Drowning in Vocabulary Lists? Stop Saving Every Word. Do This Instead.
A practical decision map for choosing what to keep, skip, and study deeply.
Saving every new word you meet is one of the fastest ways to learn almost none of them.
It feels like the opposite should be true. You are reading a book or listening to a podcast, and unfamiliar words keep coming. You don’t want to lose them, so you catch them all. You write them down, tap them on your Kindle, or screenshot the page for later. The list grows. Later never quite arrives, and when it does, you are staring at two hundred words with no idea which ones matter, half of them stripped of the sentence that gave them life.
You feel busy. You are not moving.
This is the real vocabulary problem. Learners who hoard words are too thorough. They treat every unknown word as equally worth keeping, so they keep everything, and a pile of everything works exactly like a pile of nothing. You cannot see the wood for the trees.
I have watched this happen for twenty years, in classrooms and in my own language learning. And the standard advice is almost useless in the moment that counts. “Guess from context.” “Use a frequency list.” Fine. But you are on page forty with a word in front of you right now, and what you need is a decision. Keep it or let it go. And if you keep it, what then?
So that is what this piece gives you. A decision map you can run in the second after a word appears, grounded in what the research actually shows about how words get learned. By the end, you will know which words to save, which to wave through, what to write on the card, and how the kind of reading you are doing changes every one of those answers.
The first thing the map does is take the weight off your chest. You were never supposed to catch them all.
New here? I write about the science of how languages are actually learned, one idea at a time. Subscribe, and the next one lands in your inbox.
Step 1. First, name what you are doing
Before you decide on a single word, answer one question about the activity itself.
Are you working extensively or intensively?
The right move for a word depends entirely on which one you are in, and the channel you are using, whether you are reading on a screen, reading on paper, or listening with or without a transcript, changes the practical steps.
So separate the two modes before you start.
If you are going extensive (flow and volume)
Extensive means input for pleasure and quantity: a novel on the train, a podcast while you cook, lots of language you mostly understand. The goal is momentum and comprehension, not capture. Your rule here is to mark lightly and decide later, never to break the flow.
What that looks like in practice:
On an e-reader (Kindle and similar): tap the word, read the gloss, keep reading. Most e-readers automatically log every word you tap into a saved list (Kindle calls it Vocabulary Builder, though not every device does this, so check yours). That list is raw material, not a study list. Do not sort it mid-book. You triage it after the session.
On paper: read with a pencil in your hand, the way you already annotate ideas. When a word genuinely stops you, underline it or put a dot in the margin, and read on. Do not look anything up mid-page. At the end of the chapter, flip back through your marks and decide which few are worth checking. Marking now and judging later is the compromise the research supports: noticing a word helps you learn it, but stopping to look up every one destroys the volume that makes extensive reading work in the first place.
Listening with a transcript: listen first for meaning. Then scan the transcript and highlight only the words that recurred or blocked you. The transcript is your capture tool, so you do not have to freeze the audio every few seconds.
Listening without a transcript: you cannot catch what you cannot see, and that is fine. Extensive listening trains your ear through sheer volume. If a word keeps surfacing and keeps blocking you, note it (a quick phone memo, a timestamp) and look it up afterward. Everything else, let it wash through.
The thread running through all four: in extensive mode, you capture lightly and judge afterward.
If you are going intensive (study and harvest)
Intensive means a shorter text you work through closely, on purpose, to pull language out of it. Here, stopping is the whole point. You slow down, you look words up, you mine.
What that looks like in practice:
Read the sentence first, then check. Let a guess form before you reach for the dictionary, because the guess is half the learning. The dictionary confirms or corrects what your brain already tried.
Mark as you go. Highlight the words you choose to work on, or write them straight into a working list beside the text.
Keep the passage short. A single page worked hard gives you more vocabulary than ten pages skimmed. Depth beats distance in this mode.
Do not translate every word the instant you meet it. That reflex is the flow-killer in pleasure reading and the time-sink in study reading. Reach for meaning after your brain has tried, not instead of it.
This is the mode where deliberate saving actually happens. The rest of this map is mostly for here, plus the rare word in extensive mode that earns the same treatment.
Step 2. The save-or-skip rule (the part to screenshot)
Here is the hard center of the whole thing, the part that was missing from your instincts before.
