Lesson 3 of 5
Cutover validation
10 min
During the cutover weekend the data team loads master data and open items into production. Before the business starts transacting, the process owner must accept that the data is complete and correct. That acceptance is a sign-off in its own right, and it is the one with the least time to think. Preparation is everything.
What is migrated and how it is checked
In SAP S/4HANA new implementations, data is commonly loaded with the SAP S/4HANA Migration Cockpit. It works with predefined migration objects (business partners, products, open items, balances, stock and so on), validates data against the same checks as the standard applications, lets the team simulate a load before posting it, and points to the post-processing apps used to check each object. Technical success in the cockpit is necessary but not sufficient: the business must reconcile the result to the legacy system.
- Extract and freezeLegacy data extracted after the business freeze; legacy totals recorded
- Validate and simulateMigration Cockpit checks data and simulates the load
- LoadMaster data first, then open items, balances and stock
- Technical reconciliationRecord counts and load errors reviewed by the data team
- Business reconciliationTotals and samples compared with legacy by process owners
- Data acceptanceProcess owner signs, or lists conditions
The reconciliations a process owner reviews
| Object | Reconciliation | Contoso example |
|---|---|---|
| Customers and suppliers | Record counts by account group or role; sample of key fields (address, payment terms, bank details, tax data) | 1,240 customers in ECC = 1,240 business partners with customer role; 25 sampled |
| Products | Counts by plant and type; sample of valuation, MRP and sales data | CMC-ALT-100 extended to plants 1710 and 1720 with standard prices $150 and $156 |
| Open receivables and payables | Total by reconciliation account and by customer/supplier; aging buckets | AR total and per-customer balances match ECC; Great Lakes Fleet Services checked line by line |
| G/L balances | Trial balance by account, company code 1710, migration period | Every account matches ECC; differences explained |
| Stock | Quantity and value by plant, storage location and material; total inventory value to G/L | Storage locations 171A, 171B, 172A quantities and values match; inventory G/L ties to stock value |
| Open orders | Counts and values of open sales and purchase orders | Open sales orders for 1710 / 10 / 00 match; quantities still to deliver correct |
Business reconciliations for data acceptance.
| Reconciliation result | Meaning | Process owner action |
|---|---|---|
| Matches | Counts and totals agree; samples correct | Accept |
| Difference, explained | For example, items deliberately not migrated (closed, written off) and documented | Accept with the explanation attached |
| Difference, unexplained but small and isolated | Not yet understood; low impact | Accept with condition: owner and deadline to explain or correct |
| Difference, unexplained and material | Not understood; could misstate balances or block operations | Withhold acceptance until resolved or escalated for formal decision |
Process owners should also check that data is usable, not only that it matches. A customer can reconcile perfectly and still be unable to order because a sales area extension or a credit segment is missing. Include a handful of smoke tests on migrated data in the acceptance: create a sales order for Great Lakes Fleet Services from migrated master data, post a goods receipt for a migrated purchase order from Lakeside Components, and display a migrated open item and its due date. These take minutes and catch gaps that totals cannot.
Preparing before the weekend
Data acceptance cannot be improvised at 2 a.m. Prepare it during the mock cutovers. Agree the reconciliation reports, the sample sizes, the tolerance for differences (often zero for financial balances), who checks what, and how results are recorded. Run the full reconciliation in each mock cutover, and time it: if reconciliation took nine hours in the last rehearsal, plan nine hours in the cutover plan.
What would you do?
At 3 a.m. on cutover Sunday, migrated open supplier items for Lakeside Components are $18,400 higher than the ECC total. All other suppliers match. Luis needs the data acceptance by 6 a.m. What should Dan do?
- Data acceptance is a sign-off: the process owner confirms migrated data is complete and correct.
- Migration Cockpit success is technical; business reconciliation to legacy totals is still required.
- Reconcile counts, totals and samples for master data, open items, balances, stock and open orders.
- Prepare and time reconciliations in mock cutovers; financial balances must tie exactly or be explained.
Check your understanding · not graded
Lesson check
1. Which SAP tool is commonly used to load data in new SAP S/4HANA implementations?
2. Why is a successful Migration Cockpit load not enough for data acceptance?
3. What should total inventory value reconcile to after migration?
4. Migrated payables for one supplier are $18,400 higher than legacy, unexplained. What should the process owner do?