Legacy ERP modernization case study

Legacy ERP Modernization Case Study (Microsoft GP → NetSuite)

A $90M distribution company on Microsoft Dynamics GP for 12 years migrated to NetSuite OneWorld with full integration suite. We extracted historical data, replaced 14 custom integrations, redesigned the chart of accounts, and stood up modern AP/AR automation in 5 months.

Client profile: Composite case study based on a $90M B2B distribution company on Microsoft Dynamics GP for 12 years, 4 distribution centers, ~1,800 active SKUs, 14 active integrations to GP (custom .NET integrations, ODBC connections), 8-person finance team. Microsoft mainstream support ending; modernization mandatory.
See measured resultsTalk through your situation
Company contextBefore — what was actually brokenWhat Ledger Summit implementedLegacy ERP migration mechanicsImplementation timelineMeasured results

Company context

The client is a $90M B2B distribution company that had been running on Microsoft Dynamics GP since 2013. Microsoft's end-of-mainstream-support window forced the modernization decision; the company had been deferring it. By the time we engaged, GP had 12 years of accumulated customizations: 14 custom integrations to upstream/downstream systems, a heavily modified chart of accounts, custom fields documenting domain logic, and a SQL-server-back-end that finance accessed via direct ODBC for ad-hoc reporting.

Legacy ERP migrations differ from QBO → NetSuite migrations in three ways: (1) data extraction is harder — older databases use complex schemas without modern API access; (2) custom integrations need to be inventoried and rebuilt against modern APIs; (3) tribal knowledge of the system lives in 1–2 long-tenured employees who become single points of failure during migration.

  • $90M B2B distribution
  • 12 years on Microsoft Dynamics GP
  • 4 distribution centers
  • 1,800 active SKUs
  • 14 custom integrations (EDI, banking, payroll, e-commerce, manufacturing data)
  • 8-person finance team
  • Microsoft mainstream support ending
  • Modernization mandatory; not optional

Before — what was actually broken

  • Microsoft Dynamics GP back-end on SQL Server (2008/2014)
  • 14 custom integrations including EDI, ACH, e-commerce platforms
  • 600-account COA accumulated over 12 years
  • 60+ custom fields documenting business logic
  • Direct ODBC access for finance ad-hoc reporting
  • 2 senior accountants holding tribal knowledge
  • No documented integration inventory
  • No documented business rules in custom fields
  • Annual maintenance fees rising
  • Microsoft mainstream support window expiring

What Ledger Summit implemented

  • Data extraction strategy: 12 years of GL summary balances + transaction-level for last 3 years; sub-ledger detail (customers, vendors, items, projects, fixed assets); custom field semantics documented and migrated where applicable
  • Chart of accounts redesign: 600 → 280 active accounts, normalized for NetSuite roll-up
  • Custom integration replacement: 14 inventoried, classified (rebuild / replace / retire)
  • EDI integration: rebuilt against NetSuite via Boomi or DiCentral
  • E-commerce integration: rebuilt via Celigo / native NS connectors
  • Banking integration: rebuilt via NetSuite's native bank connectivity
  • Manufacturing data: bridged via Boomi to NetSuite Advanced Manufacturing
  • AP automation: Bill (BILL.com) deployment with three-way match
  • Fixed-asset register: extracted, validated, opening balance loaded
  • Tribal knowledge capture: long-tenured employees interviewed, business rules documented
  • Parallel close: 1 month parallel run before cutover
  • Hypercare: 8 weeks daily standups

Legacy ERP migration mechanics

ChallengeApproach
Data extractionSQL queries on legacy SQL Server; custom ETL where APIs absent
Schema mappingEach legacy field mapped to NetSuite field; custom field semantics documented
Integration inventoryEach integration: source, destination, frequency, owner, business purpose
Integration replacementRebuild: native connectors. Replace: middleware (Boomi/Celigo). Retire: obsolete
Tribal knowledge captureInterview long-tenured employees; document business rules; reverse-engineer custom fields
Historical data — summaryGL monthly summary balances by account-sub-dept for 12 years
Historical data — detailTransaction-level for 3 years; older stays in legacy read-only
Sub-ledger detailCustomers, vendors, items, projects, fixed assets
Cutover strategyParallel close; reconcile to zero; documented rollback plan
User retrainingRole-specific training; recorded walkthroughs; hypercare ticket queue

