CUSTOMER SUPPORT

Cluster monthly deflection failures into doc gaps

Monthly, it groups all failed-deflection questions by topic, finds clusters where no single article answers a recurring need.

CategoryCustomer Support
Enginepaperclip
Difficultyadvanced
Triggerschedule
Steps5
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerMonthly schedule
  • ActionRead month's failed-deflection records from PostgresPostgreSQLPostgres
  • LogicCluster questions by intent; isolate true content gaps
  • ActionCreate Asana task per gap clusterAsanaAsana
  • OutputDraft stub article in ReadMe and link to taskReadMeReadMe

What it does

Finds the missing articles, not just the weak ones. Once a month it takes every question that escalated past the help center, clusters them by intent, and detects clusters that no existing article covers well. For each real gap it opens an Asana task and seeds a stub page in ReadMe so the docs team can start from a defined need instead of a blank page.

When to use it

Use this when failures aren't about one bad article but about content that doesn't exist yet. It converts a month of scattered failing questions into a short, deduplicated list of net-new articles worth writing.

How it works

  1. 1A monthly schedule starts the run.
  2. 2It reads the month's failed-deflection records from Postgres, including each question and the article shown.
  3. 3An agent clusters the questions by intent and separates true content gaps from cases an existing article should have handled.
  4. 4For each gap cluster it creates an Asana task in the Docs board describing the unmet need and example questions.
  5. 5It drafts a matching stub page in ReadMe and links it on the Asana task for the writer to flesh out.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect PostgresAny Postgres URL — query, write, migrate.
  2. 2
    Connect AsanaTasks, projects, milestones — everywhere.
  3. 3
    Connect ReadMeAPI docs, changelog, auth.
  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.