Can You Cheat Harver? Retakes Get Linked
The shortcut Harver’s format invites is applying twice, once to see the questions. Harver built a detector for exactly that.
Harver’s fraud detection links applications it suspects belong to the same candidate, its proctoring flags tab switching and blocks copy and paste, and the result that decides your next step is a match against one job, not a high score to copy.
The one shortcut the format invites, and the detector for it
Most of a Harver journey has no clock: the situational judgement test, Personality Print and the video job previews are untimed. So the obvious way to game it is not speed but repetition: apply once to see the scenarios and the questionnaire, then apply again with a second email address and better answers.
Harver’s fraud detection answers that directly. It runs a machine-learning Retry Detection that flags applications suspected to belong to the same candidate and links them, and it cross-references every application to the vacancy regardless of where it came from. It is an add-on the employer switches on, so it is not everywhere, but a candidate cannot see whether it is on. A linked second application does not look like a fresh start; it looks like a candidate who applied twice.
Proctoring watches the window, not just the answers
Since April 2025 an employer can switch on proctoring for a vacancy. It comes as a set: periodic photos of the candidate, copy-and-paste blocking, and focus-change detection. In Harver’s words, whenever the candidate changes focus by switching tabs or windows, it detects it; the first few times it reminds the candidate not to, and if it happens too often it notifies the recruiter.
That closes the practical routes on the timed modules. Looking an answer up on a cognitive subtest means leaving the window, and pasting a sentence into the typing test is blocked by the test itself before proctoring even starts.
There is no single high score to copy
The result that decides your next step is not a subtest score. Harver’s employers screen on a matching score against the vacancy, with their own Good Fit and Great Fit brackets, and on a standard job profile each cognitive subtest is only one input to it. Personality Print makes the point most clearly: Harver’s report guide says high scores are not necessarily better and that each role has its own ideal range. An answer sheet for one employer’s job profile is the wrong answer sheet for another’s.
The situational judgement test is the same. A custom test is written for one employer from a job analysis with its own staff, and the three standard tests are keyed for different roles. A scenario answered for a retail vacancy can be keyed differently in a contact centre.
What actually moves a Harver result
Familiarity with each module’s format, because the journey is short and a first sight of an unusual item costs a large share of it. On the cognitive subtests, that means knowing that Components always wants two shapes and that Checking needs a steady rhythm. On the SJT, it means reading every option and finding the harmful one first. On the chat simulation, it means answering the customer who has waited longest and opening the policy before promising a refund. None of that needs a shortcut, and all of it survives the interview afterwards.