The platform

Built to integrate, secure by design, fair by evidence

ATS integration Platform security AI validation
API

Connect iTestHub to your ATS

Our client API is live. Your system orders an assessment, we return the candidate's link and its expiry, and you pull back a structured result carrying the exact test version and norm group it was scored against. You can order identified candidates — or anonymously, sending only your own reference, so no candidate personal data ever reaches us.

  • Five endpoints: the catalogue, ordering, the order listing you reconcile against, results and report PDFs
  • A signed assessment.completed webhook, so you do not have to poll
  • One key per client — scoped, hashed at rest, rotatable and revocable
  • Idempotent ordering: a retry never creates a second candidate

The API in full — quickstart, payloads and how an integration is built

Example — order a test from your ATS
POST /api/v1/orders
{ "candidate": { "first_name": "Ada",
    "last_name": "Lovelace", "email": "…" },
  "assessments": ["verbal-reasoning"],
  "reference": "REQ-4471" }
201 Created
{ "id": "ord_9c1d4f0b…",
  "candidate_url": "https://app.itesthub.com/…",
  "expires_at": "2026-10-15T17:00:00Z" }
Platform security

Your candidates' data, protected

Assessment data is sensitive personal data. The platform is built around that fact.

Two-factor authentication
Every client and staff account is protected by mandatory two-factor authentication with recovery codes
Encryption throughout
Data encrypted in transit and at rest, with UK/EU hosting and GDPR-compliant processing
Least-privilege access
Candidates reach tests through unique secure links — no accounts, no passwords to lose. Admins see only their own client data
AI validation

Confidence that the score is the candidate's

Sophisticated measures detect whether someone is using AI tools or other methods to cheat — so the result you act on reflects the candidate, not their browser tabs. Candidates are told up front that assessment activity is monitored to protect the fairness and validity of results.

Behavioural signals
Response times, answer patterns and in-test behaviour are analysed for signatures inconsistent with unaided human performance
On-screen protection
Copy protection on test pages deters pasting questions into AI tools — with documented exceptions for screen readers and assistive technology
Statistical validation
Our Continuous Research & Development Programme routinely investigates AI-assisted responding, so detection keeps pace as tools change
Flag, don't just reject
Suspect sessions are flagged for your review with the evidence behind the flag — you stay in control of the decision

Fairness comes first: approved assistive technology and agreed reasonable adjustments are always supported, and detection methods are validated to avoid disadvantaging any group.