What the job involves day to day
You get data out of the old system, clean it, reshape it and load it into Dynamics 365. Microsoft's implementation guidance splits migration data into master data, such as customers, products and vendors, and open transactions, such as sales orders, purchase orders, stock on hand and open balances. The source can be an old ERP or CRM, or a pile of Excel workbooks. On most projects you also build or support the integrations that keep Dynamics 365 talking to payroll, e-commerce, banks, warehouses and reporting.
The planning work matters as much as the loading. Microsoft's guidance says the plan has to cover discovering sources and volumes, documenting scope and cutover, sizing environments, mapping fields and values, running ETL loads, and validating data quality. It also warns that migration runs disrupt other testing, so they should not overlap with it. In practice that makes you the person who negotiates test windows with the project manager every few weeks.
The finance and operations toolset
In Dynamics 365 Finance, Supply Chain Management and Commerce, the main tool is the data management framework, opened from the data management workspace. You build data projects that hold configured data entities with their mappings. Running a project creates a data job, and a job can produce a data package: one compressed file with a manifest and data files. A data entity presents several normalized tables as one business object, such as Customers, so you load against business concepts rather than the physical schema.
Load order is a big part of the skill. Entities inside a package run in a set sequence. Microsoft's example is that sales tax codes have to load before sales tax groups. The framework lets you skip failed records, keep the good ones and fix the rest later. Entities can also be exposed as OData services for synchronous integrations. For whole environments, database movement operations can restore a signed-off golden configuration from a sandbox to production for go-live.
Dual-write, Dataverse and Azure
Dual-write is Microsoft's built-in, near-real-time, two-way link between finance and operations apps and Dataverse, which is where the customer engagement apps store their data. A change on either side writes to the other. It installs as two Marketplace solutions on Dataverse and adds concepts such as company and party to the Dataverse schema. Customer engagement architects need to understand those changes early, or they end up redoing design work. The standard maps cover customers, vendors, products, prospect-to-cash and procure-to-pay.
For bulk movement and transformation, Microsoft's guidance lists data import and export wizards, Azure Data Factory or Azure Synapse Analytics, SQL Server Integration Services, and non-Microsoft integration products as common ETL options. One documented Azure pattern exports Dataverse data to Data Lake Storage Gen2 with Azure Synapse Link for Dataverse, then transforms it in Data Factory dataflows. Many teams also use Logic Apps and other Azure integration services for event-driven interfaces, so expect clients to ask about them.
Why this role decides go-live dates
Configuration can be finished and the processes signed off, and the project still can't go live if opening balances don't reconcile or half the item master fails validation. Microsoft recommends a dedicated, properly sized migration environment, keeping all databases in the same region, and testing the migration at least once in both SIT and UAT. Experienced consultants go further. They run full cutover rehearsals, time every step and agree the reconciliation reports with finance before the real weekend.
Microsoft's guidance names three roles here. The data migration analyst plans the work and gathers requirements. The data steward owns the definitions and rules for each data item. The data migration architect or developer builds the packages, transforms the data and checks its quality. Consultants in this role usually cover the analyst and developer parts, while the client provides the steward. That split is why so many cleansing tasks end up waiting on client decisions.
Pay
USD a year, before tax. D365 Market advertised pay from US adverts that state pay, snapshot 2026-11-05, Dynamics 365 category (48 adverts, 34 employers). There is no separate data migration category, so these are broad Dynamics 365 advertised figures, not earnings and not an official rate, and they skew toward US states with pay transparency laws.
What you need
- Data management frameworkComfort building data projects, mapping entities to staging, sequencing loads and reading job history to fix errors in finance and operations apps.
- Dataverse and dual-writeKnowing how Dataverse tables, the dual-write maps and the company and party concepts behave when two apps share the same records.
- SQL and ETL toolingStrong SQL for profiling and cleansing source data, plus hands-on use of at least one ETL option such as Azure Data Factory or SSIS.
- Business process knowledgeEnough finance and supply chain understanding to know why open balances, stock on hand and open orders must reconcile at cutover.
- Cutover planningAbility to write a timed cutover runbook, run rehearsals and agree reconciliation sign-off with the business.
- CertificationNo mandatory certification applies to this work. Microsoft certifications are optional proof of product knowledge.
How to get there
- Year 0-2Build data foundationsWork in SQL, reporting or a legacy ERP or CRM support role. Learn to profile, cleanse and reconcile data.
- Year 1-3Join a Dynamics 365 project teamTake load tasks on a partner or end-user project. Learn data entities, data packages and load sequencing in a sandbox.
- Year 3-5Own a migration workstreamWrite the migration strategy, mapping documents and cutover runbook. Run test migrations in SIT and UAT.
- Year 4-6Add integration designDesign dual-write, OData and Azure Data Factory interfaces, and learn the error handling and monitoring behind them.
- Year 6+Move to architect or leadLead data and integration architecture across several projects, or move into a solution architect role.
Open now on D365 Market
Questions people ask
Is data migration a functional or a technical role?
It sits between the two. You need SQL and ETL skills to move data, and enough process knowledge to know what correct data looks like for finance or sales. Most good migration consultants come from one side and pick up the other.
What is the difference between the data management framework and dual-write?
The data management framework imports and exports data entities in finance and operations apps through data projects, jobs and packages, and it handles most bulk migration work. Dual-write is a near-real-time, two-way sync between finance and operations apps and Dataverse. Projects usually use the framework for the initial load and dual-write for ongoing sync.
Which ETL tools do Dynamics 365 projects use?
Microsoft's implementation guidance lists data import and export wizards, Azure Data Factory or Azure Synapse Analytics, SQL Server Integration Services, and non-Microsoft integration products. The right choice depends on the data sources, types and volumes in scope.
Do I need a Microsoft certification for this job?
No law requires one, and no certification covers data migration on its own. Employers look first for evidence that you have run loads and cutovers on real projects. A product certification can help when you are moving into Dynamics 365 from another platform.
How many test migrations should a project run?
Microsoft's guidance says to test the migration at least once in both the SIT and UAT environments. Experienced teams usually add full timed rehearsals of the cutover so the go-live weekend holds no surprises.
Sources
- Microsoft Learn: Data management overview (finance and operations)
- Microsoft Learn: Data entities overview
- Microsoft Learn: Dual-write overview
- Microsoft Learn: Configuration data and data migration (implementation guide)
- Microsoft Learn: Manage data in finance and operations apps
- Microsoft Learn: Ingest exported Dataverse data with Azure Data Factory
- D365 Market: advertised salaries
