Two clocks
Ask two people who both play word games on their phone what they actually do, and you will often find they share almost nothing. One of them opens a single puzzle after breakfast, solves it or does not, and closes the app until tomorrow. The other opens the same category of game on a train and plays until the train arrives, with no target and no end.
Our desk at The Weekly Report has come to think of these as two different genres wearing similar tiles. They reward different design, they fail in different ways, and reviewing them against a single checklist produces nonsense. This piece sets out how we separate them, and where the dictionary — the part almost nobody inspects — decides both.
Reading which clock a game runs on
Before scoring anything we establish the shape of the game, because that determines what counts as a fault.
- Play a full session without touching settings. Whatever the game offers by default is its real thesis; everything behind a menu is an accommodation.
- Note the moment the game wants you to leave. A well-built daily puzzle has a clear closing beat. An infinite mode should have obvious, low-friction exit points rather than a wall of continue prompts.
- Return four hours later. Does the game greet you with new content, with a countdown, or with pressure? The greeting tells you what the designers think your attention is for.
- Return after skipping two days. A daily-rhythm game should absorb the gap without ceremony. If missing two days visibly costs the player something, the appointment has become an obligation.
- Only then start scoring. By this point we know which of the two things we are holding, and the rest of the log is written against the right standard.
Step four catches more titles than the rest combined. A great many word games are pleasant for a week and quietly coercive by the third, and the coercion almost always arrives dressed as a streak.
Why the word list carries everything
Whichever clock the game runs on, the same component decides whether players stay: the dictionary.
Consistency over size. A list of two hundred thousand entries that accepts one plural and refuses its neighbour is worse than a smaller list that behaves predictably. Players do not count entries. They build a mental model of what the game will accept, and every inconsistency demolishes it.
Legibility of refusal. When a word is rejected, the game must say so in language that distinguishes "not a word here" from "not a word". The first is a rule; the second is an insult, and players react accordingly.
Evidence of curation. Raw corpora contain abbreviations, fragments and technical residue. A list somebody pruned by hand feels different within a few sessions — the accepted words are ones a reader might plausibly meet in a book.
Teaching, quietly. The best implementations offer a one-line gloss on an unfamiliar accepted word without forcing it into view. That single feature converts guessing into learning and is the closest thing this genre has to a progression system.
We test all four with the same probe list on every candidate, so that our notes across a year of reviews are comparable rather than impressionistic.
The case for scarcity
A daily puzzle is doing something structurally unusual for a phone game: it is refusing to sell you more of itself. That refusal is the product. It gives the session an ending, and an ending is what turns a distraction into a small ritual.
It also produces the only genuinely comfortable social layer left in mobile gaming. When everyone faces the same grid on the same day, comparison happens in conversation rather than in a leaderboard, and nobody has to be told to invite a friend. Our desk marks this highly, and marks down any daily-format game that bolts a competitive ladder onto it — the ladder converts a shared object into a ranked one, and the conversation stops.
The case for the bottomless well
Infinite modes get unfairly patronised, and we think that judgement is wrong. There is a real, distinct pleasure in a format with no terminus: the player is not completing anything, they are practising. Skill improves in a way daily formats cannot show, because one puzzle a day is too thin a signal to feel yourself getting better on.
What an infinite mode owes the player in exchange is an honest exit. That means no interstitial pleading at the moment of leaving, no escalating reward that is always one round away, and a state that survives being closed mid-word. We check the last one on every candidate by force-closing the app in the middle of a move.
Alone and asynchronous
Word games remain the genre where asynchronous play works without apology. A turn answered two days later is fine here, and games that treat that delay as a problem to be solved with reminders have misread their own format.
We test solo play as a first-class mode rather than a tutorial. Many readers will never play against a person, and a solo experience with no variety is a genuine defect regardless of how good the multiplayer is.
What our number means
Our editorial score reflects logged play by our own testers and is deliberately not the Play Store star average — the two are measuring different things, and this genre in particular collects star ratings that reward familiarity over craft. How the headings are weighted is documented on our methodology page, the current standings are on the ranking, and the people writing these logs are listed on the about page.