CHATBOTS

Reorder Assistant: Escalate Ambiguous Reorders to a Human

When a reorder request can't be matched to a single past order with confidence, the bot summarizes the customer's history and hands off to a human agent in Slack…

CategoryChatbots
Enginepaperclip
Difficultyintermediate
Triggerchat
Steps6
Setup~15 min

How it runs

The automated pipeline, trigger to output.

  • TriggerReorder message received in IntercomIntercomIntercom
  • ActionPull matching past orders from PostgresPostgreSQLPostgres
  • LogicScore match confidence and order value
  • LogicBranch ambiguous or high-value to escalation
  • ActionPost agent handoff with candidates to SlackSlack
  • OutputLeave holding reply in Intercom threadIntercomIntercom

What it does

Keeps the bot from guessing on risky reorders. When a customer's request is ambiguous — multiple matching past orders, vague wording, or a high-value cart — the assistant stops, compiles the relevant order history, and routes a clean handoff to a human agent rather than placing the wrong order.

When to use it

For brands that want automation on the easy 80% of reorders but a safe human escape hatch for the ambiguous or high-stakes ones, protecting against wrong-item charges.

How it works

  1. 1A reorder message in Intercom triggers the flow.
  2. 2The flow pulls the customer's matching past orders from Postgres.
  3. 3Logic scores match confidence; a single clear match would continue to a normal reorder reply.
  4. 4Low-confidence or high-value requests branch to escalation.
  5. 5The flow assembles a summary of candidate orders, SKUs, and the customer's words.
  6. 6It posts the handoff to a Slack channel for agents and leaves the Intercom thread with a "a teammate is confirming this" holding reply.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect IntercomConversations, contacts, articles.
  2. 2
    Connect PostgresAny Postgres URL — query, write, migrate.
  3. 3
    Connect SlackChannels, DMs, threads, mentions.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    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.