Automating User Provisioning with ServiceNow RPA and Document Intelligence

Posted on June 27, 2025 Virtuxient Technologies

In contemporary businesses, the user-providing mechanism is still one of the slowest, inefficient, and most prone to mistakes processes within IT business operations. Even though sophisticated ITSM tools are available, when introducing new users to systems such as GLOS, one can often encounter repetitive data input, multiple logins into the system, reliance on human administration, and inconsistencies in the processes.

Our team addressed this by deploying an end-to-end automation strategy that incorporates ServiceNow, Robotic Process Automation (RPA), and Document Intelligence to automate building the GLOS IDs, leading to a reduction in manual efforts.

It takes the form of a long-form case study describing the problem, design of the solution, technical implementation, and measurable business impact.

Background

The IT support department of the client had the responsibility of generating GLOS IDs for newly onboarded workers in different branches. The legacy process entailed:

• Collecting data manually via email
• Document verification by HR or L1 support
• GLOS manual navigation to develop user records

This way caused:

• Delays in employee onboarding
• High rate of mismatch of data and errors in provisioning
• Outcries by corporate stakeholders
• Anxious support teams and SLAs that turned over widely

Using this process, we aimed to have a fully automated end-to-end process using a scalable, secure, and compliant architecture that would be within the ServiceNow ecosystem.

Automation Objectives

The aims were:

• Automate the form of request reception, validation, and user creation.
• Remove manual control over documents and implement AI-based checks.
• Minimize the time spent on provisioning and IT support personnel.
• Ensure Traceability and transparency by updating centrally via ServiceNow.
• Create automation that will form future, reusable provisioning use cases.

Solution Architecture

At Virtuxient, we planned a tiered automation stack structure:

• ServiceNow: It was utilized to submit forms, orchestrate workflows, create an audit trail, and give notifications.
• Document Intelligence AI: It is a technology that is used in extracting and validating employee details from the Welcome Letters uploaded.
• RPA Hub (ServiceNow): Utilized to log into GLOS, to find the name of any existing users, and create a new record where needed.

Workflow Implementation

Step 1: Request Submission via ServiceNow

A ServiceNow catalog item was created to collect the user’s necessary details:

• User ID, Designation, Name, Branch, Mobile Number, Email ID.
• Requested For (dropdown)
• Mandatory Document Upload, like Welcome Letter (PDF or screenshot)

Most of the fields were auto-populated based on the “Requested For” user to minimize human input errors.

Step 2: Document Verification Using AI

Uploaded Welcome Letters were scanned using Document Intelligence. The AI model extracted identifiers (such as employee ID or name) from the letter and compared them against the ServiceNow form values.

• If values matched, the process proceeded.
• If the values mismatched, the Workflow was halted and a message was logged in the request:

“Employee details in the Welcome Letter do not match the selected user. Please review the request.”

No further action was taken unless corrected by the requester.

Step 3: GLOS System Login via RPA

On successful validation,  the RPA bot initiated an unattended login session into the GLOS system and navigated to the following path:

Profile → Sysadmin → Administration → Security → Users

The bot searched for the User ID.

• If already present, the system updated the request:
  “User ID already exists. No creation performed.”
• If not found, the bot was started to create the User.

Step 4: Automated User Creation

RPA entered the following details:

• User ID
• Full Name
• Default Password
• Required checkboxes
• Additional fields such as branch, designation, etc., are fetched from the ServiceNow form

Access profiles were assigned based on designation. This logic was designed to be configurable and extensible during full rollout.

Step 5: ServiceNow Request Closure and Notifications

Upon successful user creation, the RPA ServiceNow request with:

• Status: Completed
• Description: User created and credentials assigned
• Notifications are triggered to the requestor and the user

The default password was sent securely through the designated channel as per internal policy.

Technical Setup

The following modules were installed and configured:

1. RPA Desktop Design Studio

• Installed on Windows 10 Pro (Core i5+, 8GB RAM)
• .NET Framework 4.7.1+
• Integrated with the ServiceNow instance via Edge WebView2

2. Unattended Robot and Login Agent

• Installed on persistent Windows VMs
• High-density mode is enabled for multi-session handling
• Configured with both Basic and mTLS authentication options
• Proxy settings added for corporate network routing

3. Guided Setup and Plugin Management

All required Plugins and dependencies for the RPA Hub were installed using the Guided Set Up module in ServiceNow. This ensured consistency across dev, test, and production instances.

4. Compliance and System Properties

• WebSocket communication was enabled in the instance
• glide.cometd.websocket.enabled set to true
• All users and session data were stored within ServiceNow for audit and rollback.

Results & Business Impact

Reduction in Processing Time

• The previous average provisioning time was 2-3 hours.
• Post-automation provisioning time: <15 minutes

Elimination of Errors

• Providing manual mistakes: ~15 percent before
• Post automation errors: less than 1 percent.

Resource Optimization

• Effort on the support team was cut by 90 per cent.
• The IT team might also dwell on handling exceptions as opposed to conventional provisioning.

SLA Compliance

• Pre-automation: ~65-70 percent SLA conformance
• Once automation: 100 percent SLA compliance in the creation of new users.

Security and Audit

• ServiceNow collected all the provisioning activities.
• No confidential information had been transferred through the email or unsecured mediums.
• The integrity was provided through role-based access to the automation.

Reusability and Future Plans

The framework that is implemented to provide resources in this project has been extended to:

• Offboarding Automated Deactivation.
• Role and asset provisioning based on HR data
• Multi-level dynamic access workflows
• Azure AD and Active Directory integration

Key Takeaways

• Smart automation needs a close integration of the form data entry with document AI and back-end activities.
• It is no good relying on RPA without verification. Document Intelligence was instrumental with regard to risk mitigation.
• The RPA hub and Guided Setup offered by ServiceNow supported scalable forms of deployment with less scripting.
• The major metrics in enterprise automation include standardization, traceability, and SLA compliance.
• Automation should not make life less burdensome, but it should also enhance governance.

Conclusion

The project represents that RPA can automate a manual process that can be eradicated by other supporting technologies, such as Document Intelligence and ServiceNow, to make the process a secure, scalable, error-free cycle

When your enterprise still manually manages user provisioning, now it is the turn to get a modern and integrated automation solution.