Now accepting guest editorial from SAP, Oracle, Workday and ServiceNow practitioners.Pitch a guide
The Sign-Off Table

Lesson 1 of 5

Deciding what data to bring

13 min

Every migration program is eventually asked the same question: how much of the old data do we bring? The honest answer is "as little as the business needs to run, report and comply, and no more." Getting there is not a technical exercise. It is a negotiation between operations, finance, legal, tax, audit and IT, and at expert level your job is to run that negotiation so that it ends in written, owned decisions rather than a vague "migrate everything to be safe".

Contoso Motor Co. moved its own SAP ECC 6.0 system to SAP S/4HANA Cloud Private Edition by system conversion, so its own history came across in place; the volume question there was what to archive before converting. The scope negotiation this course is about arises in Wave 2. Contoso acquired Alternadores del Norte S.A. de C.V. in Monterrey, Mexico, and loads its data from a non-SAP legacy ERP into the converted system as company code 1910 (MXN), plant 1910 Monterrey. That load migrates master data and open items by default, and closed history stays behind. The default still leaves dozens of questions: which of the 3,180 legacy customers count as active (1,946 ended up in scope), whether one-time customers come across, how much purchasing history buyers need, what happens to 15 years of closed finance documents. Dan (finance), Maria (order to cash) and Grace (PMO) own those answers with advice from legal, tax and internal audit in both countries.

Why "everything" is the wrong default

Cost of more dataWhat it looks likeAlternadores del Norte example
RuntimeLonger loads, longer cutover window, more risk of overrunEvery extra year of history adds hours of load and validation
QualityOld data carries old problems: duplicates, obsolete codes, missing fieldsCustomers from closed regions with invalid tax jurisdiction codes
TestingEvery migrated record must be reconciled and usableHistory in the new system needs its own validation and test cases
PrivacyMore personal data in more systems for longerContact persons of customers inactive for a decade
Sizing and costMemory and storage in SAP HANA are not freeLarger database sizing for the private cloud contract

The costs that a "migrate everything" default quietly accepts.

A decision framework

For each data set, ask four questions in order: does the business need it to run processes after go-live, to report in the new system, to retain for legal, tax or audit reasons, or can it be left behind? The answer decides where the data lives after go-live.

  1. Is the data needed to run processes after go-live?Yes → migrate (open items, active master data) · No → next question
  2. Do users need it inside the new system for reporting, returns or warranty?Yes → migrate a limited history (only possible on some paths) or provide reporting from another store · No → next question
  3. Is there a legal, tax or audit retention requirement?Yes → keep it accessible: read-only legacy, archive or retention warehouse · No → next question
  4. Is there any other documented need?Yes → owner states it and accepts the cost · No → leave behind and plan destruction at end of retention
Run, report, retain or leave behind: the core scope decision for each data set.
The Sign-Off Table · thesignofftable.com/academy/sap-dmg-401

Notice that "retain" is not the same as "migrate". Most retention obligations can be met by keeping the data accessible somewhere other than the new production system. That single distinction removes most of the pressure to migrate history.

Data volume is a decision input, not an afterthought

Volume shapes the path as well as the plan. In a system conversion, every gigabyte in the source is converted during downtime and lands in SAP HANA memory, so programs often archive or delete data in the source before converting. In a new implementation, volume drives load runtime and validation effort. In selective data transition, volume is chosen through company code and time-slice selection. SAP Readiness Check reports data volume and usage information about the source system that helps size these decisions; ask the Basis team for the latest results before the scope workshop.

Challenge itHigh cost, low value: long finance history, inactive master data. Retain elsewhere.
Decide deliberatelyHigh cost, high value: recent sales history for warranty. Owner accepts cost, or find another store.
Leave behindLow cost, low value: obsolete reference data. Not worth the test effort.
MigrateLow cost, high value: active master data, open items. Core scope.
Positioning data sets by value and cost. The top-left box is where most scope arguments end up.
The Sign-Off Table · thesignofftable.com/academy/sap-dmg-401

Write decisions down

Each scope decision becomes a scope decision record: the data set, the decision (run, report, retain, leave behind), the selection rule, where the data lives after go-live, the retention period if any, the decision owner and the advisers consulted, and the date. These records are what auditors, successors and future you will read when someone asks in two years why a report cannot show 2019 sales.

Data setDecisionWhere it lives after go-liveOwner
Customers with activity in the last 24 months or open itemsRun: migrateSAP S/4HANAMaria
Customers inactive for over 24 months, no open itemsRetainRead-only legacy ERP, then retention storeMaria with legal
Closed sales orders and billing documentsRetainRead-only legacy ERP, then retention storeMaria
Open receivables and payables, G/L balancesRun: migrateSAP S/4HANADan
Closed finance documentsRetainRead-only legacy ERP, then retention storeDan with tax and internal audit
Obsolete test customers in the legacy ERPLeave behindNot kept beyond legacy shutdown, subject to legal reviewMaria

Extract from the scope decision log for the Alternadores del Norte load. Selection thresholds are fictional.

What would you do?

At the Alternadores del Norte scope workshop, its sales director asks for ten years of its closed sales orders in SAP S/4HANA "for trend analysis". What is the best response?

Key points
  • Migrate as little as the business needs to run, report and comply.
  • Ask run, report, retain or leave behind for each data set; retention rarely requires migration.
  • Volume drives conversion downtime, load runtime, testing and sizing; use SAP Readiness Check data.
  • Record every scope decision with owner, advisers, rule, location after go-live and retention period.

Check your understanding · not graded

Lesson check

  1. 1. Why not migrate all historical data by default?

  2. 2. A legal retention requirement exists for closed invoices. What does that usually require?

  3. 3. What is a scope decision record?

  4. 4. In a system conversion, why do programs often archive before converting?