Standardising Dynamics 365 and Power Platform Across Regions That Work Differently
For groups whose regional units run independently: what to make identical, what to leave local, the solution architecture that holds the line, and a rollout regions will actually adopt.
If your regional units run independently and resist a single model, standardise only what the group genuinely needs to be identical: the customer master, opportunity stages and reporting dimensions. Leave tax and legal identifiers, consent and retention rules, partner structures, language, currency, approval thresholds and local process steps to the regions. Build that split into the architecture: one core managed solution owned centrally, with regional extension solutions layered on top that add but never edit core components. Govern it with a design authority and a single release path so no region can fork the core. Then pilot in one region with measurable outcomes before rolling out. The failure to avoid is a template designed centrally and never used locally.
Why do global Dynamics 365 templates fail in regions that work differently?
Most global templates fail quietly rather than dramatically. Headquarters designs a model in workshops with the largest region, calls it the global template, and deploys it everywhere. Smaller regions find that it cannot record their tax identifiers, does not reflect how their distributors work and asks for approvals their market does not need. They comply on paper and keep the real work in spreadsheets, local tools or a copy of the solution someone modified directly in production. Two years later the group has one CRM in name and several in practice.
The cause is almost never the platform. It is a template that standardised too much of the wrong things and too little of the right ones, with no legitimate way for a region to be different.
- Designed centrally, used locally by nobody: the template matches the headquarters process and no region adopts it fully.
- Everything declared global: every local difference becomes a change request, so regions stop asking and start working around the system.
- Nothing declared global: each region customises freely, and group reporting depends on manual mapping every quarter.
- Unmanaged changes in regional production: a quick local fix edits core components, and the next core release overwrites it or cannot be installed.
- No owner for the core: nobody has authority to say no, so the template drifts toward whichever region shouts loudest.
What must be identical across every region, and what should stay local?
The test is simple: a thing is core if the group cannot report, share customers or move people between regions without it being identical. A thing is local if it is driven by law, market structure or language, or if making it identical would only move the difference into a spreadsheet. Your compliance and tax teams define the legal requirements per country; the table is about where each requirement lives in the system.
| Element | Core or local | Why |
|---|---|---|
| Customer master: account and contact structure, identifiers, duplicate rules | Core | One customer served in two regions must be one record, or group reporting and key account management break. |
| Opportunity stages and their exit criteria | Core | Pipeline across regions is only comparable if a stage means the same thing everywhere. |
| Reporting dimensions: region, legal entity, product line, segment, industry | Core | Group reporting needs the same values, owned in one place. |
| Product catalogue structure | Core structure, local prices | Product identity is shared; price lists and availability vary by market. |
| Tax and legal identifiers | Local values in a core pattern | Each country has its own registration and tax identifiers; the data model to hold them should not be reinvented per region. |
| Consent and retention rules | Local rules in a core pattern | Legal bases, consent wording and retention periods differ by jurisdiction. |
| Partner and distributor structures | Local | Direct sales, distributors, resellers and agents differ by market maturity. |
| Language and currency | Local | Users work in their language and quote in local currency, while reports convert to a group currency. |
| Approval thresholds | Local values, core mechanism | The approval flow is shared; limits and approvers differ by region and entity. |
| Local process steps | Local | Extra checks, documents or hand-offs required by a market, added without changing the core stages. |
How do you settle core versus local when regions disagree?
Disagreement is normal and useful, provided it is resolved by criteria written before the arguments start rather than by seniority. We run the core-versus-local decision as a short, documented series of workshops with each region represented, and each contested element goes through the same questions.
- Does group reporting, a shared customer or a cross-region process break if this differs? If yes, it is core.
- Is the difference required by law or regulation in that country, confirmed by your own legal or compliance team? If yes, it is local, inside a core pattern where possible.
- Is the difference a genuine market difference, or a habit? Habits can be changed; market structure usually cannot.
- Would making it identical move the real process into a spreadsheet? If yes, keep it local.
- Can the difference be expressed as configuration data, such as a threshold or a list of values, rather than as a different design? If yes, make the mechanism core and the values local.
- Record each decision, its reason and its owner, so the next region to join does not reopen it from scratch.
What does a core managed solution plus regional extension solutions look like?
The architecture turns the core-versus-local decision into something the platform enforces. The core is built once, released as a managed solution, and installed in every production environment. Each region has an extension solution that depends on the core and adds what that region needs. Regions add; they do not edit core components. The mechanics of solutions, source control, managed deployment and pipelines are covered in our Power Platform ALM guide and are not repeated here.
| Layer | What it contains | Owned by |
|---|---|---|
| Core managed solution | Customer master tables and columns, global choices for stages and reporting dimensions, core business process flows, shared plugins, the approval mechanism, security role templates, core reports and dashboards. | Central design authority and core delivery team. |
| Shared integration solution | Connections and flows to group systems, with environment variables and connection references for values that differ per environment. | Central integration owner. |
| Regional extension solution | Region-specific columns and tables (for example local identifiers or consent records), additional forms and views, local process steps, local flows and region-specific apps. | Regional product owner, reviewed by the design authority. |
| Configuration data | Approval thresholds, approvers, local choice mappings, price lists and currencies, loaded per environment as data rather than built as customisation. | Region, within limits set by the core. |
Which design rules keep regional extensions from breaking the core?
