SCIM provisioning
Without SCIM, every joiner, mover, and leaver depends on someone remembering. What provisioning automation covers, and where gaps stay manual.
of manual offboarding misses SaaS applications
2026
•
Reco
SaaS applications in the average organization
2026
•
Zylo
of employees still have access to a previous employer's account
2022
•
Beyond Identity
Apps without provisioning depend on a human remembering every joiner, mover, and leaver, in every system, every time.
what it is
SCIM (System for Cross-domain Identity Management) is a protocol that allows an identity provider to automatically manage user accounts in connected SaaS applications. When a new employee is added to Microsoft Entra ID, Google Workspace, or Okta, a SCIM-enabled application creates the user's account automatically. When the employee changes roles, SCIM updates the account. When the employee leaves and the directory account is deactivated, SCIM deactivates the application account.
SCIM is the technical implementation of the joiner-mover-leaver (JML) process at the application layer. It removes the dependency on manual steps and on someone remembering to act.
Without SCIM, each application requires a separate, manual action for every JML event. A new hire needs an account in each tool. A role change needs permissions updated in each tool. A departure needs the account closed in each tool. As the number of applications grows, the number of manual steps grows with it.
why it accumulates
The gap between what SCIM could automate and what actually gets automated has several causes.
SCIM support is not universal. Many SaaS tools support SCIM only at higher pricing tiers. Organizations with mixed plans across their tool stack often have SCIM available in some applications and not in others, creating an inconsistent posture.
SCIM setup requires initial configuration. Enabling SCIM between an identity provider and an application requires technical setup: configuring the SCIM endpoint, granting the right permissions, testing provisioning flows. For each new application, someone needs to do this work. It often gets deprioritized when a tool is first adopted.
Applications get added faster than SCIM gets configured. A team deploys a new tool. SCIM configuration is on the list but not the immediate priority. The tool runs on manual provisioning. More tools are added. The backlog of non-SCIM applications grows.
what it costs you
Offboarding omissions. Applications without SCIM rely on someone manually closing accounts when an employee leaves. In environments with many applications, this step is often missed for some tools. The result is active accounts for former employees in applications that IT is not actively managing.
Provisioning delay for new joiners. Without SCIM, new employees may wait for account setup across applications because each one requires a separate manual action. Access provisioning depends on someone's availability rather than on the directory event that should trigger it.
Role change drift. When an employee changes roles, their permissions in applications should update to reflect the new role. Without SCIM, this requires manual action in each application. The action is frequently missed or deferred, leading to permissions that no longer match the current role.
Audit evidence gaps. Auditors checking access provisioning and deprovisioning expect to see systematic, timely handling of JML events. Manual processes produce inconsistent evidence. Some applications are handled promptly. Others have gaps. The inconsistency is visible in the audit trail.
what works
The starting point is a support map: each application in the registry checked for whether SCIM is supported and on which pricing plan, information most major vendors publish in their developer documentation or admin help center. The map usually splits the stack three ways: applications where SCIM is live, applications where it is available but unconfigured, and applications where it does not exist or sits behind a tier upgrade.
Risk ordering does the prioritization. Applications holding customer records, financial data, employee information, or source code come first, because those are the tools where a missed offboarding carries real consequence. Enablement itself follows a known shape: a service account or app registration in the IdP, the SCIM endpoint configured in the application, IdP attributes mapped to application fields, and a test user pushed through the full provisioning flow before anything goes live.
Group-based provisioning is what makes the automation cheap to operate. Access tied to IdP group membership means an employee added to the Sales group gets the sales tools automatically and loses them on removal, with no individual assignments to manage and role changes handled by the same mechanism as joiners and leavers.
The applications that remain manual deserve an explicit list rather than an assumption. Every tool without active SCIM belongs on every offboarding checklist and in every access review, precisely because nothing automatic will catch a miss there. A visible list also tends to shrink over time, since each entry is a standing argument for the plan upgrade or the configuration work, while an invisible gap just persists.
practical guides you might find useful
let's start with a conversation
Most first conversations start with not quite knowing what you have or where to begin. That's normal, and it's exactly where we're useful.
Tell us what prompted this. An upcoming audit, an incident, a client's security questionnaire, or just a sense that things have gotten messy.
We'll take it from there

+48 783 762 997
julian@unshadowit.com



.svg.png)


