When you need this

  • An ERP, CRM or other core system is being replaced. The new environment exists, but the data side is not yet anyone's responsibility.
  • An old system has to be switched off. A licence ends, support ends, or hardware is being replaced, and the data has to reach its new home before that.
  • Two systems are being brought together. After a merger or a reorganisation, the same data sits in two places and in two different shapes.
  • A previous attempt stalled. A test migration was run, but the numbers did not add up and nobody knows where the difference arose.
  • A tender includes a migration item. The description has to be written so that bidders are pricing the same volume of work.

What we do

  • An inventory of the source data — which systems, tables and files actually hold what has to move, and who owns them.
  • Source-to-target data mapping — field-level mapping with transformation rules, mandatory fields and default values.
  • Data-quality rules — duplicates, missing mandatory values, broken references, format errors. Every finding comes with a decision: fix it, leave it, or do not migrate it.
  • Decisions on history and retention — what goes into the new system, what goes to the archive, and what stays read-only. The retention period for personal data belongs here too.
  • Test migrations — repeated dry runs with real data, each one with a measurable result.
  • Reconciliation and checks — documented checks that let you say the source and the target agree: volumes, totals, keys, and sample-based content checks.
  • A cutover plan and a rollback plan — a timed sequence of steps, the people responsible, the decision points, and exactly what happens if the switchover has to be stopped.
  • Running the migration — driving the work to the agreed plan, with status reports and a log of open questions.

For the bulk build of transformation and load pipelines we bring in a partner who works from the same mapping and the same quality rules. The mapping, the quality rules and the reconciliation checks stay our responsibility, so the basis for acceptance does not depend on who wrote the scripts.

Why this is analysis work

A failed migration looks technical, but it is almost always a question of substance. Two systems understand the same concept differently: a customer is a contract in one place and a person in the other, a date is the time of the transaction in one place and the time of the entry in the other, and one system has three statuses where the other has seven.

These are questions of concepts and models, and they are settled before the first load. The same chain of artefacts repeats across all our services: source, rule, check. In analysis it is requirement, control and evidence. In migration it is field, transformation rule and reconciliation check.

What you receive

  • Data mapping

    Source and target fields mapped, with transformation rules and exceptions.

    FormatsXLSXMDSQL
  • Data-quality report

    Findings with volumes, impact and decisions.

    FormatsXLSXMDPDF
  • Test migration results

    A summary of each run: what moved, what got stuck, what was fixed.

    FormatsXLSXMD
  • Reconciliation checks

    A list of checks with their queries and expected results, repeatable after the switchover as well.

    FormatsSQLXLSXMD
  • Cutover and rollback plan

    Steps, times, the people responsible, decision points, and the conditions for stopping.

    FormatsDOCXMDPDF
  • Decision log (ADR)

    What was decided about history, exceptions and missing data, and why.

    FormatsMDSHEETS

All deliverables are handed over in editable formats. The reconciliation checks can be re-run a year later, when somebody asks where a number came from.

Security in this work

A migration is a period when real data moves more than usual and exists in two places at once. Part of the work is therefore about who sees which data in test environments, when a masked or reduced copy is used, and how long test copies are kept.

The other part is the trace. Which data was moved, when and by whom, which rule changed it, and on the basis of which check it was accepted. That trace is also what an auditor or a data protection officer asks for, and it is easier to collect while the work is happening than afterwards.

Where this comes from

The founding analyst has 13+ years of practice with systems, data models and interfaces in banking, telecom, IT consulting and logistics. Those are exactly the fields where bringing data together, quality rules and reconciliation are daily work, because a number has to add up even when it is asked about a year later.

Qualification: BCS Foundation Certificate in Business Analysis. Working languages Estonian and English.

Further reading

We write in Estonian first. The pieces below have not been translated yet and open in Estonian.

Ways to work together

Each of these is a starting point; we settle its scope in conversation and fix it in a written offer. The steps can be taken separately, or one after another in a logical sequence.

System replacement readiness assessment

A fixed-scope written assessment, before the decision or before the tender. It answers the question of what data you are dealing with, what is in order in it, and what has to be resolved before the switchover.

What it includes:

  • An inventory of the source systems and data sets
  • A first picture of data quality, with volumes
  • Risks, dependencies and assumptions for the switchover
  • A recommended approach and order of work

Data mapping and quality rules

The mapping and rules layer you can load from and check against. This is a deliverable that can be handed to any implementer.

What it includes:

  • Field mapping with transformation rules
  • Quality rules, and decisions on the exceptions
  • Decisions on history, archiving and retention
  • Reconciliation checks with their expected results

Migration analysis and management

Driving the substance of the whole migration, from mapping through to acceptance. One person accountable, holding the rules, the test migrations and the checks together.

What it includes:

  • A migration plan with its stages and decision points
  • Running the test migrations and analysing the results
  • Reconciliation, and the basis for acceptance
  • Status reports and a decision log throughout the work

Switchover support

Support during the cutover week and after it, when decisions have to be made quickly and in writing. Suits you when the implementer is in place but the data side needs an independent head.

What it includes:

  • A review of the cutover plan and the rollback plan
  • Presence during the switchover, in an agreed form
  • Running the checks and confirming the results
  • Follow-up checks and closing open questions after the switchover

Describe your situation in one message.

Write down which system the data moves from and which it moves to, and by which date the old one has to close. We talk the scope through and send a written offer. We reply within one working day, and the first discussion is free.