Layering works when a few rules are applied without exception. Most of the damage in multi-region estates comes from one region editing a core component because it was quicker, which creates a layer the next core release has to fight.
- Add, do not modify: regional solutions add forms, views, columns and flows rather than editing core forms or core logic. If a region needs a different form, it gets its own form, assigned by security role.
- Shared choice lists stay core: stages and reporting dimensions are global choices owned by the core; regions do not add values to them locally.
- Mechanism core, values local: approvals, validations and routing read thresholds and approvers from configuration data, so a new limit is a data change, not a solution change.
- Local identifiers in a related table: a registration identifier table with a type and country, rather than a new column on the account for every country.
- No unmanaged changes in production, in any region, including "just this once".
- Environment variables and connection references for anything that differs between environments, so the same solution installs everywhere.
- One publisher and prefix convention for the core and a documented convention for regional extensions, so ownership of every component is visible.
Should regions share one Dynamics 365 environment or have their own?
Both designs work, and the choice is driven more by data residency and operational independence than by preference. Some settings, such as the base currency and base language of an environment, are fixed when the environment is created, so this decision has to be made before the build rather than revisited later.
| Design | Fits when | Watch out for |
|---|---|---|
| One global production environment, regions as business units | Data can be held in one geography, customers are shared across regions, and group reporting in real time matters. | Security design carries the regional separation; release timing is shared, so one region cannot delay or rush a release alone. |
| One environment per region, same core solution | Data residency requires separate geographies, regions need independent release windows, or regional volumes and integrations differ greatly. | A shared customer exists in several environments, so a customer master synchronisation and a group reporting layer are needed. |
| Hybrid: shared environment for most regions, separate ones where law requires | Most regions can share, but one or two jurisdictions cannot. | Two operating models to support; keep the core solution identical in both. |
What governance stops regions from forking the template?
Governance has to make the legitimate route faster than the workaround. If a regional change request takes a quarter, regions will fork; if it takes a sprint and has a clear answer, most will not.
- A design authority with regional representation that owns the core and decides whether a request is core, regional or declined, against the written criteria.
- Tiered change requests: configuration data changes handled by the region, regional extension changes reviewed lightly, core changes planned into the core release.
- A regular core release cadence that regions can plan around, with release notes and a test window in each region.
- Regional product owners with real authority over their extension, so they own local adoption instead of blaming the centre.
- Deployment only through pipelines, with production access for makers removed, and Power Platform data policies and environment strategy applied consistently.
- A quarterly review of drift: unmanaged layers, unused regional components and requests that keep being declined, which usually signal a core decision to revisit.
How should a multi-region rollout be sequenced?
One pilot region, measurable outcomes, then waves. Picking the pilot is a judgement call: it should be representative enough to test the core, supportive enough to engage, and not the headquarters region, because the headquarters region is the one the template was already designed around.
| Phase | What happens | Evidence to move on |
|---|---|---|
| 1. Core versus local decisions | Workshops with every region; decision log signed off by the design authority. | Every contested element has an owner and a recorded decision. |
| 2. Core build and pilot region extension | Core solution and the first regional extension built together, so the extension pattern is tested from day one. | The pilot region extension installs on the core without modifying it. |
| 3. Pilot region live | Pilot users work in the system; local steps and configuration data adjusted during a defined period. | Agreed outcomes measured: core stages used as defined, customer master quality, group report produced from core dimensions, and local work no longer in side spreadsheets. |
| 4. Core adjustments | Lessons folded back into the core before anyone else goes live. | Core release published; decision log updated. |
| 5. Regional waves | Remaining regions onboard in waves, each with its own extension and data migration. | Each wave meets the same outcomes before the next starts. |
What if regions already run different CRMs or acquired companies bring their own?
Then standardisation starts with consolidation, and the order matters. Where an acquired company must meet a reporting deadline before its CRM can be retired, our guide to merging two CRMs after an acquisition covers the reporting layer, key conforming and retiring systems one at a time. The core-versus-local model on this page is what the surviving system should look like afterwards.
We recommend the right solution - whether that's Microsoft Dynamics 365, Power Platform, or a custom-built CRM. Some businesses need the Microsoft ecosystem. Others need full control without licensing. We deliver both. Some regions or subsidiaries do not fit Microsoft licensing at all, for example large numbers of light users or hosting rules that require their own infrastructure. For those, a custom CRM on React, Node.js, PostgreSQL or .NET can implement the same core customer master, stages and reporting dimensions and feed group reporting, so the standard is preserved even where the platform differs.
Should every region run on Dynamics 365, or do some need a custom CRM?
