Practical guides · Practical guide

Data migration: testing time zones and daylight saving changes

Different times in source and target may indicate a correct time zone conversion or a migration error. Determine whether the sequence of manufacturing, testing and release was preserved. A comparison of the displayed characters gives little evidence on its own.

Table of contents

At a glance

How do you test timestamps during GxP data migration?

Define the meaning, source format, time zone, UTC offset and required precision of each time field. Compare the represented point in time and event order, not just the character string. Test daylight saving transitions, ambiguous local times, missing zones and rounding. Preserve the original historical information and transformation rule to the extent needed.

First, clarify what the time field means

A time field may represent measurement time, entry time, receipt time, approval time, or technical change date. These meanings must not be confused during migration. For example, a target import timestamp does not replace the original approval time. Therefore, capture the field name, professional definition, source, and usage. Additionally, verify whether a field denotes a date only—without time.

Separate timestamp from local display. A single timestamp may appear differently across time zones. Conversely, the same local time may represent different timestamps. For reconciliation, it must be defined which information is required: the actual event time, the local time display, or both. A general rule such as "convert all date fields to UTC" is inappropriate when certain fields represent simple calendar dates or professionally agreed local deadlines.

Sources: [1] [2]

UTC offset and time zone designation differ

A UTC offset denotes the difference at a specific point in time, for example +02:00. A time zone identifier such as Europe/Berlin refers to rules that may include historical changes. The IANA Time Zone Database is updated when borders or rules change. Therefore, the current offset alone is insufficient for interpreting all historical data of a region.

Document the regulatory basis upon which the migration procedure is based. Assess historical time periods and locations that may differ from current configurations. Avoid ambiguous acronyms such as CST without additional definition. If the source already contains a clearly defined UTC offset, treat it with controlled validation. If it only records a local time, an appropriate additional mapping must be in place. Missing information cannot be substituted with a technically plausible assumption.

Sources: [3]

Sample calculation for autumn transition

The following example uses Europe/Berlin according to the verified rules for 2026. On October 25, at 02:30, an event occurs twice. At 02:30 with a timezone offset of +02:00, the local time corresponds to 00:30 UTC; at 02:30 with a timezone offset of +01:00, it corresponds to 01:30 UTC. Local time strings without spacing appear identical, but the events occur one hour apart. An import that treats both occurrences as the same timestamp loses this distinction.

For the spring period, a different edge case will be evaluated: On 29 March 2026 at 02:30 UTC in the Europe/Berlin time zone, there is no regularly occurring local time. The procedure must define a specific handling for such inputs. A silent adjustment to another time point would constitute a substantive transformation that must not occur without proper notice. The specific calendar events will be derived from the active time rules for all relevant regions and time periods; not all regions use the same transition dates.

Expected results for unambiguous, repeated and nonexistent local times
Source informationExpected InterpretationDecision on testing
25.10.2026 02:30 +02:00 00:30 UTC Earlier of the two events
25.10.2026 02:30 +01:00 01:30 UTC Later event received
25.10.2026 02:30 without additional information Not solely unambiguous Clarify additional context
02:30, 29 March 2026, Berlin Time Zone A local clock time that does not occur during the normal transition Handle defined failure scenario
Date without time Calendar day according to professional definition No unjustifyecd daily shift
Transfer time at destination Time the data was imported into the target system Do not present it as the original event

Multivalent and incomplete sources addressed

Prepare a list of unexplained time fields and known edge cases. Verify whether additional data unambiguously determines the timestamp: source system configuration, recorded interval, event sequence, or a linked original protocol. Any assignment must be clearly justified. The current time zone of a server is not sufficient evidence to determine which rule was applied to an older record.

If the information cannot be reliably reconstructed, uncertainty is explicitly assessed. Retain the original entry and label any derived interpretation within the appropriate context. Avoid inventing seconds or arbitrary daylight saving time adjustments solely to achieve technically unambiguous values. The impact on process, history, and required decisions is evaluated by the responsible parties. A successful database conversion does not close this professional gap.

Sources: [1] [2]

Verify accuracy, sequence, and identical time stamps

Compare resolution and rounding rules of source and target. If milliseconds are lost, multiple events may be assigned the same displayed timestamp. Verify that additional sequence numbers or relationships preserve the required order. The target does not need to inherently possess greater accuracy than the source; it may, however, not lose any required information without notice. A artificially precise timestamp does not enhance historical accuracy.

Review the sequence of technical events: result recorded, modified, reviewed, and approved. A time conversion should not infer an earlier action based on a later approval. Additionally, consider that unsynchronized source systems may have previously exhibited differing time stamps. Data migration and original time accuracy are distinct issues. Do not silently correct historical discrepancies; instead, evaluate and document any required corrective actions.

