Every integration involves migrating systems. And every migration contains traps for the unprepared.
According to Gartner, 83% of data migration projects fail or exceed their budget and timelines. KPMG estimates an average budget overrun of $300,000 per data set in failed migrations.
This isn't because migration is impossible. It's because the non-obvious problems (the structural mismatches, the Conversion Trap, the process vs. data gaps) catch teams who thought they were prepared.
This article covers the principles and pitfalls for the core systems: email, CRM, ERP, HR, and operational tools. Not every platform or scenario, but the things that derail otherwise solid plans.
This article is adapted from Chapter 6 of The Roll-Up Integration Playbook, our free guide to post-merger technical integration.
Migration Sequencing: Why Order Matters
The order you migrate systems matters. Some migrations create dependencies for others. Some are lower risk and good for building confidence before tackling harder ones.
Typical Sequence
| Order |
System |
Timeframe |
Rationale |
| 1 |
Email & Identity |
Days 1-30 |
Security perimeter, affects everyone, relatively straightforward |
| 2 |
CRM |
Days 30-60 |
Sales visibility, customer data unified early, tests migration process |
| 3 |
Finance/ERP |
Days 45-90 |
Requires fiscal calendar alignment, parallel running |
| 4 |
HR & Payroll |
Days 30-60 |
Often parallel with CRM, driven by payroll cycles |
| 5 |
Operational Systems |
Days 60-90+ |
Most complex, benefits from earlier lessons, may stay separate |
Why This Order Works
Email and identity come first because they affect everyone and establish your security perimeter. They're also relatively low risk compared to business systems.
CRM comes next because sales visibility is critical and you want customer data unified early. It's also a good test of your migration process before tackling finance.
Finance waits for a natural cutover point (month-end, quarter-end) and requires more preparation. It often runs parallel with later stages of CRM migration.
HR can run parallel with CRM since they're largely independent systems. Timing may be driven by payroll cycles.
Operational systems come last because they're often the most complex and benefit from lessons learned on earlier migrations.
When to Deviate
- If finance reporting is your biggest pain point, ERP may come earlier
- If the acquired company's CRM is actively losing deals, that becomes urgent
- If security concerns are severe, you might do a Day 1 email cutover
- If operational systems are simple, they might move earlier
Let your business priorities drive the sequence, not a rigid playbook. For the overall execution timeline, see The First 100 Days After an Acquisition.
Email and Identity: The Conversion Trap
Email migration seems simple. It rarely is.
The platform decision isn't always obvious. If you're on Microsoft 365 and they're on Google Workspace, migration seems straightforward. But if they're a small team who lives in Google Docs, forcing them onto Microsoft may create more friction than it's worth.
The real danger is the Conversion Trap.
What the Conversion Trap Looks Like
When migrating from Google Workspace to Microsoft 365, Google Docs, Sheets, and Slides must be converted to Office formats. This sounds simple but causes real problems:
| Google Feature |
What Happens in Conversion |
| Complex formulas with IMPORTRANGE |
Breaks completely |
| Google Apps Scripts |
Stops running |
| Embedded links to other Google files |
Broken references |
| Heavy formatting in Google Docs |
May not convert cleanly |
| Query functions in Sheets |
Often fail or produce errors |
Before migration, scan for complex files that need manual remediation. Don't discover these problems on migration day.
Example: The Spreadsheet That Runs the Business
An acquired company has a Google Sheet that pulls data from three other sheets using IMPORTRANGE, calculates sales commissions with custom formulas, and triggers email notifications via Google Apps Script. It's ugly but it works, and the sales manager depends on it.
When you migrate to Microsoft 365, this sheet becomes a broken Excel file. The IMPORTRANGE functions fail. The Apps Script stops running. The formulas may convert incorrectly. Nobody told you this spreadsheet existed because "it's just a spreadsheet."
Before any email/identity migration, audit for these dependencies. The "simple" migration becomes complex when you discover what people have built on top of the platform.
