What Is a Good HireVue Game Score? How the Bar Is Really Set

TL;DR: There is no universal "good score", because there is no published pass mark. Your raw performance becomes a trait estimate, which becomes a percentile against other applicants, which is compared to a bar the employer set for that specific role. That bar is invisible, it moves between companies, and it moves between roles at the same company. What you can control is where you land on the curve — and most of the curve is made of people who walked in cold.

Why nobody can tell you the number

Three separate layers sit between what you do and whether you advance, and each one blocks a simple answer.

Layer one: your raw play becomes a trait estimate. The games are adaptive. Get a sequence right and the next is longer; miss and it eases back. The software is not counting correct answers, it is triangulating the highest level you can hold reliably. Two candidates who "got 14 right" can receive very different estimates depending on where in the run those hits fell.

Layer two: the estimate becomes a percentile. You are ranked against a comparison group — typically other applicants for that kind of role. This is the step that makes an absolute score meaningless: the same working memory span is a strong result in one applicant pool and an ordinary one in another.

Layer three: the employer chooses the cut. A quantitative trading desk and a retail graduate scheme buy the same games and screen at completely different points on the curve. Some employers cut hard at the game stage; others use it as one weighted input beside interviews or a strengths assessment. None of them publish where the line sits, and a recruiter who told you would be handing you an unfair advantage over other candidates.

So when a forum post claims "you need the 70th percentile to pass HireVue", it is describing at best one employer, one role, one cycle — and quite possibly nothing at all.

What "good" actually means in practice

Strip away the mythology and a usable definition remains: a good score is one comfortably above the median of a largely unprepared applicant pool, and consistent across the whole battery.

Two parts of that sentence do the work.

"Above the median of an unprepared pool." This is the opportunity, and it is bigger than most candidates believe. A large share of any cohort has never seen these game types before and loses the opening seconds of each timed task decoding the rules. In a percentile ranking, not losing those seconds moves you up before ability is even measured. Familiarity is not a marginal edge here; it is most of the gap.

"Consistent across the whole battery." Your result is a profile, not a single number. One weak game rarely sinks you on its own, but a jagged profile — excellent memory, poor attention control, erratic speed — reads worse than an even one at the same average, because employers are matching a shape, not a total. The cheapest points are always in your worst game, never your favourite.

Judging your own runs without a pass mark

You cannot see the employer's bar, but you can absolutely see your own position — and that is the number worth working with. Every game here scores server-side against fixed norm tables and returns a percentile, so a practice run tells you where you sit relative to a comparison curve rather than giving you a bare score.

Used properly, that gives you three things a "target score" never could:

• A baseline. Play each game once, cold. That is your honest starting point and it takes about fifteen minutes across the set.
• A ranked weak list. Sort your percentiles. The bottom two are where your remaining preparation time belongs, regardless of which games you enjoy.
• A trend. Repeat runs show whether a category is actually moving or whether you are just having good and bad days. Working memory span, mental arithmetic speed and attention control all respond measurably within days; that improvement is real, and it is the part of your score that preparation can genuinely shift.

Start with a free practice run across the games, then dig into the specific categories: digit span for working memory, numerosity for rapid quantity judgement, mental math for numerical fluency under a timer, and reaction time for raw processing speed.

What does not move your score

Worth saying plainly, because a lot of preparation energy gets spent here for nothing.

Chasing a rumoured threshold. Optimising toward "the 80th percentile" tells you nothing about what to do differently. Training your weakest category does.

Trying to game the behavioural tasks. Where an assessment includes risk or trait tasks, there is no high score to reach, and manufactured answers tend to produce an inconsistent profile that reads worse than an honest one.

Grinding the game you are already best at. It feels productive and moves your overall profile least.

Taking it at 1 a.m. on the deadline. Working memory and reaction speed are both sleep-sensitive, and the deadline hour is when the least prepared half of the pool submits. Take it when you are sharp, inside the window you were given.

The bottom line

Stop looking for the number. The bar is relative, invisible, and set by someone whose priorities you cannot see — but it is set against a field that is mostly unprepared, and that is the whole opportunity. Find your baseline, fix your weakest category, and make your one attempt with nothing about the format unfamiliar.

If your assessment is close, the 24-hour prep plan sequences the time you have; can you fail a HireVue game assessment explains what happens on the other side of the bar; and the complete guide covers the full battery game by game.

FAQ

What is a good score on HireVue games? There is no published pass mark. Results become percentiles against other applicants and are compared to a bar the employer sets per role, so "good" means comfortably above the median of a largely unprepared pool and even across the whole battery — not a specific number.

Is there a minimum percentile to pass? Not a universal one. Different employers, and different roles at the same employer, cut at different points, and none of them publish where. Any specific threshold you read online describes at most one employer in one cycle.

Will I be told my score? Usually not. Some employers send a short feedback summary; most simply advance or reject without a number. A week or two of silence is normal screening time rather than a rejection.

Does one bad game fail the whole assessment? Rarely on its own — you are scored across the battery as a profile. But an uneven profile reads worse than a level one at the same average, which is why your weakest category deserves your remaining practice time.

Can I improve my score with practice? For the cognitive games, yes and measurably — working memory span, numerical speed and attention control all respond within days. For behavioural or risk tasks, no; those read traits, and trying to perform a profile usually backfires.

How do I know where I stand before the real test? Play the practice games and read your percentile. Every run here is scored on the server against fixed norms, so you get a position on a curve rather than a bare score, which is the same shape of information the real assessment produces.

GamePrep is an independent practice platform and is not affiliated with HireVue or any assessment vendor. Percentiles shown here come from our own fixed norm tables and are not the employer's benchmark — no vendor publishes its cut scores. We provide practice only — no leaked questions or answers.