What it is
The Third-Party Access Request Form is a structured intake that turns an informal 'can you give the vendor a login?' into a reviewable, time-boxed, least-privilege grant. It captures who is requesting access (the internal sponsor accountable for it, their manager, the vendor company, the named individuals and their emails, and the contract/PO/SOW reference), exactly what they need to reach (specific hosts and applications, environment, access method, privilege level, and any PII/PHI/PCI involved), why (business justification), for how long (hard start and end dates), and under what controls. It ends with a controls checklist and a signature/approval block, so a completed form is the audit evidence that every external connection was sponsored, scoped, and approved before it existed.
The controls section is the heart of the form: least privilege (access limited to named systems, no broad network reach), time-boxed (auto-expires on the end date, no standing access), phishing-resistant MFA on every login, device posture (vendor device meets policy or access is via a managed jump host/VDI), session recording for privileged or production targets, named identity (tied to a person, not a shared vendor account), a signed acceptable-use agreement, and an offboarding trigger to revoke on contract end or task completion. Each is checked off before access is provisioned. The approval block requires sponsor signature, system/data-owner approval, security review for production or sensitive data, and a record of who provisioned the access, when, and with what expiry.
It embeds a short review-questions section that prompts the approver to challenge the request: does the scope match the task or is it broader than needed; is there a hard end date and an early-revoke trigger; is the vendor connecting from a managed surface; will privileged sessions be recorded. Used consistently, the form is how third-party access stops being the orphaned account a breach rides in on. It operationalizes the access tiers in the policy template and pairs with the acceptable-use agreement the vendor signs.
What it's used for
Third-party access is one of the most common breach vectors precisely because it's often granted casually and never cleaned up. The form imposes the discipline of sponsorship, scoping, and expiry on every external grant. It's used to:
- ✓ Intake and approve a vendor, contractor, or partner remote-access request through a single reviewable document, replacing ad-hoc 'just give them a login' messages.
- ✓ Force least-privilege scoping by naming the exact systems, environment, and privilege level — trimming the request from 'broad network access' down to what the task actually needs.
- ✓ Set a hard expiry and an early-revoke trigger so no vendor grant becomes standing, permanent access that outlives the work.
- ✓ Require the right controls before provisioning: phishing-resistant MFA, device posture or a managed jump host/VDI, session recording for production, and a named (not shared) identity.
- ✓ Route the request through the right approvals — sponsor, system/data owner, and a security review for production or sensitive data — so no one party can wave access through.
- ✓ Capture the data sensitivity (PII/PHI/PCI) the vendor will touch, so compliance obligations are flagged at request time rather than discovered in an audit.
- ✓ Create the audit trail that proves every external connection was sponsored, scoped, time-boxed, and approved — the evidence an auditor or incident responder asks for first.
Who uses it
Vendor access requests cross the boundary between the business that wants the access and the security and system owners who must vouch for it. The form gives each party their role.
Context & good to know
Vendors and contractors are a structurally weak point because their access is requested by the business, granted as a favor, scoped generously to avoid friction, and then forgotten. The result is the pattern that shows up in breach after breach: an external account, often over-privileged, often shared, often still active long after the work ended, connecting from a device you don't control. The form attacks every part of that pattern — named sponsor, named individuals, trimmed scope, hard expiry, managed surface, recorded sessions.
Scope-trimming is the single biggest risk reduction on the form, and it's also the one most easily skipped. Vendors routinely ask for more than the job requires — 'admin, just to be safe' — because broad access means fewer follow-up tickets for them. The form's review question 'does the requested scope match the task, or is it broader than needed?' exists to make the approver push back. Pinning access to the named systems at the minimum privilege the task needs is what limits the blast radius if that vendor account is ever compromised.
Expiry is the second pillar. Access without an end date becomes the orphaned account a breach rides in on — the contractor who finished six months ago whose login still works. The form requires both a hard end date and a trigger that revokes early if the work finishes sooner, plus an offboarding notification so an owner actually pulls the access. This turns 'we'll remember to remove it' into a built-in expiry that doesn't depend on memory.
The managed-surface decision is where this form connects to the rest of your zero-trust controls. You can't posture-check a device you don't own, so an unmanaged vendor laptop is an unknown quantity — which is why the form steers privileged or production access through a managed jump host or VDI rather than direct connection. Combined with session recording (your only after-the-fact account of what an external party did) and a signed acceptable-use agreement, the form ensures third-party access is contained, observed, and consented to before it ever exists.