How to Play WordRank

Almost everyone hits the same wall the first time they play WordRank: you type a word, and instead of a right-or-wrong tick you get a number — something like #2,481. That is not a bug. That number is the game. This guide starts exactly there and gets you genuinely playing in about three minutes.

What that number actually means

WordRank is a semantic word game, not a spelling game. Every round has one hidden secret word. When you type any word, the game measures how close it is in meaning to the secret word and returns a rank: rank #1 is the answer itself, #2 is the next-closest word in meaning, and so on. So #2,481means there are 2,480 words in the game’s vocabulary that are closer to the answer than your guess. The smaller the number, the closer you are.

This mechanic belongs to the same family as Contexto, built by a developer in Brazil, and the earlier Semantle. WordRank differs by being unlimited, multilingual, and by showing distance as one continuous color gradient rather than a few discrete bands. For the underlying semantic-similarity idea, the WordRank home page has a short explainer.

What “close in meaning” really means

Picture every word as having a circle of friends — the words it tends to show up with. “Coffee” hangs out with morning, cup, bitter, awake, shop. “Tea” hangs out with almost the same crowd. That is exactly why the game treats coffee and tea as close: not because they are the same drink, but because people talk about them using the same words.

This is also why you will hit surprises. Guess “king” and “queen” ranks close — no shock there. But “chess” ranks close too. Why? Because people constantly mention kings and chess in the same breath. The game learns how words are actually used together, not their dictionary definitions. “Close” means used in similar ways— not “same kind of thing.”

That flips your strategy. A strong next guess is not another word in the same category — it is a word that often appears alongsideyour warmest guess. If “ocean” is warm, “wave,” “sailor,” and “tide” may beat “lake,” even though a lake is more obviously “the same kind of thing.” Behind the scenes this is powered by a language model trained on billions of sentences — but you never have to think about that to play.

Start playing in three steps

• Type any word. It does not need to relate to the answer — your first word is for scouting, not for scoring.
• Read the rank. Green (1–300) means very close, amber (301–1500) means warm, red (1500+) means far away.
• Narrow down from the feedback.When a word’s rank drops sharply, keep testing words along that same meaning.

There is no guess limit and no penalty for a bad word. The fastest way to play is not caution — it is to probe widely first with words of very different meanings, locate roughly which semantic region the answer lives in, then close in.

Why spelling does not matter at all

This is the most counter-intuitive part for new players. In WordRank, “sea” can rank higher than “seal” even though they look nearly identical. The game judges how words are actually used in real language, not which letters they share. Close in meaning is not the same as similar in spelling — which is why synonyms, broader and narrower terms, and words that frequently appear together can all rank highly. Once this clicks, you have made the mental shift from guessing letters to guessing meaning.

What to do when you get stuck

Everyone gets stuck. When you do, stop circling the same word — jumping to a completely different semantic category (say, from “emotions” to “objects”) often breaks the round open. You can also use the past puzzles as a practice ground where the answer is known, purely to train your opening-word instincts. When you are ready, try today’s puzzle— or, if you want a change of pace, browse other games like Contexto.