is about.
- Retire what is solid. A word you answer instantly three times in a row does not need the rest of the ladder. Taking it out is what makes the real load lighter than this page's arithmetic.
- Separate the stubborn ones. The words that keep failing are a small fraction of the list and deserve a short list of their own, seen daily. Cycling the whole set to catch them is expensive.
Reading the numbers back
If the busiest-day figure is larger than you would actually sit through, lower the intake rather than the ladder. Halving the intake halves every number on the page and roughly doubles the length of the new-word phase, which is usually the better trade for anything that is not a test in a fortnight. The ceiling field does that division for you.
For a plan measured in pages or lectures rather than items with a review tail, the Study Plan Calculator is the closer fit, and the Study Hours Target Calculator handles the case where the binding constraint is hours in the week rather than items in the day.
Questions people ask
Can I do 100 new words a day?
For a short target with a fixed date — a vocabulary test in two weeks — it is a real strategy, and the arithmetic works: the ladder never fully stacks before the test arrives. For a long-term vocabulary goal it usually loses to 30 or 50 a day, because the peak load at 100 a day is around 500 items and few people hold that for months. The total learnt after a year is what matters, and a rate you keep beats a rate you abandon.
I missed several days of reviews. Start over?
No. Clear the backlog first and pause new words while you do it, then resume. A review done four days late is still a review; the interval ladder is a convenience, not a rule that breaks when a day slips. The one thing worth avoiding is adding new words on top of a backlog, because the load compounds and the backlog never clears.
Why 1, 3, 7 and 14 days?
Because it is a widely used practical ladder that spaces the reviews out and is easy to keep track of on paper. It is not an optimum. Other published ladders use different numbers and there is no settled answer about which is best; what the evidence supports is the general shape — several reviews, with gaps that grow — rather than any particular set of days.
Does this account for words I already know?
No. The target is whatever number you enter, and every word in it is treated as new. If a third of a word list is already familiar, subtract it from the target before you start, or the plan will schedule reviews for words that never needed them.