CUSTOMER SUPPORT

Reverse Deflection: Flag Tickets No KB Answers

When a Zendesk ticket is solved, checks whether any Confluence article could have deflected it; if not, files a Linear doc-gap task so the missing self-serve content gets written.

CategoryCustomer Support
Enginesim
Difficultyadvanced
Triggerevent
Steps6
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerZendesk ticket solvedZendeskZendesk
  • ActionSearch Confluence for related articlesConfluenceConfluence
  • ActionJudge if any doc enables self-serveOpenAI
  • LogicContinue only when no adequate doc exists
  • ActionDraft proposed article title and outlineOpenAI
  • OutputFile Linear doc-gap task and notify SlackLinearLinear

What it does

This is the inverse of a deflection scan: instead of finding tickets a new doc covers, it catches resolved tickets that no existing doc could have deflected. Those are your documentation gaps. For each gap it opens a Linear task describing the missing article, turning real support pain into a docs backlog.

When to use it

Use it to continuously grow your knowledge base from actual ticket demand. Ideal for teams that want their self-serve content prioritized by what customers really ask, not guesswork.

How it works

  1. 1A Zendesk ticket-solved event triggers the run.
  2. 2The flow searches Confluence for articles related to the ticket subject and body.
  3. 3An LLM judges whether any returned article would have let the customer self-serve.
  4. 4A logic gate continues only when no adequate doc exists.
  5. 5For confirmed gaps it drafts a proposed article title, scope, and outline.
  6. 6The draft is filed as a Linear doc-gap task and announced in Slack.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect ZendeskTickets, queues, knowledge base.
  2. 2
    Connect ConfluenceSpaces, pages, blueprints.
  3. 3
    Connect OpenAIModels, embeddings, files.
  4. 4
    Connect LinearIssues, projects, cycles, triage.
  5. 5
    Connect SlackChannels, DMs, threads, mentions.
  6. 6
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  7. 7
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  8. 8
    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.