IT systems · Practical guide

Business Central: GxP data migration and reconciliation

A successful import into Business Central gives little evidence that the meaning of the data survived. Identify the authoritative source for critical values. When ERP, spreadsheets and partner applications disagree, the responsible team must determine which value applies and how the target will use it.

Table of contents

At a glance

What should you test when migrating GxP data to Business Central?

For a GxP migration to Business Central, link source data and transfer rules to their actual use in the target process. Check units, status values and relationships. Successful processing of a configuration package does not establish completeness of the business data. Define in advance what will be transferred, what remains outside the target and how results will be reconciled.

Scope and Applicability

Microsoft describes configuration packages as a means of transferring setup data. The appropriate method for your source data and target version must be determined on a project-specific basis. This article is not a comprehensive import template for all tables.

Sources: [1] [2]

The technical significance must withstand the system transition.

Pay attention to sequences and dependencies. A movement dataset may reference material, storage location, unit, and batch. If these relationships are incompletely transferred, a formally present dataset remains technically unusable. Therefore, verify selected complete business transactions rather than isolated rows.

The reconciliation should clearly identify expected exceptions. Discarded data sets, intentionally retained historical records, and subsequent corrections require transparent handling. If sums do not align, a general rounding justification must not obscure differences in units or duplicate imports.

Sources: [1] [2]

Technical import and professional assumption differ

A successful import confirms that a tool has processed certain data. It does not verify whether all required objects, relationships, and status values are correctly present in the target system. Therefore, begin with the intended process: What data are required after the change for material usage, traceability, or quality decisions?

Business Central configuration packages enable selection of tables and fields. Microsoft also outlines limitations regarding automatic population of linked tables and settings for field validation. Consequently, for internal review: the package content and settings must be deliberately evaluated. The "Validate" tool function does not constitute a general GxP validation of the entire migration process.

Capture alongside master data open processes, batch or series references, units, and extension fields. Whether historical data is actively transferred and which data remains accessible elsewhere is a professional judgment. A one-size-fits-all transfer of all previous tables may be equally inappropriate, as may exporting only the visible primary fields.

Sources: [1] [2]

Prepare a mapping and reconciliation matrix

Link the source object, target object, transformation rule, and inspection method. Assign keys to ensure data records can be uniquely compared. When new numbers are assigned, a traceable mapping between old and new identifiers is required. A total count alone is insufficient for this purpose.

Distinguish unchanged receipts from deliberately modified values. A unit may be recalculated, a status may be rerepresented, or a field may be consolidated. For each critical transformation, appropriate normal and boundary cases should be defined. Define expected results prior to execution and justify the scope of automated and manual reviews.

Own work aid: Prepare a mapping and reconciliation matrix
Example objectCritical interdependenceIn-house verification
Article Identity, Unity and Relevant Additional Fields Key and Field Comparison
Batch stock Article, quantity, location, and available status Inventory reconciliation against selected business processes
Open task Reference to material and processing status Continuation of a specified procedure
Expansion field Significance and technical implementation Verify mapping and behavior after import
Historical Record Link to the previous action Reconstruction in the target or in the path of the released archive

Sources: [1] [2]

Example of a correct total assigned to the wrong batch

In the legacy system, two batches of the same product exist, each containing 50 units. One batch is released; the other is blocked. After migration, Business Central shows a total of 100 units. This quantity reconciliation holds true even if the blocked status is incorrectly assigned to the wrong batch. Therefore, the technical review must link identity and status.

Compare the two datasets initially based on their source and target keys. Then, evaluate an authorized and a non-compliant material usage according to the specifically selected process rule. System behavior depends on configuration and installed extensions; the document does not claim to enforce universal standard blocking. The test requirement is derived from your actual workflow.

Document a difference not only with manual correction. Investigate whether mapping, key logic, or sequence may have impacted additional data sets. After root cause correction, appropriate reconciliations should be repeated. A single corrected example dataset is insufficient if the transformation rule was systematically incorrect.

Example of a correct total quantity assigned to the wrong batch
Test stepExpectationWhat the evidence can establish
Quantity reconciliation 50 plus 50 equals 100 No confirmation of correct batch assignment
Key comparison Each batch remains uniquely identifiable No confirmation of professional qualification
Status comparison Released and blocked are correctly assigned No confirmation of the actual downstream process
Business process Defined usage responds to the correct status Applies to the workflow and version tested
Check of the cause Mapping errors and affected scope have been clarified Evaluate additional transformations separately

Sources: [1] [2]

Pilot run, final acceptance and delta integration

A test run checks the tool, data rules and planned process at a designated stand. New processes often arise between test run and productive change. Determine which source is leading for the final stand, when changes end and how data created afterwards are treated. Without this reference, a successful general sample can cover up a later data loss.

Document the final package version, selected fields, relevant settings, and import reports. Verify that errors, warnings, and excluded data sets have been fully evaluated. A green overall status is considered weak if individual data sets do not meet the expected scope or are only supplemented by uncontrolled manual entries.

