Before You Freeze the LMS Migration Scope, Fix the Launch Criteria

Signals to watch

• scope drift before term start
• disconnected readiness checklists
• no shared stop/go rule

Composite industry case. This page describes a reusable operating problem and decision method. It does not represent a named customer, real conversation, contract, revenue result or testimonial.

The Unfrozen Scope Problem

You are five weeks from term start. The LMS migration project board shows every task at 80 percent or higher. Yet every planning meeting surfaces a new requirement: course shells need a different naming convention, grade pass-through rules changed in the source system, a third-party LTI tool is not responding in the sandbox, the alt-text audit on migrated content is incomplete.

None of these surprises are malicious. Each one is legitimate. And each one rewrites the launch criteria after the team thought they were settled.

The structural issue is not that scope changes — scope always changes. The issue is that no shared definition of “ready to launch” exists across the six dimensions that a semester migration touches: course content, identity and access, grade schema, tool integrations, accessibility compliance, and instructor training. Without that shared definition, every stakeholder defines readiness differently, and the scope remains unfrozen until the moment someone says “we launch anyway” — which is usually too late to adjust gracefully.

Why Teams Misread This as a Planning Problem

Most learning program leads respond to scope drift by adding more planning. They build a detailed Gantt chart, assign owners to every sub-task, and hold weekly status reviews. These steps feel productive, and they do surface information — but they do not fix the unfrozen scope.

The reason is subtle: planning tools track activity completion, not readiness across dimensions. A task can be 100 percent complete — the course migration script ran, the grade schema was exported, the SSO configuration was applied — while the launch criterion it was meant to serve is still unmet. The script ran but produced incorrect section mappings. The grade export happened before a downstream formula change. The SSO config works for faculty but not for students on mobile.

Planning says “done.” The launch criterion says “proven.” These are not the same thing, and confusing them is what keeps the scope unfrozen.

The Evidence Review Framework

Build a launch criteria sheet with five columns: dimension, criterion, evidence, owner, and decision window. Do not start with dates. Start with what must be true.

For each of the six dimensions, write one to three concrete criteria.

• Course content: Every migrated course shell has passed a spot-check comparing source and target enrollments, module structure, and at least one graded item.
• Identity: A test account for each role (student, instructor, admin, auditor) can log into the target LMS and access the correct dashboard.
• Grades: The target gradebook computes final grades within one decimal point of the source system for a sample set of twenty enrollments spanning three grading schemes.
• Tool integrations: Every LTI 1.3 link in the source system launches successfully in the target sandbox and returns a valid grade pass-back.
• Accessibility: A random sample of ten migrated courses scores above the institutional accessibility threshold on automated scan, and two courses pass a manual screen-reader check.
• Training: At least one instructor per department has completed the target LMS orientation and confirmed that their common workflows (attendance, grade entry, content update) are functional.

Each criterion earns a status: green (evidence matches), yellow (evidence partial with known gap), or red (no evidence or evidence shows failure). Yellow and red entries require a written mitigation plan before the decision window date.

Team Next Step

Schedule a two-hour working session with the LMS administrator, the instructional design lead, the IT identity manager, the accessibility coordinator, and one faculty representative. Bring a blank version of the criteria sheet — do not let anyone pre-fill rows.

Walk through each dimension together. The first round will produce rough criteria. The second round tightens the evidence column. The third round assigns owners and sets decision windows based on the calendar, not the other way around.

By the end of the session, you will have six to twelve criteria that everyone agreed to. That agreement is what freezes the scope — not the plan, not the tasks, not the status color. The agreement that these are the things that must be proven before launch.

What Automation Cannot Replace

A tool can flag that a criterion is red. A dashboard can show that the decision window has passed. A workflow can nudge the owner to upload evidence. But no tool can decide whether the gap is acceptable or whether to delay.

That human judgment — reviewing the evidence, weighing the risk, and making an explicit call visible to every stakeholder — is the one activity that continuous discovery and evidence organization cannot replace. The tool’s job is to surface the signal clearly enough that the human review produces a named action with an owner, evidence, and a decision window. The learning program lead’s job is to hold that review.

Start with the criteria sheet. The frozen scope follows.

Frequently asked questions

How many launch criteria are enough for a semester migration?

There is no universal number. The goal is one criterion per dimension that has a history of last-minute surprises — typically five to seven criteria covering content, identity, grading, integrations, accessibility, and training. Each must have a clear evidence source and an owner.

Who should own each launch criterion?

The person closest to the work. The LMS administrator owns grade schema verification; the instructional designer owns course content spot-checks; the IT liaison owns SSO login testing. Ownership means the person can produce evidence and make a go/no-go recommendation — not that they decide alone.

What if a criterion is not met two weeks before term?

The decision window forces a choice: accept the gap with a documented mitigation plan, or delay the migration for that track. The key is that the choice is explicit and visible to all stakeholders — not buried in a status meeting.