BestEver

Our testing routine, start to finish

Sixty-two games went in and five came out. This is every stage they went through, in order.

Six passes before a game gets a number

  • Shortlist

    We start each cycle with everything new or newly updated on the major browser portals, plus anything readers have sent in. This cycle that came to sixty-two titles. Anything requiring a download, an account before the first round or a payment to start is removed here and never reaches a device.

  • First-load check

    Each surviving game is opened cold on three devices with the cache cleared, and we time the gap between the click and the first frame you can actually control. Anything past twenty seconds on a fibre connection is cut. This stage alone removed nineteen games.

  • Twenty-minute play test

    A single unbroken session on a desktop, with notes taken on what the game explains, what it hides and where a first-timer gets stuck. Games that stop being interesting inside twenty minutes rarely recover later, and this is where most of the remaining shortlist falls away.

  • Mobile and controls pass

    The same game on a mid-range Android phone, in portrait and landscape, and on a five-year-old laptop trackpad. Touch layouts that require a third thumb, buttons that sit under the browser chrome and text too small to read all cost points here.

  • Ads and fairness audit

    We count interruptions across ten rounds, note whether any advertisement is unskippable, and check what money buys. A cosmetic shop is fine. A paid weapon that changes a match is not, and it caps a game below four regardless of everything else.

  • Re-check and rank

    A fortnight later, every survivor is played again by a second person who has not read the first set of notes. Where the two scores differ by more than half a point, a third person breaks the tie. Only then does a game get a number and a position.

Habits that make browser gaming better

Most complaints about browser games are really complaints about how they were opened. A few small habits remove the majority of the friction, and none of them require any technical knowledge.

  • Play a new game on the device you will actually use it on, not the best one you own
  • Clear the tab and reload before deciding a game is slow, browser caches lie
  • Check the settings menu first, most browser games hide a quality toggle there
  • Use a wired connection for shooters and mobile data for everything else
  • Mute the tab rather than the game, browser volume survives a reload
  • If a game asks for an account before the first round, close it and move on
Three test devices lined up on a desk beside a stopwatch

What happens after a score is published

A published score is not the end of the process. Browser games update silently and frequently, and a title that earned its position in March can be a different product by June. Every game on the board is re-opened at the start of each quarter and run through the first-load check and the ads audit again. Those two stages catch almost every meaningful regression, because the changes that hurt players most are slower loading and heavier advertising.

Reader reports matter more than they probably should, and we would rather that than the alternative. Three people telling us that a game now shows an unskippable advertisement before the first round is enough to trigger an immediate re-test rather than waiting for the quarter to end. If you tell us which device and which browser you were using, that report is worth several times more, because half of the problems people report turn out to be specific to one browser version rather than to the game.

We do not accept payment for a position, and no game on this board knows it is being tested. Where a link to a game earns us a commission, that is disclosed in the footer of every page and it has no bearing on the score, for the simple reason that the scoring happens before anyone looks at whether a commission exists. A game we earn nothing from can and does outrank one we do.