Lesson 5 of 5
Testing material ledger results
9 min
Listen to this lesson
Material ledger testing has one non-negotiable end point: inventory values in the ledger reconcile with the inventory balances in the general ledger, for every plant and every currency you keep. Everything else in this lesson builds the evidence for that statement, plus evidence that differences went where the design says they should go. Because the ledger and the G/L share the Universal Journal, a mismatch is rarer than in ECC, but it is not impossible: wrong account determination, manual G/L postings to inventory accounts and conversion residues all produce gaps.
What to test
| Test | How | Expected result |
|---|---|---|
| Price difference at invoice | Post a PO at $40, GR, invoice at $42 for CMC-RTR-200 in 1710 | Price difference account debited $2,000; ledger shows $2,000 on receipts for the rotor |
| Production variance | Run a CMC-ALT-100 order with extra scrap; settle | Variance posted; ledger shows it against the alternator receipts |
| Currencies | Post a receipt in company code 1910 (MXN) after Wave 2 | USD values at the posting date rate; no revaluation at month-end rates |
| Actual costing run (pilot) | Run all steps for plant 1710 in CMR | All materials settled or explained; PUP for the rotor = $42 |
| Revaluation postings | Review post-closing documents | Ending inventory +$400 rotors, +$400 alternators; COGS +$1,200 |
| Reconciliation | Compare ledger stock values with G/L inventory accounts per plant | Zero difference, or each difference explained |
A material ledger test set for Contoso. The bottom row is the exit criterion.
Design expected results before you run anything. For the rotor case Priya's team wrote the expected periodic unit price, the expected split of $2,000 into $400 and $1,600, and the expected roll-up to alternators on paper first. A test that says "ran the costing run, no errors" proves very little. A test that says "expected PUP $42.00, actual $42.00; expected COGS adjustment $1,200.00, actual $1,200.00" proves the logic.
- Pick samplesHigh-value, high-volume and multilevel materials: CMC-ALT-100, CMC-RTR-200, plus one purchased-only part
- Calculate by handExpected differences, PUP and distribution in a spreadsheet the business reviews
- Run and capturePostings, CKM3 screens, run logs and post-closing documents per sample
- ReconcileLedger values against G/L inventory accounts per plant and currency
- Explain or fixEvery difference has a cause, an owner and a decision
Reconciliation as evidence
Reconciliation is the evidence a controller accepts. Take the stock value per plant and valuation class from the ledger side (a stock value list or inventory value app) and compare it with the balances of the inventory G/L accounts for the same company code and period. Do it in each ledger currency. Document the result as a table: plant, account, ledger value, G/L value, difference, explanation. If you test right after a conversion mock, include a comparison with the ECC stock values from the extract taken before the conversion, so you can show that the conversion did not change inventory values.
| Evidence | Weak | Strong |
|---|---|---|
| Run result | "Costing run completed" | Run log with every material settled or listed with an owner |
| Prices | "PUP looks reasonable" | Expected PUP vs actual for each sampled material |
| Postings | Screenshot of one document | Post-closing documents tied to expected amounts |
| Balances | "Inventory looks right" | Ledger vs G/L per plant, account and currency, differences explained |
Negative and regression tests
Two more kinds of test round out the set. Negative tests show the ledger and the run fail safely: post a goods issue that would drive stock negative where negative stock is not allowed and confirm it is blocked; leave a component without a valid price in the pilot and confirm the run lists it and holds the materials above it rather than settling them at a wrong value. Regression tests protect the result after go-live: when Contoso applies its first feature pack or changes account determination, rerun the rotor and alternator cases with the same data and compare the numbers with the stored expected results. Keep the expected-results spreadsheet under version control next to the test cases in SAP Cloud ALM so the comparison is a five-minute job rather than a re-derivation.
What would you do?
During mock conversion 2, the reconciliation shows plant 1720 Chicago inventory in the ledger is $3,150 higher than the G/L inventory account. Luis wants to know if it blocks the dress rehearsal. What should Priya do first?
- The exit criterion: ledger inventory values reconcile with G/L inventory balances per plant and currency.
- Write expected differences, PUPs and distributions before running anything.
- Sample high-value, high-volume and multilevel materials; capture CKM3, logs and post-closing documents.
- After a conversion mock, compare total stock values with the pre-conversion ECC extract.
Check your understanding · not graded
Lesson check
1. What is the strongest single piece of evidence that material ledger values are correct?
2. When should expected periodic unit prices be calculated?
3. Why compare stock values with the ECC extract after a conversion mock?