Phase 01 · Initiation & Strategy

The authorizing document

Project Charter

Artifact · Charterv1.0 · approved by Steering

Purpose

The legacy SQL Server EDW is at end-of-life — capacity-bound at peak, costly to license and maintain, and unable to support the analytics the business now needs. Migrate the data foundation and all reporting to Snowflake Cloud, then retire the legacy platform.

Objectives & success criteria

Migrate ~1,200 objects across 8 functions with zero-defect reconciliation on all financial and revenue metrics.
Migrate or rebuild every in-scope report with proven old-vs-new parity.
Adoption by ~4,500 users, measured, before legacy decommission.
Decommission the SQL Server EDW within a defined confidence window post go-live.

In scope

Finance, Sales, Marketing, Manufacturing Ops, Supply Chain, Services, Quality, Compliance — data objects, ETL/code conversion, reports, and analytics.

Out of scope

Retired/duplicate reports (see rationalization), source operational systems, and net-new analytics use cases — deferred to a post-migration backlog.

Sponsor & roles

Executive sponsor: VP, Enterprise Analytics. Program lead: Technical Program Manager. Workstream leads: Data Engineering, BI/Reporting, Testing, Change Management.

Approach

Wave-based, hybrid Waterfall/Agile. Phase gates govern the program; sprints deliver the build inside each wave. A dual-run period de-risks the switchover.

Top risks (headline)

Data-quality gaps in the source; reconciliation failures on financial metrics; code-conversion complexity; SME availability; adoption resistance at scale.

The justification

Business case & approach decision

Three migration approaches were weighed. The recommendation balances speed against the cost of carrying legacy debt into the new platform.

Artifact · Options analysisrecommended: hybrid

Approach	What it is	Trade-off	Verdict
Lift & shift	Replicate the SQL Server schema as-is on Snowflake.	Fastest, but carries every legacy flaw and un-tuned model forward.	Rejected
Re-architect	Redesign the full data model for Snowflake from scratch.	Best end state, but timeline and risk balloon; business waits too long.	Rejected
Hybrid	Migrate structure faithfully; selectively re-model high-value domains and retire dead objects.	Balanced speed and quality; rationalization pays for itself.	Recommended

The financial case rests on retiring SQL Server licensing and infrastructure and maintenance, offset by Snowflake consumption and one-time migration cost, with the payback driven as much by retiring 150 dead reports as by the platform swap.

Decision rights

Stakeholder map & RACI

Who decides is settled before delivery starts — so ambiguous calls resolve once, in the right forum, rather than repeatedly at the keyboard.

Artifact · RACIR responsible · A accountable · C consulted · I informed

Key decision / deliverable	TPM	Exec Sponsor	Data Eng Lead	Function SME	BI Lead	Change Lead
Field-mapping sign-off	A	I	C	R	C	I
Report disposition (migrate/retire)	R	I	I	A	C	I
UAT sign-off	A	I	C	R	C	I
Go / no-go decision	R	A	C	C	C	C
Legacy decommission	R	A	C	C	I	I

The highest-leverage early move

Rationalize before you migrate

The current-state assessment inventories both the source (tables, views, stored procs, ETL) and every downstream consumer — and who actually uses each one. In a real EDW, a large share of reports are dead or duplicated. Retiring them is the cheapest scope reduction available: nobody thanks the TPM who faithfully migrated 400 reports when 150 hadn't been opened in a year.

400

Reports inventoried

150

Retired (dead / duplicate)

250

Carried into scope

Golden thread

The Sales/Revenue close-amount report is confirmed high-use and in-scope. Its owner — a Sales Finance SME — is named here and stays accountable for the field's meaning through mapping, UAT, and cutover.