CHATBOTS

Escalate Unanswered Benefits Questions to HR with Logging

When the benefits bot can't ground an answer in Confluence, it opens a tracked escalation to HR in Teams and logs the gap in Postgres so the knowledge base can be improved.

CategoryChatbots
Enginesim
Difficultyintermediate
Triggerchat
Steps5
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerEmployee asks the benefits bot in TeamsMicrosoft Teams
  • ActionSearch Confluence and score answer confidence with OpenAIConfluenceConfluence
  • LogicBranch: confident answer vs. knowledge gap
  • ActionLog the unanswered question to PostgresPostgreSQLPostgres
  • OutputEscalate to HR channel and notify employee in TeamsMicrosoft Teams

What it does

Catches the questions your benefits bot cannot confidently answer from Confluence, routes them to the HR team channel in Teams, and records each unanswered question in Postgres. Over time this builds a backlog of documentation gaps you can prioritize.

When to use it

When you already run a grounded benefits bot and want a safety net: no employee should hit a dead end, and HR should see exactly which topics are missing from Confluence so they can write the page.

How it works

  1. 1An employee asks the bot a question in Teams.
  2. 2The flow searches the HR Confluence space and runs an OpenAI confidence check on the retrieved context.
  3. 3If confidence is high, it answers normally and stops.
  4. 4If confidence is low or no page matched, it posts the question into the HR escalation channel in Teams, tagging it for a human.
  5. 5It writes the unanswered question, timestamp, and asker into a Postgres table for trend analysis.
  6. 6The employee gets a message that HR has been notified and will follow up.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect Microsoft TeamsChannels, chats, files.
  2. 2
    Connect ConfluenceSpaces, pages, blueprints.
  3. 3
    Connect OpenAIModels, embeddings, files.
  4. 4
    Connect PostgresAny Postgres URL — query, write, migrate.
  5. 5
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  6. 6
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  7. 7
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.