- Post History
- Subscribe to RSS Feed
- Mark as New
- Mark as Read
- Bookmark
- Subscribe
- Printer Friendly Page
- Report Inappropriate Content
a week ago
Agentic AI for Employee Journey Management: Overview
This article presents an overview of agentic AI workflow capabilities available within Employee Journey Management (EJM) in ServiceNow HR Service Delivery (HRSD).
Agentic AI within HRSD spans a broad range of applications, from case management and hiring to employee development and lifecycle transitions. Within the context of EJM, agentic AI use cases are purpose-built to support managers and employees at the moments that matter most in the employee lifecycle — welcoming new hires with structured, personalized ramp-up plans, and preserving critical institutional knowledge when employees depart.
These agents do not just assist; they act. They gather data, make decisions, generate outputs, and coordinate handoffs across multiple systems — all with deliberate human checkpoints at points where judgment matters most. The result is reduced administrative burden for managers, faster time-to-productivity for new hires, and greater organizational continuity during offboarding.
The following Agentic AI workflows are available out of the box for Employee Journey Management:
- Generate Onboarding Ramp-Up Plan (Manager persona)
- Generate Offboarding Knowledge Transfer (Manager and Employee persona)
I. Generate Onboarding Ramp-Up Plan
Purpose
As a manager, use the Generate Onboarding Ramp-Up Plan agentic workflow to automatically produce a structured, personalized ramp-up plan for a new hire — without starting from scratch.
Creating onboarding plans is one of the most time-consuming responsibilities a manager faces when a new hire joins their team. The burden compounds when plans need to be tailored to the employee's specific role, skill gaps, and team context. If the plan falls short, managers often end up spending even more time on ad-hoc training. This agentic workflow eliminates that cycle entirely.
The workflow triggers automatically when a new hire's joining date approaches, runs a team of AI agents to collect data, analyze skill gaps, recommend learning, catalog items and generate a structured multi-stage plan — then surfaces it to the manager for review and refinement through a conversational interface before it reaches the new hire.
How it works
The workflow is driven by an AI Agent Orchestrator and employs two AI agents working in sequence.
Step 1: Auto-trigger
The system continuously monitors active onboarding journeys. When a new hire's joining date is five or fewer days away, the workflow triggers automatically — no manual initiation required. The day-count threshold is configurable, giving admins flexibility to adjust the lead time based on organizational needs.
Step 2: Ramp-Up Plan Generation AI Agent
Once triggered, the Ramp-Up Plan Generation AI Agent runs autonomously across four functions:
Data collection — The agent gathers a comprehensive picture of the new hire and their context. It pulls from the onboarding journey record, HR cases, team plans, user and manager profiles, job details, and available learning resources. It also extracts the employee's existing skills from their resume and identifies skill gaps by cross-referencing the job description and interview notes.
Learning recommendations — Based on the identified skill gaps and role requirements, the agent retrieves targeted learning courses from the connected learning platform. Courses are matched to close specific gaps, not just general onboarding content.
Plan generation — The agent builds a structured onboarding plan organized into up to four stages:
- Stage 1: Relevant for Your Role — role-specific learning courses (up to 5 per stage)
- Stage 2: Trending Courses — broader learning trending across the organization
- Stage 3: Get to Know Your Team — always included; contains three core team-related tasks plus up to six 1:1 meeting tasks with teammates (nine tasks maximum in this stage)
- Stage 4: Initial Setup — Frequently requested catalog items & order guides in the manager's team
Any stage without available courses is skipped automatically, keeping the plan clean and relevant.
Scheduling and task assignment — The agent assigns tasks and learning items to both the employee and manager, with calculated due dates to ensure timely completion relative to the start date.
Step 3: Manager reviews the draft plan
Once the plan is generated, the draft stages are added to the journey in a DRAFT state — meaning the new hire cannot see them yet. The manager receives a notification in Now Assist in Virtual Agent: "Ramp-up plan stages added — review now." The manager can view the stages directly in the Now Assist chat window or open the Journey Detail page for a full view.
Step 4: Ramp-Up Plan Reviewer AI Agent (conversational editing)
If the manager wants to make changes, they interact conversationally with the Ramp-Up Plan Reviewer AI Agent. This agent guides the manager through a structured four-phase interaction:
- Present plan — Displays the current ramp-up plan with full context from the employee record and journey.
- Collect input — Accepts manager instructions to create, update, or delete tasks and stages. All changes are confirmed before applying.
- Review — Redisplays the updated plan and checks whether further changes are needed.
- Exit — Ends the session once the manager is satisfied.
If no edits are needed, the manager can proceed directly without engaging this agent at all.
Step 5: Publish
Once satisfied, the manager signals completion through Now Assist. The manager navigates to the Journey Detail page and clicks Publish to activate the stages — at which point the new hire can see and begin working through the plan.
The manager stays in the loop at two deliberate checkpoints: draft review and final publish. Everything in between — data collection, skill gap analysis, course matching, plan generation, and conversational editing — is handled by the agents.
AI agents at a glance
|
AI Agent |
Role |
|
Ramp-Up Plan Generation AI Agent |
Runs autonomously after trigger. Collects employee, journey, team, and job data. Identifies skill gaps from resume and interview notes. Retrieves matching learning courses. Generates a multi-stage onboarding plan with scheduled tasks for both employee and manager. Notifies manager via Now Assist when the draft is ready. |
|
Ramp-Up Plan Reviewer AI Agent |
Activated when the manager requests changes. Presents the current draft plan, collects editing instructions conversationally (create, update, or delete tasks and stages), applies and confirms each change, redisplays the updated plan, and exits when no further changes are needed. |
End-to-end flow Overview:
Figure: Example conversation flow in the instance
Prerequisites and dependencies
The following plugins must be installed and active for the workflow to function:
|
Plugin / Component |
Identifier / Notes |
|
Now Assist for HR Service Delivery |
sn_hr_gen_ai |
|
Skills Foundation |
sn_skills_int |
|
Learning |
sn_lep |
|
Hiring Core |
sn_ta_hiring_core — required for resume and interview note access |
|
Journey Designer |
sn_jny |
In addition, the Onboarding Ramp-Up trigger must be configured to use the Employee Center portal and set to Active in AI Agent Studio.