Sources: [1]

Separate transformation and display alignment checked

Verify the stored timestamp and the displayed surface. A target system may internally store time in UTC, but each user profile may display a local time. Testing using only the developer's profile does not cover other authorized settings. Use the intended roles and relevant regions. Specify whether printouts and exports must unambiguously indicate time offset or time zone.

The reference comparison is prepared independently of the migration code. Use controlled, known examples and appropriate comparison methods. Only after a defined conversion is applied and source and target values are found to be equivalent is the specific rule substantiated. A simple text comparison would incorrectly flag a correct case as a failure. A comparison based solely on date could, in contrast, overlook an actual time discrepancy. The inspection method must align with the technical and functional requirements of the field.

Sources: [3] [1]

Delta transfers and cross-system chains are considered.

During a migration, additional events may occur. Define the cut-off as a clear point in time and verify its usage across all involved systems. A local timestamp without a time offset may create ambiguity precisely on the cutover day. Describe which events belong before, at, and after the boundary. Test the handling of equality and late-arriving messages.

For multiple systems, evaluate the entire data chain. ERP, laboratory application, and archive systems may employ different representations. A unified search must still demonstrate the correct temporal relationship. Account for receipt times separately from the original event times. If a dataset is imported on the following day, its substantive origin must not appear shifted solely due to the technical receipt date. Requirements for traceability evidence are defined prior to productive implementation.

Sources: [1] [2]

Test protocol and permanent investigation

Document field semantics, formats, rule levels, conversion, reference cases and findings. The conclusion indicates unexplained sources and their impact. If migration introduces a new representation, you will explain it to later reviewers. A historical expression with 02:30 and a target display with 00:30 UTC can describe the same process; without an understandable rule, this acts as a contradiction.

Retain the required original information and transformation description as per the approved concept. Changes to time zone rules or display programs may affect historical representations and require appropriate evaluation. The suitability determination should demonstrate that events, sequence, and required accuracy are preserved. Identical numbers of imported time fields or a flawless conversion run alone do not prove these attributes.

Sources: [1]

Project experience and collaboration

Preserve the instant in time when the displayed clock time changes

With IT and process owners, we select historical events whose time and order matter to later review. We compare the source value, time zone, conversion rule and target display. Daylight saving changes reveal where a locally stored time is ambiguous. A few affected fields may need only focused comparison. A fundamental change in time formats or rules also requires preservation of the original information and later access to it.

Frequently asked questions about Time zones and timestamps in data migration

Is a different timestamp in the target always a migration error?

No. A correct time zone conversion may result in a different local display. Verify the actual timestamp, UTC offset, and the meaning of the field. A simple text comparison is inadequate for this purpose. The conversion logic and expected display must be clearly defined and validated.

Is the current server time zone sufficient for historical data?

Not automatically. Historical rules and previous system configurations may differ. Verify time zone designation, time period, and available source information. A current UTC offset must not be universally applied to all past events. The rule basis employed belongs to the documented transformation.

Sources: [3]

What does one do with a duplicate local time?

Look for appropriate additional context, such as recorded UTC offset or a verifiable event correlation. Without such information, local timestamps may remain ambiguous. Assess and document this limitation. Arbitrary selection of an earlier or later time point does not generate reliable historical conclusions.

Must empty date fields be converted to UTC?

Not as a general rule. A calendar date can have a different meaning from a point in time. An unjustified conversion can shift the date. Define the meaning of each field and test the appropriate transfer rule, including for fields with similar technical formats.

Can the objective of millisecond removal be achieved?

Only if the required statement and sequence remain intact and the decision is justified. Verify identical target time points and any additional required sequence information. A loss of critical resolution must not occur without detection. Additional artificial precision without a source is likewise not an appropriate substitute.

Which summer season cases are included in the inspection?

Relevant spring and autumn transitions, ambiguous or non-occurring local times, and affected historical regions. Extend interval boundaries, delta cut-off values, and user representations. Specific cases are derived from source, time period, and applied time zone rules, not from a universal calendar list.

Sources and review date

Status: October 14, 2026. Regulatory classification is based on the cited original sources. Examples and test matrices are internal support tools; they are adapted for actual usage purposes.

  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. Annex 11 No. 4.8 requires, upon transfer to another format or system, verification that the value and meaning of the data have not been altered. The time-based examples are individual test cases.
  2. PIC/S: PI 041-1, Good Practices for Data Management and Integrity (2021)Inspection Guidance on Data Integrity; in particular electronic data, transmission, access, and review.
  3. IANA: Time Zone DatabaseHistorical UTC spacing and daylight time rules are changing; taking into account time zone identification and control level. Tested on 14.10.2026.

Sources checked: .

Daniel Herrmann, Daniel Herrmann Consulting

Daniel Herrmann Consulting

How we can support you

Daniel Herrmann Consulting combines professional assessment with hands-on work on the specific system. During the first meeting, we clarify your intended use, existing documentation, and the next most appropriate action step.

  • 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 →