Implementation timeline

  • Months 1: Discovery: integration inventory, COA review, tribal knowledge capture, project plan
  • Month 2: NetSuite configuration: COA, dimensions, integrations design, AP automation
  • Month 3: Data migration: master data, opening balances, transaction-level history
  • Month 4: Parallel close: NetSuite alongside GP for one month; reconcile to zero
  • Month 5: Cutover; production close on NetSuite; hypercare for 8 weeks

Measured results

MetricBeforeAfterDelta
ERP systemMicrosoft Dynamics GPNetSuite OneWorld—
Custom integrations14 (custom .NET / ODBC)14 (modern APIs / Boomi)—
Chart of accounts600 accounts280 active accounts−53%
Close cycle12 days5 days−58%
Manual JE volume~120 / close~30 / close−75%
Reporting capabilityDirect ODBC + ExcelSuiteAnalytics—
Microsoft maintenance fees~$80K / yr$0−100%
NetSuite license—~$50K / yr—

Alternatives considered

OptionTimeCostStrengthsWeaknesses
Microsoft Dynamics 365 Finance (cloud upgrade)9–12 months$420K–$680KMicrosoft pathMigration depth similar to NetSuite
Sage Intacct7–10 months$280K–$420KStrong financialsLess inventory depth
SAP Business One9–12 months$320K–$520KBrandMid-market unfit at $90M
Stay on GP—Rising maintenance + riskNo migrationForced eventually
Ledger Summit + NetSuite (selected)5 months$240K–$380KRight-sized; controller-ledMaintenance ongoing

When this approach fits

  • $25–250M companies on legacy ERP (Dynamics GP / SL / NAV, JD Edwards, Sage 50/100, Oracle EBS)
  • Vendor end-of-support window approaching
  • Material custom integrations needing replacement
  • Distribution / manufacturing / multi-location services
  • Existing technical capacity to participate in extraction
  • Tribal knowledge holders willing to engage

Lessons learned

  • Tribal knowledge capture is mandatory. Long-tenured employees know business rules nobody else does; document before migration.
  • Integration inventory at week 1. Surprise integrations during migration produce timeline slips.
  • Don't copy COA verbatim. Legacy COA accumulated over 12 years; redesign for the future state.
  • 3 years of detail; older summary. Full transaction-level migration of 12 years rarely worth the cost.
  • Parallel close is non-negotiable. Even with confidence, run NetSuite alongside legacy for one cycle before cutover.

Frequently asked questions

Why not stay on legacy?

End-of-support means rising risk, no security patches, no new functionality, vendor support quality declining. Eventual migration mandatory.

Why not Microsoft's upgrade path (D365 Finance)?

Valid alternative; migration cost similar. Choice depends on existing Microsoft footprint, ecosystem fit, and team preferences.

How much history to migrate?

3 years of transaction-level + 12 years of monthly summary balances is typical pattern. Older transaction-level rarely worth the cost.

What about custom code?

Each custom integration / report / form: rebuild, replace with native, or retire. Inventory at week 1.

How do you handle tribal knowledge?

Interview long-tenured employees; document business rules in writing; reverse-engineer custom fields. Critical pre-migration work.

What's the typical cost?

$240K–$680K depending on scope. Plus ongoing license + integration maintenance.

How long does this take?

5–9 months for $50–250M companies. Larger / more complex run 9–12 months.

What about integration with manufacturing systems?

Inventoried separately; bridged via middleware (Boomi, Workato) until full replacement.

Do you provide post-migration support?

Yes — hypercare for 8 weeks; ongoing managed support optional.

Can we phase the migration?

Sometimes — finance first, operations later, or vice versa. Increases project complexity but reduces risk for some shapes.

Stuck on legacy ERP with end-of-support coming?

A 30-minute call walks your stack and tells you the realistic migration timeline.

Book a free call