Timing: Email migration typically takes 2-4 weeks from planning to completion. If doing Day 1 cutover for security reasons, preparation must happen pre-close.
CRM: Data Quality and Structural Mismatches
CRM migration is often the highest-risk migration for business disruption. Get it wrong and your sales team loses visibility, deals fall through cracks, and customer relationships suffer.
The Data Quality Reality
Expect problems. Research shows that data quality issues affect most migrations:
- Duplicate records - the same customer entered multiple times
- Incomplete data - required fields empty or filled with placeholders
- Outdated information - contacts who left years ago
- Inconsistent formatting - phone numbers, addresses, names entered differently
Budget significant time for data cleanup before migration. A "quick" CRM migration that skips cleanup becomes a slow migration when you're fixing problems post-go-live.
Deloitte research shows that 32% of executives identify data integration as the single biggest challenge in M&A. CRM is where this challenge is most visible.
Process vs. Data Mismatch
The bigger challenge is often process, not data. The acquired company's sales process may be fundamentally different from yours:
- They might track different pipeline stages
- They might use different fields
- They might define "opportunity" differently
- Their lead qualification criteria might not match yours
Simply migrating data from their system to yours doesn't solve this. You need to decide:
- Are they adopting your process?
- Are you creating a hybrid?
- Are you running separate processes within one system?
Make these decisions before migration, not during.
Example: HubSpot to Salesforce
This is one of the most common CRM migrations, and it's more complex than it appears.
| HubSpot Concept |
Salesforce Concept |
The Problem |
| Companies & Contacts |
Accounts & Contacts |
HubSpot allows flexible associations; Salesforce has stricter parent-child hierarchy |
| Deals |
Opportunities |
Stage definitions, pipeline structures, and required fields are completely different |
| Activity tracking |
Activities |
Data stored in different structures |
| Custom properties |
Custom fields |
Different data types, picklist values need recreation |
The result: you can't just export from HubSpot and import to Salesforce. You need field mapping, data transformation, decisions about structural differences, and often significant data cleanup in between.
Timing: CRM migrations typically take 8-12 weeks including data cleanup and training. Simple migrations (clean data, similar processes) can be faster. Complex ones take longer.
Finance and ERP: Parallel Running and Dimensional Mismatches
Finance migration keeps CFOs up at night, and for good reason. Errors in financial data have regulatory implications, affect reporting to investors, and can take months to untangle.
Parallel Running Is Essential
For finance systems, run both systems in parallel until you're confident the new system is producing accurate results. This typically means:
- Both systems active for at least one full accounting period
- The same transactions entered in both (or one system feeding the other)
- Reports generated from both and reconciled
- Cutover only when outputs match consistently
Yes, this is extra work. It's also how you avoid a financial reporting disaster.
Research shows that only 57% of successful acquirers fully integrate systems and processes. For finance, partial integration without proper parallel running is how errors compound.
Chart of Accounts Alignment
The Chart of Accounts (CoA) is where technical migration meets business decisions. The acquired company's CoA won't match yours.
Mapping between them requires decisions about:
- How to categorise revenue and expenses
- What level of detail to maintain
- How to handle historical data under old categories
This isn't a purely technical exercise. Finance leadership needs to drive CoA mapping decisions. Start early because these conversations take longer than expected.
Example: Xero to NetSuite
A common scenario: your platform runs NetSuite, and you acquire a company on Xero. Both are accounting systems. How hard can migration be?
| Xero |
NetSuite |
The Gap |
| Flat chart of accounts |
Multi-dimensional (account + class + department + location + custom segments) |
NetSuite's power doesn't exist in Xero |
| Tracking categories (limited) |
Full dimensional reporting |
Don't map cleanly |
| Bank reconciliation method |
Different reconciliation approach |
Process change required |
| Multi-currency handling |
More complex multi-currency |
Transformation needed |
| Simple tax codes |
More granular tax configuration |
Mapping required |
The decision: do you flatten NetSuite to match how Xero worked (losing reporting capability), or do you enrich historical data with dimensions it never had (requiring business decisions about how to categorise old transactions)?
