Test converted balances before going live
Run test imports, sample transactions, parallel comparisons and exception resolution before staff rely on the new ledger.
Exam scope and assessment notes
Cambridge International AS & A Level Accounting 9706, 2026–2028 syllabus, Version 2. These lessons cover the process of transferring business accounts to a computerised accounting system and ways to maintain data integrity during conversion. No specific accounting application is required. Practice is self-checked, not an official Cambridge grade. Full timed Papers 1–4 and human subject-teacher review remain outstanding.
Before you begin
- Use the Topic 1.2 accounting-system skills: understand double entry, trial balances, control accounts, source documents and why complete records matter.
Work on paper. Try each question before revealing help, and explain why your method works.
Quick readiness check
What should happen if an unresolved critical balance fails a pre-agreed go-live test?
What you will learn
- Design a test migration and risk-based sample
- Compare old and new system outputs
- Use a stop/go decision based on verified exceptions and approval
Test high-risk data and workflows first
A test conversion uses a copy of data or a controlled test environment so the team can inspect how records map before live use. Test opening balances, customer and supplier details, inventory, tax codes, recurring journals and transaction cutover. Include edge cases such as credit balances, inactive accounts and large amounts.
A sample should cover different account types, values, periods and transaction outcomes. Trace selected records from source to destination and back, checking the account, amount, date, reference and reporting treatment. A sample complements, but does not replace, full control totals and reconciliations.
Compare results before the stop/go decision
A parallel run records transactions in both systems for a defined period or performs agreed comparisons on the same source data. Reconcile financial reports, control accounts, transaction counts and exception lists. Investigate differences before choosing which system becomes authoritative.
Define acceptance criteria in advance: what must reconcile, which exceptions can remain, who approves them and what blocks go-live. If a critical balance or data-integrity issue remains unresolved, delay or limit go-live and follow the fallback plan rather than hiding the difference.
Pause and explain
Why is a parallel comparison useful?
Worked example
The test migration reconciles the trial balance but five customer accounts show receipts allocated to the wrong invoices. Should the project go live? Explain the decision and next tests.
Show the worked solution
- The aggregate trial balance can reconcile while customer-level allocation is wrong.
- Wrong allocations affect ageing, collection activity and customer statements, so the issue is material to the process even if the control total matches.
- Correct the mapping or allocation, rerun affected customer reports and samples, and require reviewer approval against pre-agreed criteria.
Answer Do not approve the affected receivables process yet. Correct the allocations, retest customer balances and statements, then proceed only after the exception meets acceptance criteria.
Try a topic test
Choose Easy, Normal or Difficult. Save your answers and open worked review after finishing.
Test this topic →From guided practice to a new situation
Write your answer and reasoning first. Use a hint only if you are stuck. The model solution is for self-checking; your written work is not automatically marked.
A sample of 40 migrated supplier accounts finds three with correct totals but wrong payment terms. Describe the issue, consequence and retest.
Give me a hint
Fields can be wrong even when account balances agree.
Compare my reasoning
- The conversion preserved balances but failed a supplier-master-data field check.
- Wrong terms can cause missed discounts, late payments or disputes.
- Correct the field mapping or source values, test other supplier records and reissue a controlled sample report for review.
The data is not fully accurate. Correct terms using approved records, expand the test and obtain review before the supplier process goes live.
Look for these in your work
- The conversion preserved balances but failed a supplier-master-data field check.
- Wrong terms can cause missed discounts, late payments or disputes.
- Correct the field mapping or source values, test other supplier records and reissue a controlled sample report for review.
Acceptance criteria require all control accounts to reconcile and no unapproved critical exceptions. The test finds one $4,200 unexplained inventory difference and two formatting issues in non-critical descriptions. State a reasoned go/no-go decision.
Give me a hint
Distinguish a critical balance difference from a cosmetic issue, but preserve the exception records.
Compare my reasoning
- The $4,200 inventory difference is unexplained and affects a key balance, so it fails a required control.
- The project should not fully go live until the cause is resolved or formally assessed under approved criteria.
- The description-format issues may be handled under an approved exception process after confirming they do not affect identification or reporting.
No-go for the full inventory-dependent launch until the $4,200 difference is resolved. The formatting issues can be reviewed separately under the approved non-critical exception process.
Look for these in your work
- The $4,200 inventory difference is unexplained and affects a key balance, so it fails a required control.
- The project should not fully go live until the cause is resolved or formally assessed under approved criteria.
- The description-format issues may be handled under an approved exception process after confirming they do not affect identification or reporting.
Management suggests dropping the parallel run because old and new systems produce the same trial balance. Give two additional comparisons needed before relying on the new system.
Give me a hint
Check detailed balances and the reports users depend on, not only the trial balance.
Compare my reasoning
- Compare customer and supplier subledgers, allocations and ageing reports.
- Test transaction dates, tax or inventory fields and trace samples back to source documents.
- Review rejected records, audit logs and user access, then have an independent person approve the result.
Compare subledgers and operational reports, trace data fields and samples, inspect exceptions and access controls, and obtain independent sign-off.
Look for these in your work
- Compare customer and supplier subledgers, allocations and ageing reports.
- Test transaction dates, tax or inventory fields and trace samples back to source documents.
- Review rejected records, audit logs and user access, then have an independent person approve the result.
Common mistakes
- Testing only the total trial balance
- Choosing a sample that contains only small, simple transactions
- Changing acceptance criteria after an error appears
What can you explain now?
Why test individual records as well as totals?
Compare with the explanation
A total can agree while dates, references or account allocations are wrong
Record-level tracing checks the fields that determine how an item is reported and used.
After trying it yourself, choose your next review. This is your self-assessment.
Your review choice appears on Today. Sign in to sync it across devices.
Make it stick
What evidence should a project team review before deciding go or no-go?
If you needed the solution, close it and solve the problem again from a blank page. A correct answer is stronger when you can explain the reason for each step.