When you meet an unknown word, Nation (2022) trains learners to choose deliberately between four actions, instead of reflexively saving:
Wave it through. Ignore it, or guess from context, and read on.
Check and release. Look it up, understand it, save nothing.
Park it. Mark or list it now, decide later.
Work on it now. Stop and learn it deliberately.
Most words deserve action 1 or 2. A few earn action 3. Very few earn action 4. Three questions tell you which, and you run them in order.
Will I meet this word again?
This is the strongest signal there is. A word that is common in the language, or common in your particular life, will return on its own, and every return teaches you a little more for free. A word from your job, your hobby, or your daily routine earns a place even if a frequency list calls it rare. A word that is rare everywhere and absent from your world will not come back. Don’t ask whether a word is useful in general. Ask whether it is useful to you, and often.
Can I anchor it to something I already know?
Words you can hook onto existing knowledge, a shared root, a cognate, a near-synonym, are cheap to learn and worth grabbing. A word that is strange the whole way through, sitting in a sentence where three other words are also strange, is telling you the text is above your level, not asking to be saved. A word with nothing to attach to will not stay attached.
Did not knowing it actually stop me?
Your brain runs ahead and fills gaps, which is why you can finish a chapter without looking up every word. If you understood the sentence without it, the word was scenery. Keep the words that genuinely broke the meaning, or the ones you reached for in your own speech and could not find.
The rule, in its hard form:
Save a word only if it passes all three: it will recur, you can anchor it, and it actually blocked you.
Skip it if it fails even one: a one-off, or unanchorable because the text is too hard, or a word you understood without.
When in doubt, skip. A genuinely useful word you skipped will come back and ask again. A useless word you saved just clogs the system and hides the words that matter.
Real examples. A podcast says a word three times in five minutes, and you still cannot get it: recurring and blocking, so keep it. A novel drops an ornamental adjective once, and you guessed it fine: let it go, with no guilt.
Most words fail at least one question. That failure is the correct result. Saying no is the skill nobody teaches.
Every article and PDF guide here takes weeks of research, writing, and rewriting to turn dense science into something clear you can use tonight. I’m one person doing this work with care. Upgrade to paid to get full access to everything I create and help me keep building it.
Step 3. Turn the messy capture into a clean list
After the session, you have raw captures: marks in the margins, highlights on a transcript, a logged list on your Kindle, scribbles beside a textbook. This is where the sorting happens, and the sequence matters.
Gather everything into one short list. Copy your paper marks into a single list, or open the e-reader’s saved words. The act of rewriting them by hand is not busywork. Pulling each word back out of your memory as you write it is already a small dose of the retrieval that makes words stick.
Run each word past the three questions and cross out the failures. Be ruthless. Expect to delete most of them. A list of forty that becomes a list of six is the system working, not failing.
The survivors become cards. Nobody else.
A short, chosen list beats a long, automatic one every time, which is exactly what both Nation (2022) and Ur (2024) keep telling teachers: be selective, and better still, let the learner do the selecting.
Step 4. Put the right thing on the card
This is the rule I keep coming back to: save the chunk, not the naked word.
Write down “ta hänsyn till” (to take into account), not “hänsyn.” Save the word inside the sentence you met it in, so the grammar, the collocation, and the register travel with it. A word learned alone is a word you can recognise but never quite use, because you do not know what it leans on or whom it keeps company with. The phrase carries all of that for you.
The exception is basic, concrete vocabulary. An apple, a table, a chair. There, a single word and a picture are plenty, and a sentence would only clutter it. Save chunks for everything that behaves like language, and single words only for things you could point at.
What else belongs on a strong card, and what to leave off, is a craft of its own. I put the whole design system into a paid companion piece, How to Design Flashcards That Actually Stick, if you want to build cards that do the remembering for you.
Step 5. How to actually learn them (what Nation recommends)
A saved word is not a learned word, and most learners lose everything right here, at the gap between collecting and studying. If you want the fuller case for why a deck you simply reread barely teaches you anything, I wrote it up in my free post Why Your Flashcards Aren’t Working.
Nation (2022) lays out a word-card method built on decades of memory research, and it is specific. Three moves.