Plan a break and restart path. A second import must not create unchecked duplicate operations or overwriting changes. Test suitable interruption and repeat cases in a controlled environment. Which technical method is used depends on data object, tool and concrete business central version.

Sources: [1] [2]

Acceptance and operation following change

The acceptance process includes technical operation, technical and specialist reviews, evaluated exceptions, and release of the intended operation. Identify open activities and, if applicable, restricted usage scopes. A global go-live decision without visible residual items hampers collaboration between QA, functional departments, and IT.

Obtain source data, mapping records, reconciliation reports, and decisions in an appropriate format. The subsequent question "Why does this historical order appear different?" must be resolvable. Data retention and data access obligations do not cease solely due to a successful migration. For information not actively transferred, appropriate access must be provided.

Additionally, assign responsibility for identifying first-time operational deviations. A reported incorrect status following a system change may arise from migration, a new configuration, or a subsequent modification. Clear statuses and logs assist in distinguishing between these root causes. Therefore, the initial production phase should include a clear and traceable root cause investigation process.

Sources: [1] [2]

Draft: Integrate configuration package and subject matter alignment

For the final run, compile the exact package status, tables, fields, and settings through technical alignment. Use the following table as a control record and complete a full business case including critical relationships. A report from a general batch must not be renamed alone: the final statement must reflect the actually received inventory.

Sample filling example – internal work aid
LauffeldExample entryAcceptance question
Source Extraction procedure performed with two batches and known status values. Is the final source version unambiguously defined?
Package Tables, enclosed fields, and relevant settings are identified. Are not critical fields accidentally excluded?
Transformation Source status is depicted according to documented procedure. Is the required significance maintained?
Reconciliation Key, quantities, units, and status are compared. Have differences been fully resolved?
Process Permitted and limited use will be reviewed. Does the actual target workflow behave as intended?
Residual data Historical context remains accessible via a designated pathway. Is a subsequent investigation practically feasible?

Sources: [1] [2]

Next steps

  • Define data sources, scope, and exceptions.
  • Provide field mapping and conversion rules for release.
  • Perform test transfer involving continuous business activities.
  • Final data status, reconciliation and deviations documented.
Further reading:

Project experience and collaboration

Check imported units, status values and relationships

With IT and process owners, we prepare a known business transaction whose quantities and mappings were established independently. After a trial migration into Business Central, we compare units, batch, status and relationships as well as totals. Historical information may be migrated into active use or made available in a suitable archive, depending on need. We assess exclusions and unresolved differences together. Final reconciliation covers the records actually transferred in the specific import run.

Frequently asked questions about Business Central data migration

Is a sample sufficient?

The scope must be justified based on data criticality, implementation logic, and available automated controls. Critical rules require targeted counterexamples; a random or arbitrary selection is not sufficient justification.

Sources: [1] [2]

Must historical records be actively transferred to Business Central?

Not mandatory. Decide based on subsequent usage, storage requirements, and need for reconstruction. A validated archive access may be appropriate for certain historical information when content, context, and required readability are preserved. An unvalidated export or a future inaccessible legacy system database does not guarantee this.

Sources: [1] [2]

Who verifies the technical accuracy?

The department responsible for the relevant process within the defined release structure. IT can verify technical completeness, but does not substitute for professional assessment.

Sources: [1] [2]

Is a fault-free configuration package import sufficient?

No. Review selected tables, fields, settings, and the technical application of the transferred data. Technical field validation and GxP evidence strategy serve distinct purposes. Appropriate comparisons are required for completeness, relevance, relationships, and critical processes. The scope is determined by data risk and specific transformation requirements.

Why should excluded fields be explicitly documented?

Since a technically successful transfer does not necessarily confirm its content. An excluded field may be non-critical or contain essential quality information. The decision requires professional justification. Additionally, verify whether the objective establishes a standard value that could mask an unconfirmed transfer.

What do we do when a final import must be restarted?

First, define the current target state achieved. Assess whether further processing adds, overwrites, or duplicates data. Use the intended retry pathway and perform appropriate reconciliations. A new technical success does not automatically confirm that prior partial processing remained free of impact.

Sources and review date

The sources provide the technical foundation. Tables, checklists, and examples are derived methodological tools, not verbatim regulatory requirements. Manufacturer information must be referenced to the specific version and configuration in use.

  1. European Commission: EU GMP Annex 11, Computerised Systems (2011)Current version as per EudraLex inventory as of 14.10.2026. Application, risks, operation, and data lifecycle; specific testing proposals are derived independently.
  2. Microsoft: Prepare a configuration package, Business CentralSupplier source for product functionalities. Availability and behavior depend on version, edition, license, and configuration; no general compliance certification.

Sources checked: .

Daniel Herrmann, Daniel Herrmann Consulting

Daniel Herrmann Consulting

Verify professional data transfer from Business Central

We link the actual import volume to the functional area and IT through mapping, data reconciliation, and selected business processes. Technical import and subject-matter verification remain distinct.

  • GxP system decommissioning →

    Retire legacy systems while preserving data, context and auditability. We plan and manage the decommissioning so that data, metadata and audit trails remain readable and retrievable after shutdown.

Send a project enquiry Customer testimonials →