The 5 Phases of an Access Database Migration
Every Access migration that goes wrong does so in one of the same predictable ways. Understanding the five phases — and where each one typically breaks down — is how you avoid them.
The most common mistake in Access migrations is treating them as a single task: "migrate the database."
That framing is what leads to projects that start confidently and then stall somewhere in the middle when requirements that nobody documented surface during testing, or when the VBA turns out to be three times more complex than anyone estimated, or when users reject the new system because nobody involved them until go-live.
A migration is a project with distinct phases. Each phase has its own goals, its own deliverables, and its own characteristic failure modes. Understanding the phases does not guarantee a smooth migration — but it does mean you will not be surprised by the things that routinely derail them.
Phase 1: Discovery (2–4 Weeks)
Discovery is the most important phase and the most frequently rushed. Its goal is to understand exactly what you have before you decide what to build.
The reason organizations rush it is understandable: discovery does not produce anything visible. You are not building anything. You are reading VBA, interviewing users, and documenting things that already exist. It feels like delay.
It is not delay. Every hour spent in discovery saves three hours in later phases. Requirements discovered during testing cost ten times more to address than requirements discovered during discovery. This is not an estimate — it is a pattern I have seen repeat across enough projects to be confident in it.
What happens in discovery:
Database inventory. Document every table, query, form, report, macro, and VBA module. Note the purpose of each, who uses it, and how often. You will find things no one remembers building — and things that are critical but completely undocumented.
Data profiling. Examine the actual data in each table. Look for data quality issues: inconsistent formats, values that should be in lookup tables but are stored as free text, fields that are supposed to be required but have nulls, date fields stored as text. These issues need to be fixed before or during migration — discovering them in Phase 4 is expensive.
Business rules documentation. The most valuable and most time-consuming part of discovery. Interview the people who use the database. What validation rules exist? What are the edge cases? What does the database do that is not obvious from looking at the tables and forms? This knowledge is often in people's heads, not in the database.
Integration mapping. Does the database import data from other systems? Export to Excel for reporting? Connect to external data sources? Every integration needs to be accounted for in the new system.
Deliverable: a discovery document that describes the current system in enough detail to design the new one.
Where this phase breaks down: skipping or shortening it to save time. The time is not saved — it is deferred to a later phase where it costs more.
Phase 2: Design (2–3 Weeks)
Design translates the discovery findings into a blueprint for the new system.
What happens in design:
Data model design. Map the Access tables to the target platform. Resolve the data quality issues found in discovery. Normalize where needed. Define the new primary keys, relationships, and constraints.
Application design. Decide which Access forms become Power Apps screens, which reports become Power BI dashboards, which VBA automation becomes Power Automate flows. Not everything needs to be rebuilt — some Access functionality may be retired if it is no longer used.
Security design. Define who needs access to what. In Access, security is often "everyone can see everything." In SQL Server or Dataverse, you have the opportunity to implement proper role-based security. Design it now, not after go-live.
Migration strategy. Decide how data will move from Access to the new system. What is the cutover plan? What happens to historical data?
Deliverable: a design document with the new data model, application architecture, security model, and migration strategy. This document should be reviewed and approved by stakeholders before Phase 3 begins.
Where this phase breaks down: designing the new system to look exactly like the old one. Migration is an opportunity to fix problems in the current design. A form that was confusing in Access should be redesigned, not replicated. A table structure that was wrong should be corrected, not copied. The organizations that treat migration as a business process redesign — not just a technology swap — consistently get better outcomes than those that try to recreate what they had.
Phase 3: Build (4–12 Weeks, Depending on Complexity)
Build is where the new system is constructed according to the design.
What happens in build:
Database build. Create the SQL Server or Dataverse schema. Set up tables, relationships, indexes, views, and stored procedures. Configure security roles.
Data migration scripts. Write and test the scripts that will move data from Access to the new database. This is more complex than it sounds — data type conversions, lookup table mapping, handling nulls and edge cases, and validating that the migrated data matches the source.
Application build. Build the Power Apps screens, Power Automate flows, and Power BI reports. This is typically the longest part of the build phase.
Integration build. Rebuild any integrations with external systems.
