Student Success by DesignA practitioner’s guide to theory, data, and AI in student success

7.8 · Student Success and Data Governance

Monitoring, escalation, and retirement

Approval is the start. These templates cover what to watch after launch, what to do when something goes wrong, and when to stop.

Short shows the summary, takeaway, and questions.
Template · Incident escalation
  1. Level 1Data error

    TechnologySteward corrects at the source within 5 working days; affected staff informed

  2. Level 2Harm to a student

    TechnologyPause the affected feature, contact the student, correct records, report to the council

  3. Level 3Systematic disparity or breach

    TechnologySuspend the tool, convene the council, notify students and legal counsel, publish a summary

Name an incident owner for each level before launch.

Template · Monitoring plan

Each term
Outcomes by student group; reach and response time for flagged students; student complaints and corrections.
Each quarter
Model accuracy and error rates by group; share of flags acted on.
When upstream data changes
Re-check inputs and accuracy before scores are used again.
Each year
Full review against the original purpose by the governance council, with student members.

Retirement criteria: retire or redesign when

  • The tool no longer serves its documented purpose.
  • Accuracy or outcomes for any student group fall below the threshold set at approval.
  • Staff can no longer respond to what it surfaces.
  • The vendor changes data use terms in ways the council has not approved.
  • A comparison shows no benefit after two full evaluation cycles.
  • On retirement, confirm data deletion and tell students what changed.