Can afford licensing and want the Microsoft ecosystem
Dynamics 365
Microsoft 365, Teams and Outlook integration, a mature partner ecosystem, Copilot, and apps for sales, service and field operations that are configured rather than built.
Need full control and zero licensing
Custom CRM
A CRM built on React, Node.js, PostgreSQL or .NET that you own outright: your data model, your hosting, no per-user subscription, and features shaped exactly to your process.
Not sure which fits
We help you decide
A short discovery weighs licensing budget, process complexity, integrations and long-term ownership, then recommends one path. We deliver both, so the recommendation has no reason to lean.
What part of a multi-region programme does Solzet deliver?
Senior consultants and full-stack developers with 8+ years of Dynamics 365 Customer Engagement and Power Platform delivery run the core-versus-local workshops, design the core and regional solution architecture and environment strategy, build the core and regional extensions, set up ALM and pipelines, and support regional waves including data migration. Delivery is remote from Yerevan, Armenia, working with regions in the UK, Germany, the Netherlands, the US and elsewhere through Teams, with any onsite workshops arranged case by case. See our Dynamics 365 consulting service for how engagements run.
- Covered: Dynamics 365 Sales, Customer Service, Field Service and Customer Insights, Dataverse, Power Apps, Power Automate, Power Pages, integrations and PCF controls.
- Not covered: Dynamics 365 Finance, Business Central, Finance and Operations or any ERP rollout; we integrate with the ERP your group runs.
- Not provided: tax or legal advice; country requirements come from your legal, tax and compliance teams, and we implement them.
- If an earlier regional rollout has stalled, a health check of the existing estate is usually the first step.
What do people ask us?
How do you standardise Dynamics 365 across regions with different processes?
Standardise the customer master, opportunity stages and reporting dimensions, and keep tax and legal identifiers, consent and retention, partner structures, language, currency, approval thresholds and local process steps regional. Implement that split as a core managed solution with regional extension solutions that add but do not edit core components, govern it through a design authority and a single release path, and pilot in one region before rolling out.
What is a Dynamics 365 global template?
A standard design that every region or subsidiary uses as its starting point: the shared data model, process stages, reporting dimensions, security role patterns and integrations, packaged as a managed solution. A useful template also defines how regions extend it, because a template with no legitimate route for local differences tends to be bypassed rather than adopted.
How do you handle different tax identifiers and regulations per country in one CRM?
Use a core pattern with local values: for example a related registration identifier table with a type and country instead of new account columns per country, and a core consent model whose purposes, wording and retention rules are configured per jurisdiction. Your legal, tax and compliance teams define the requirements; the architecture keeps them from fragmenting the data model. Solzet does not give tax or legal advice.
Should each region have its own Dynamics 365 environment?
Only if something requires it. A single global environment with regions as business units suits shared customers and real-time group reporting. Separate environments suit data residency requirements, independent release windows or very different regional volumes, at the cost of synchronising the customer master and building group reporting. Base currency and base language are fixed when an environment is created, so decide early.
How do we model different partner and distributor structures across countries?
Keep the account and contact master core, and model partner relationships as a regional concern: relationship records or connection roles describing distributor, reseller or agent roles, with regional forms, views and processes on top. Group reporting then reads the core account and the core reporting dimensions, whatever channel structure a region uses.
How do we stop a region changing the core solution directly in production?
Deploy only managed solutions through pipelines, remove maker and customiser access in production, give regions their own extension solution and a fast review route for it, and run a regular drift review for unmanaged layers. Most forking happens because the legitimate route is slower than the workaround, so the release cadence matters as much as the permissions.
Which region should pilot a global Dynamics 365 template?
One that is representative enough to test the core, engaged enough to give honest feedback, and not the headquarters region the template was designed around. Agree measurable outcomes before go live, such as core stages used as defined, customer master quality and a group report produced from core dimensions, and fold the lessons into the core before the next wave.
Can a subsidiary that cannot use Microsoft licensing still follow the group standard?
Yes, if the standard is defined as data and process rather than as a product. A custom CRM can implement the same customer master, stages and reporting dimensions and feed group reporting. Solzet builds these on React, Node.js, PostgreSQL and .NET, and we recommend whichever platform fits each region.
Where should you go next?
Power Platform ALM with PAC CLI
Managed solutions, source control and pipelines that make a core and regional layering model deployable.
Dynamics 365 consulting and development
Senior consultants and developers for design, build and rollout on Dynamics 365 Customer Engagement.
Merging two CRMs after an acquisition
Reporting layer first, then key conforming, matching and retiring one system at a time.
Custom CRM Development
CRM on React, Node.js, PostgreSQL and .NET for organizations that need full control without Microsoft licensing.
Dynamics 365 health check
An evidence-based assessment of an existing estate before a standardisation programme builds on it.
Which solution is right for your business?
Tell us what you need. A senior consultant replies within one business day with a recommendation - Dynamics 365, Power Platform, or a custom-built CRM - not a sales script.