The chart of accounts mapping alone typically requires several sessions with finance leadership. Then you need to transform the data, validate it, and run parallel periods to ensure the numbers match.
Timing: ERP migrations take 12-20 weeks and should align with fiscal calendar. Month-end or quarter-end cutovers are typical. Year-end is ideal if timeline permits. Coordinate with auditors if relevant.
HR, Payroll, and Operational Systems
A few critical points on these systems:
HR and Payroll
Payroll accuracy is non-negotiable. People notice when their pay is wrong. Run parallel payroll calculations before cutover, verify tax codes and pension contributions, and have a rollback plan.
The hidden complexity is policy harmonisation:
- Different leave accrual methods
- Different leave years
- Different pension schemes
- Different benefits structures
You can't just migrate the data. You need HR and legal decisions about how to align policies first. The technical migration follows the business decisions.
Under GDPR, employee data requires legitimate basis for processing. Don't migrate more than you need.
Operational Systems
Not every operational system needs to migrate.
If the acquired company runs specialist software that genuinely works for them, consider keeping it separate and integrating at the data layer via API. Jobs and invoices can flow into your finance system for unified reporting without forcing everyone onto a single platform.
The risk is accumulating technical debt: too many systems connected by custom integrations.
Use a Sunset Policy to ensure these decisions are revisited. For the framework, see How to Decide the Right Integration Level for Each Acquisition.
When to Bring in Specialists
Not every migration needs external help. But some are complex enough that specialist expertise pays for itself in speed and reduced risk.
Consider Specialists When
| Situation |
Why Specialists Help |
| Platform complexity |
Enterprise platforms (Salesforce, NetSuite) where configuration expertise matters |
| Data complexity |
Large datasets, complex transformations, significant cleanup required |
| Integration complexity |
Multiple systems with dependencies, custom API work needed |
| Time pressure |
Aggressive timeline where parallel effort accelerates delivery |
| Skill gaps |
Team lacks experience with specific platforms or migration patterns |
| Risk sensitivity |
Finance or other high-stakes migrations where errors are costly |
Keep In-House When
- Small, straightforward migrations with clean data
- Your team has genuine experience with the platforms involved
- Timeline is flexible enough to learn as you go
- The migration is mostly configuration, not complex data work
The Hybrid Model
Most mid-market integrations use a mix:
- Internal team owns project management, business decisions, and user communication
- Specialists handle technical migration work for complex systems
This gives you control without requiring deep technical expertise in every platform.
For more on resourcing options, see In-House, Contractors, or Partners: Who Should Handle Post-Merger Integration?.
Key Takeaways
- Sequence matters. Email/identity first, then CRM, then finance, then HR, then operational systems. Deviate based on business priority, not convenience.
- Watch for the Conversion Trap. Google Docs with complex formulas or scripts will break when converted to Office formats. Scan and remediate before migration.
- CRM data is messier than you think. Budget significant time for cleanup. Process alignment is often harder than data migration.
- Parallel running is mandatory for finance. Run both systems for at least one full period before cutover. Coordinate with fiscal calendar and auditors.
- Not every system needs to migrate. Keeping operational systems separate while integrating via API can preserve value while providing visibility.
- Know when to bring in specialists. Complex platforms, large datasets, tight timelines, and high-stakes migrations often justify external expertise.
Get the Full Playbook
This article covered system migration principles and pitfalls. For the complete guide, including the 100-day execution timeline, training and adoption guidance, and ready-to-use checklists, download The Roll-Up Integration Playbook.
Download the Playbook
About PMI Stack
PMI Stack helps small-to-mid cap roll-ups unify systems, data, and workflows across their acquired companies. We specialise in the technical side of post-merger integration: data migration, system consolidation, and the change management that makes new tools stick.
If you're planning a system migration and want to talk through your specific situation, book a free discovery call.
Statistics cited from Gartner, KPMG, Deloitte, and PwC. For the full research compilation, see 50+ Post-Merger Integration Statistics (2026).