Choosing. Keep the cards few and avoid interference. Do not learn words that are too alike at the same time, no clusters of synonyms, opposites, or “all the colours,” because similar items cross-wire and block each other while they are fresh (Nation, 2022; this is also why Ur warns against teaching words in neat thematic sets).
Making. Word or chunk on one side, meaning on the other, with a picture where it helps. Keep each card simple and small enough to carry. The two-sided design exists for one reason: so you can hide the answer.
Using. This is the engine, so it is worth doing right:
Recall, don’t reread. Look at one side, pull the answer from your own memory, then check. That act of retrieval is what strengthens the word. Psychologists call it the testing effect (Pyc & Rawson, 2007), and it beats passive review by a wide margin (Kang et al., 2013).
Space the reviews. Return to a word after a gap, and stretch the gap as the word gets easier. Spacing beats cramming (Webb & Nation, 2017).
Go receptive first, then productive. First, recognise the meaning of the word, and later produce the word from the meaning.
Work in small packs and shuffle them. Start with a handful, grow the stack as it gets easier, change the order so you are not learning by sequence, and set aside the ones you already know to concentrate on the hard few.
Say them aloud and use them. Drop each word into a sentence of your own. A word you put to work outlives a word you copied ten times, because deeper processing lasts longer (Craik & Lockhart, 1972).
If you are not willing to come back and retrieve a word like this, do not save it in the first place. An unreviewed card is just a tree standing in front of the wood.
The map, in one breath
When a word stops you, walk it down the path.
Name the activity first. Extensive, for pleasure and volume? Mark lightly and read on, tap freely on a screen but keep almost none of it, dot the margin on paper, and sort it all after the chapter.
Intensive, to harvest? Stop, guess, then check.
For anything you might keep, run the three questions: will I meet it again, can I anchor it, did not knowing it stop me. Save only what passes all three.
Gather the survivors into one short list, put them on cards as chunks rather than bare words, and then learn them by recalling, spacing, and using the way Nation lays out, not by staring at a list.
You are not a net trying to catch every word in the sea. You are a filter, and a good filter lets most of the water through on purpose.
That is not a failure of diligence. That is how you finally start to see the wood.
Thank you for reading!
A small announcement: If you have been thinking about becoming a paid subscriber, this is a good moment to join. The annual membership is currently $120 USD and will increase to $180 USD on August 1. Join before then to lock in the current rate for as long as your subscription stays active.
When you read for pleasure, what is your own rule for which words to save? If you have a sharper filter than the three here, or a trick I have missed, I would like to read it in the comments. And if you try this map on your next chapter, tell me whether it brought you more clarity or less.
Want a truly personalized learning plan based on your level, goals, and how you actually learn? I design a clear, structured, science-based system tailored exclusively for you.
Consider becoming a Founding Member.
If this felt useful, feel free to share the post and publication with someone who might need it too.
About the author
Viktoria Verde, PhD, is an applied linguist, teacher, writer, and lifelong language learner behind How We Learn Languages. She writes about second language acquisition, multilingualism, and the real process of learning languages as an adult. Her work brings together research, teaching experience, and her own life as a multilingual learner to help readers build language-learning systems that actually fit their lives.
References
Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11(6), 671–684. https://doi.org/10.1016/S0022-5371(72)80001-X
Kang, S. H. K., Lindsey, R. V., Mozer, M. C., & Pashler, H. (2014). Retrieval practice over the long term: Should spacing be expanding or equal-interval? Psychonomic Bulletin & Review, 21, 1544–1550. https://doi.org/10.3758/s13423-014-0636-z
Nation, I. S. P. (2022). Learning vocabulary in another language (3rd ed.). Cambridge University Press.
Pyc, M. A., & Rawson, K. A. (2007). Examining the efficiency of schedules of distributed retrieval practice. Memory & Cognition, 35(8), 1917–1927. https://doi.org/10.3758/BF03192925
Ur, P. (2024). A course in English language teaching (3rd ed.). Cambridge University Press.
Webb, S., & Nation, P. (2017). How vocabulary is learned. Oxford University Press.
This is great advice, just what I needed to hear today as I'm reading! Thank you
Thanks, very good tips! I always find all the words important, I will use your questions.