Remote supervision (proctoring) for high-stakes assessment
Why this is more important than ever
39%
of job candidates used AI somewhere in the application process
29%
of them used AI to generate answers to assessment questions
Gartner Voice of the Candidate survey, 3,290 job candidates, 4Q24. Published 31 July 2025.
Two ways to check a result
Built-in monitoring
Included with every aptitude assessment, at no extra cost
-
Compares responses against pre-AI completion patterns
-
Flag goes to your admins, never to the candidate
Remote supervision
Recorded oversight of the session, priced per assessment
-
Integrity concerns returned alongside the score
-
Used where a wrong decision is expensive
How remote supervision works
Supervision is set up per project rather than switched on across your account, so you use it where the stakes justify it and nowhere else.
You decide which projects run supervised. Pricing is per assessment rather than per candidate, so a candidate sitting three aptitude tests counts as three. Supervised projects are set up directly in the platform, not triggered through your ATS integration.
The session is recorded through webcam and screen capture. Candidates need a computer with a webcam and microphone, a stable connection, a quiet room, and they give permission to record.
The provider reviews the session, or flags it automatically, and any integrity concerns come back alongside the candidate’s score. What it reports is behavior worth reviewing: repeated looking away, a second voice, more than one face on screen. It is not a verdict.
Which one fits your project
Most clients use the built-in check everywhere and add supervision to the small number of projects that warrant it.
HIGH VOLUME
Early-stage sifting
SENIOR ROLES
High-stakes hiring
REGULATED
Regulated hiring
BEHAVIORAL
Behavioral assessment
Bring us one campaign
Questions we get
The questions buyers and their compliance teams ask before a supervised project goes live.
Remote proctoring, which we call remote supervision, is recorded oversight of an assessment session. The candidate completes the test while webcam and screen are recorded, the provider reviews the session or flags it automatically, and any concerns come back with the score.
No. It highlights behavior that may indicate misconduct, such as repeated looking away, a second voice, or more than one face on screen. It is evidence for a human to weigh, and the decision stays with your team.
A computer with a webcam and microphone, a stable internet connection, a quiet room, and permission to record audio and video. The experience is more controlled than an unsupervised sitting, so tell candidates what to expect when you invite them.
Per assessment, not per candidate. A candidate sitting three aptitude tests counts as three supervised assessments.
The recommended action is a supervised retest of that one candidate, not a decision on the flag itself. A flag is a prompt to look again rather than a conclusion. Resetting a single candidate on a live project carries no extra charge.
All Saville aptitude tests except workplace English. It runs on new completions, appears in your dashboards and score extracts, and is never shown to the candidate or included in a candidate report. It is part of aptitude testing at no extra cost.
Tell us which roles worry you
We will come back with which layer fits, what it costs, and what your candidates would experience.