Software migration and cutover · Checklist · intermediate

PMS migration acceptance criteria

Define pass, conditional pass, and fail evidence for records, balances, permissions, workflows, documents, and reports before cutover.
By Aptoria editorial team · 3 min read · Updated 2026-09-18 · Last reviewed 2026-09-18
Technical content review: Codex technical editorial review. Reviewed intent separation, internal consistency, original decision artifacts, fictional examples, source limits, and operational risk boundaries. No legal, tax, accounting, banking, safety, or human professional approval is claimed.
This is a technical review, not independent human or professional review.
The short answer
PMS migration acceptance criteria state, before cutover, which populations and workflows must be tested, the evidence and tolerance for each, who decides, and what happens on failure. Cover identity and relationships, balances and open items, permissions, integrations, documents, reports, and critical workflows. Record conditional passes as owned exceptions, never as silent success.

Operational checklist

Mark your progress, then save a working copy. Selections reset when you leave this page. A checked box is not an approval or evidence of completion.
0 of 7 marked

Key takeaways

  • Define criteria before reviewing results.
  • Test relationships and workflows, not just row counts.
  • Conditional acceptance needs limits, owner, deadline, and release impact.

Write criteria as reproducible questions

“Data looks good” cannot be independently checked. Name the source population, destination evidence, sampling or full-population method, tolerance, tester, and approver. Freeze source and destination versions used for the decision.
Separate migration completeness from software suitability and staff readiness. A pass in one domain must not erase failure in another.

Use a domain acceptance gate

Every domain receives pass, conditional pass, fail, or not tested. Not tested is not a pass.
PMS migration acceptance gate
DomainAcceptance evidenceFailure example
Identity and relationshipsCounts plus parent/child and cross-reference testsLease exists but points to wrong unit
Balances and open itemsControl totals, item lineage, cutoff proofTotals net to zero across properties
DocumentsInventory, association, open/read sample, exceptionsFiles counted but unreadable
PermissionsRole tests using representative usersFormer user retains access
IntegrationsSource event through external outcome and receiptConnection exists but callback fails
Critical workflowsScenario tests with acceptance evidencePayment posts without intended ledger result
ReportsDefined reports reconcile to accepted recordsOwner statement uses wrong cutoff

Bound conditional acceptance

A conditional pass states the exact exception population, business impact, compensating process, owner, deadline, and re-test criterion. If those fields cannot be stated, the condition is not controlled.
Keep the decision authority explicit. A migration vendor can supply evidence; the accountable operating owner accepts the result under the organization’s governance.

Prove what changed after the tested snapshot

Acceptance evidence expires when source records continue changing without a controlled delta process. Link the baseline extraction to the final delta register and show how creates, updates, deletes, provider events, and documents reached one supported target disposition.
Include effective-access tests and report-definition comparability. Matching row counts cannot show that permissions deny the right actions or that a familiar report label retained its historical meaning.

Define post-cutover stabilization and source retirement separately

Acceptance should name how post-cutover defects are captured, contained, assigned by domain, corrected through versioned releases, and reconciled across the affected population. A quiet support queue is not evidence that target records are correct.
Retain the source in its approved read-only or rollback role until archive retrieval, audit history, residual integrations, scheduled jobs, access, and first-cycle evidence meet explicit retirement gates. Go-live and decommissioning are different decisions.

Reserve a separate gate for the first live close

Pre-cutover testing cannot prove that migrated openings plus a full period of live receipts, charges, invoices, corrections, bank activity, reports, and integrations will close together. Define the first post-cutover close as a separate acceptance event with its own population, cutoff, difference bridge, approvers, and release impact.
Classify differences as opening, live transaction, configuration or mapping, timing, presentation, or unknown. Keep recurring-charge expectations and scheduled-job execution evidence linked but distinct: a job can run once and still produce the wrong business charge.

Edge cases

  • Source data changes during testing: identify the delta or refreeze instead of comparing moving populations.
  • A test environment differs from production: record the limitation and production verification step.
  • A legacy field has no destination: approve its disposition and retrieval path rather than silently dropping it.

Sources and references

Follow each source to check the underlying claim. Access checks and professional review are different steps.
1. Primary source · National Institute of Standards and Technology
Contingency Planning Guide for Federal Information Systems
Contingency planning connects recovery priorities, backup, alternate processing, testing, and recovery procedures. Applied here as a limited migration-planning analogy.
Source checked 2026-09-18
Automated source-access check: 2026-09-18.

Revision history

2026-09-18
Initial Phase 2 operational article with an original decision artifact, explicit failure states, primary-source scope notes, and AI-assisted technical review.
Report a correction to this resource