Capstone operating loop and readiness checklist

On this page

The capstone turns the course into a realistic operating loop. The steward is given a data-quality symptom, then asked to move through the platform in the right order, choose the correct remediation tool, verify the outcome, and decide whether anything should be escalated.

What you should get from this module

  • combine search, validation, clean, deduplication, governance, and AI patterns into one operating flow
  • practice tool selection instead of treating every issue the same way
  • self-assess readiness for real stewardship work

view_tagged_records.png

Guided walkthrough

Use a realistic scenario such as one of the following:

  • a tagged population of records with invalid phone values is rising over time
  • a saved search shows many records with missing key attributes
  • a domain contains suspicious duplicates after a new source was processed
  • a recurring typo pattern keeps returning after being fixed once

Ask the learner to work through the issue in this order:

  1. Define scope with search, saved search, glossary, or tags.
  2. Inspect evidence on the results page and sample records.
  3. Check lifecycle clues in History, validations, or review mapping.
  4. Choose remediation:
    • clean project for repeated value issues
    • deduplication for suspected duplicate populations
    • AI job for scalable tagged remediation
    • escalation for mapping or identifier defects
  5. Verify the result by returning to the saved search, tag view, or affected records.
  6. Record the lesson as a repeatable stewardship practice or an architect handoff.

The main idea is deliberate tool choice. Not every problem needs a clean project. Not every recurring issue should stay manual. Not every duplicate symptom should be merged immediately.

Role lens

By the end of this capstone, the learner should sound less like a ticket processor and more like an informed operator who understands evidence, scope, remediation, and platform consequences.

Practice assignment

Run one full scenario end to end and write a short after-action review:

  • What was the initial symptom?
  • Which UI surfaces did you use first?
  • What evidence changed your understanding of the problem?
  • Which remediation path did you choose and why?
  • What would you monitor next week to make sure the issue stays fixed?

Exit criteria

  • The learner can select the correct stewardship path for at least one realistic issue.
  • The learner verifies outcomes instead of assuming fixes worked.
  • The learner can distinguish operational remediation from structural escalation.

Suggested source material