Employee support assistant

Concepts covered

Step breakdown

1. Define the system to deployStart from this HR operating system and confirm the business outcome, source signals, owners, and review gates.
2. Create the Frontline assetsBuild the workflow canvas, agent prompt, channels, CRM records or tables, routing logic, and Max Activity evidence described in the blueprint.
3. Launch the narrow versionDeploy the smallest reliable version first, keep human review visible, then use activity and analytics to expand the system.

What you will build

An internal HR assistant that answers employee questions about policies, benefits, and processes using a Knowledge Base of approved documents; creates HR Tickets for requests requiring action; routes sensitive topics directly to the People team; and logs every interaction in Max Activity without exposing private employee data.

After deployment: an employee sends a WhatsApp question about their benefits or expense policy. The assistant replies instantly with the correct answer from the approved source. If the question requires HR action, a Ticket is created and assigned. Sensitive questions go directly to HR without being processed by AI.

When to use this

HR is answering the same policy questions repeatedly via WhatsApp, email, or Slack.

Employees don't know where to find policy documents and ask HR instead.

You need a way to handle routine HR questions at scale without adding HR headcount.

You want HR interactions to be logged and trackable without requiring manual notes.

System components

Knowledge Base: approved HR documents — benefits guide, expense policy, PTO policy, workplace guidelines, onboarding documents.

Internal HR AI assistant: answers from the Knowledge Base only, never from general AI knowledge.

CRM Ticket record: created for requests requiring HR action (leave request, benefits enrollment, policy exception).

CRM Person record: employee identity, department, manager, HR owner.

Studio workflow: triggered on new employee WhatsApp message to the HR channel.

Conditional Routing: answerable question → AI reply; action required → Ticket creation; sensitive topic → HR escalation.

Max Activity: question category, action taken, Ticket created (if any), escalation reason (if any).

Step-by-step implementation

Agent prompt

You are the People Assistant, an internal HR assistant. Answer employee questions using only the Knowledge Base provided.

You may answer questions about: PTO policy, expense reimbursement process, benefits enrollment, workplace guidelines, onboarding documents, and office logistics.

You must NOT answer questions about: individual salaries, performance review details, disciplinary actions, immigration, or legal matters. For these, set action = 'escalate'.

For questions requiring HR action (leave request, benefits enrollment, policy exception request): set action = 'create_ticket' with a brief ticket description.

Output JSON: { action: 'reply' | 'create_ticket' | 'escalate', reply_text: string | null, ticket_description: string | null, ticket_category: string | null, escalation_reason: string | null }.

Workflow logic

Question answerable from Knowledge Base: AI reply. Write Max Activity with source document name.

Question requires action (leave, enrollment, exception): create HR Ticket. Assign to HR owner. Send confirmation to employee. Write Max Activity.

Sensitive topic detected (pre-filter OR agent escalation): route to HR owner. Send empathetic acknowledgment to employee. Write Max Activity with category only — no message content.

Question not in Knowledge Base and not sensitive: create HR Ticket as 'Question — no documented answer'. Assign to HR owner to respond and add documentation.

Employee contacting multiple times about the same topic: link to existing open Ticket. Do not create duplicates.

Final operating state

Common HR questions are answered instantly from approved sources — no HR time required for routine policy lookups.

Requests requiring HR action become Tickets with an assigned owner and a tracked status.

Sensitive topics never pass through AI — they go directly to the HR owner with an acknowledgment to the employee.

Max Activity shows every HR interaction: question category, action taken, Ticket created, and any escalation — without exposing private content.

Troubleshooting

AI answering questions it shouldn't: verify the sensitive topic list in the pre-filter AND in the agent prompt. The pre-filter catches obvious cases; the agent prompt handles edge cases.

Knowledge Base answers outdated: update documents in Studio → Knowledge Bases. The AI agent uses the current version of documents — no retraining needed.

Tickets created without proper category: ensure the agent prompt specifies exact category values to use in the ticket_category field.

Employee receives no acknowledgment for escalated questions: confirm the Send Message node on the escalation branch fires before routing to HR owner — the employee should hear back within seconds even for sensitive requests.

Operational playbook

Use Employee support assistant as part of the Frontline Solutions HR operating loop: inspect the current product state, confirm the source context, and decide what should happen next.

The goal is not to memorize screens. The goal is to understand how the product surface supports repeatable work, AI assistance, and accountable handoff.

Best practices

Start with the operational job before changing configuration. Name the owner, define the trigger or source context, and decide how the result should be reviewed.

Prefer narrow, inspectable setups over broad automation. Teammates should be able to explain why the system took an action from the visible product state.

Platform layers involved

Studio defines the workflow and AI agent behavior. Channels capture the customer interaction. CRM provides customer memory. Max Activity shows what the system did and what needs follow-up.

Use the solution page as the business-facing map, then open the related product tutorials when you need configuration detail.

Outcome metrics

Track a small set of operational signals: response time, handoff rate, completion rate, escalation quality, CRM field completeness, reply rate, and repeated failure patterns.

The metric should reflect the business outcome, not only whether the automation ran.

FAQs

Related workflows