Deliverable: a working system in a development or staging environment, ready for testing.
Where this phase breaks down: building everything before testing anything. Build in increments and test each increment. Discovering that a fundamental design decision was wrong after building 80% of the application is much more expensive than discovering it after building 20%.
Phase 4: Testing (2–4 Weeks)
Testing validates that the new system does what the old system did — and does it correctly.
What happens in testing:
Functional testing. Test every form, report, and workflow against the requirements documented in discovery. Does the new system handle all the edge cases? Does it enforce all the business rules?
Data validation. Compare the migrated data against the source Access database. Row counts, totals, spot checks on individual records. Data migration bugs are common and need to be caught before go-live.
Performance testing. How does the new system perform with production data volumes? A Power App that is fast with 100 test records may be slow with 100,000 production records.
User acceptance testing (UAT). Have actual users test the system with real scenarios from their daily work. They will find things that functional testing misses — workflows that seem correct in isolation but do not match how people actually work.
Deliverable: a signed-off test report confirming the system is ready for go-live, with a list of any known issues and their resolution plan.
Where this phase breaks down: treating UAT as a formality. Users who are not genuinely engaged in testing will find problems after go-live. Make UAT meaningful: give users specific scenarios to test, track their findings, and fix issues before cutover.
Phase 5: Cutover (1–2 Weeks)
Cutover is the transition from the old system to the new one.
What happens in cutover:
Final data migration. Run the migration scripts one more time with the latest data from the Access database. Validate the results.
Go-live. Switch users to the new system. This may be a hard cutover (everyone switches on the same day) or a phased rollout (one team or location at a time).
Parallel running period. For critical systems, run the old and new systems in parallel for a short period — typically one to two weeks. Users can verify their work against the old system. This reduces risk but requires extra effort to keep both systems in sync.
Access database retirement. Once you are confident the new system is working correctly, make the Access database read-only and eventually archive it. Do not delete it immediately — you may need to reference it for historical data or edge cases.
Deliverable: the new system is live, users are trained, and the old Access database is retired or archived.
Where this phase breaks down: a cutover plan that does not include a rollback plan. What happens if a critical problem is discovered on day two of go-live? Know in advance how you would roll back to Access if necessary, even if you never need to use it.
A Realistic Timeline
For a medium-complexity Access database (10–20 tables, 5–10 forms, moderate VBA):
| Phase | Duration |
|---|---|
| Discovery | 2–3 weeks |
| Design | 2 weeks |
| Build | 6–8 weeks |
| Testing | 2–3 weeks |
| Cutover | 1 week |
| Total | 13–17 weeks |
For a large, complex database (50+ tables, extensive VBA, multiple integrations), double these estimates. For a simple database (5–10 tables, minimal VBA), cut them in half.
The most important thing: build the timeline from the phases, not the other way around. A timeline set before discovery is a guess. A timeline built after discovery is a plan.
Work With Access Evolved
If you found this article useful, Access Evolved specializes in exactly this kind of work — from hands-on Microsoft Access support to full database modernization projects.
Want a phase-by-phase plan built for your specific Access database? Schedule a free assessment call.
Thinking about migrating?
Ready to move beyond Microsoft Access?
Access is a great starting point, but growing businesses often hit its limits. Anthony Passeri specializes in migrating Access databases to modern platforms — Power Platform, Power Apps, SQL Server, Azure, and more — so you keep your data and workflows without the headaches.
Written by
Anthony Passeri, MBA
Access Expert and Editor
Bridging the gap between Access and the Future
Explore Topics
Keep reading
More on Migration Cost & Timeline
How Long Does an Access to SQL Server Migration Really Take?
The honest answer to the timeline question — with real numbers based on database size and complexity, not optimistic estimates.
5 Access Databases We\'ve Seen That Were Overdue for Migration
After working with dozens of Access databases, certain patterns appear again and again. These are the five database profiles that tell us a migration is not just a good idea — it is overdue.
How to Estimate Access Migration Cost by Database Size
Migration cost is the first question every Access developer asks. Here is a practical framework for estimating what your specific database will cost to migrate — based on size, complexity, and destination platform.