Refer to our product documentation for additional configuration details.
II. Generate Offboarding Knowledge Transfer
Purpose
As a manager, use the Generate Offboarding Knowledge Transfer agentic workflow to capture, organize, and transfer a departing employee's critical institutional knowledge to the right successors — before their last working day.
When employees leave, they take with them years of accumulated knowledge: documents, processes, project context, and organizational relationships that rarely exist anywhere else. The gap they leave behind can slow down successors and disrupt team continuity for months. This agentic workflow addresses that risk head-on.
The workflow automatically discovers documents authored by the departing employee, uses AI to categorize them into meaningful groups, creates a structured knowledge transfer record with assigned tasks, and facilitates a consent-driven review with the departing employee before any content is shared with the manager or successors. The result is a repeatable, consistent knowledge transfer process that runs without manual coordination.
How it works
The workflow is driven by an AI Agent Orchestrator and employs three AI agents across two sequential phases: manager initiation and employee review.
Phase 1: Manager initiation
Step 1: Auto-trigger
When an offboarding journey is created with the Journey accelerator plan type set to Agentic AI Offboarding Plan Type, the workflow triggers automatically. The manager receives a notification from Now Assist in Virtual Agent the next time they access the Employee Center.
Step 2: Knowledge Transfer Gather Inputs AI Agent (manager-facing)
The manager interacts conversationally with this agent to initiate the knowledge transfer. The interaction follows a structured sequence:
- Now Assist surfaces the departing employee's name, their departure date, and asks whether a knowledge transfer is needed.
- The manager confirms by entering Yes.
- The agent asks the manager to specify a time range (in months) for retrieving documents authored by the departing employee.
- The agent searches connected document sources — specifically SharePoint Online via AI Search — and retrieves up to 25 documents authored within the specified period.
- A Knowledge Transfer record is created, with individual Knowledge Transfer Resource records generated for each discovered document.
- AI-powered categorization groups the documents into meaningful clusters based on content and context — using the Knowledge Transfer Document Grouper skill configured in Now Assist Admin. Each document is assigned a relevancy score.
- A Knowledge Transfer stage is generated within the employee's offboarding journey in Journey Designer, with tasks assigned and ready to publish.
- The manager receives a link to the offboarding journey and confirmation that the departing employee will be notified to review the content.
Step 3: Manager publishes tasks
The manager navigates to the offboarding journey and publishes the Knowledge Transfer stage tasks. This makes the review tasks visible to the departing employee in their Employee Center.
Phase 2: Employee review
Step 4: Knowledge Transfer Employee Review AI Agent (employee-facing)
A second AI agent engages with the departing employee through Now Assist in Virtual Agent. The employee reviews the categorized document list, can add or remove documents, and approves the final knowledge transfer summary. Once approved, the manager is notified and receives the finalized summary to share with designated successors.
This consent-driven phase ensures that knowledge transfer is a collaborative process, not a one-sided extraction — protecting employee trust while preserving organizational continuity.
AI agents at a glance
|
AI Agent |
Role |
|
Knowledge Transfer Gather Inputs AI Agent |
Manager-facing. Confirms whether knowledge transfer is needed, collects time range input, searches SharePoint Online via AI Search for authored documents (up to 25), creates the Knowledge Transfer record and Resource records, triggers AI categorization with relevancy scoring, generates the Knowledge Transfer journey stage, and provides the manager with a tracking link. |
|
Knowledge Transfer Document Grouper (Now Assist Skill) |
Categorizes discovered documents into meaningful groups based on content and context. Assigns relevancy scores to each document. Runs as part of the document discovery step — not a conversational agent, but a key AI component in the workflow. |
|
Knowledge Transfer Employee Review AI Agent |
Employee-facing. Presents the categorized document list to the departing employee, allows additions or removals, collects approval, and notifies the manager once the knowledge transfer summary is finalized and ready for sharing with successors. |
End-to-end flow
Fig- Manager Flow in the instance
Fig- Manager Flow in the instance
Prerequisites and dependencies
The following plugins must be installed and configured:
|
Plugin / Component |
Identifier / Notes |
|
Now Assist for HR Service Delivery |
sn_hr_gen_ai |
|
Journey Designer |
sn_jny |
|
AI Search |
glide.ais |
|
External Content Connectors |
sn_ext_conn |
|
External Content Connectors — SharePoint Online |
sn_ext_conn_spo — required for document discovery from SharePoint |
The following components must also be activated or configured before use:
- AI agent triggers for the offboarding knowledge transfer workflow (Offboarding Knowledge Transfer trigger and Knowledge Transfer Record Created trigger) — both must be set to Active and configured to use the Employee Center portal in the AIA Trigger Configuration table.
- Agentic AI Offboarding Plan Type field value in the Journey accelerator plan type field on the journey configuration.
- Knowledge Transfer Document Grouper skill activated in Now Assist Admin.
- Now Assist in Virtual Agent added to the Employee Center display experience.
Refer to our product documentation for step-by-step configuration instructions.
Related Links
We have other out-of-the-box (OOB) agentic AI workflows and agents for HR as well. Refer to the list of out-of-the-box AI agents article for the cumulative list.
Refer to the Now Assist for HR quick start guide for more information on all Now Assist for HR offerings, including generative AI capabilities.
