Microsoft Cloud · Data and analytics
We bring your Dynamics 365 and Dataverse data into Microsoft Fabric and build the access model there before the first report is shared — because business units, security roles and record ownership are Dataverse concepts and do not travel into the lakehouse. Five building blocks, each with a result that stays with you.
What we deliver
Scope, order and duration are set after we have seen your environment, not before.
Link to Fabric as the path for new links, a recommendation on which workspace the link lands in, authentication via workspace identity rather than a user account, table selection and change tracking, initial synchronization in a low-load window. If an Azure Synapse Link path exists, retiring it belongs in the same plan.
Result. A running link in your workspace, plus a record of which path was chosen and why — traceable even when someone else asks later.
Microsoft Entra ID groups, workspace roles, OneLake security roles, row-level security on the gold layer, column restrictions in the semantic model rather than in a OneLake column rule, a fixed identity for Direct Lake models, the default read permission DefaultReader kept with a reason or switched off, diagnostics and monitoring logs enabled, the result documented.
Result. A documented permission model: which group sees what, on which layer it is enforced, and how you spot a deviation.
Silver transformations for choice fields, lookups, delete rows and record versions; gold tables per business domain; file sizes and clustering following the documented Microsoft guidance for each layer.
Result. Gold tables per business domain for your reports to build on — without every report author having to know the quirks of the raw Dataverse data.
An SKU proposal measured against your consumer count and your model sizes, pay-as-you-go versus reservation, a pause plan for non-production, reconciliation with your existing Power BI licensing — together with our License Advisory.
Result. An SKU proposal with the arithmetic behind it, which you can check against your own volumes.
Workspace model, separation of development and production, versioning via deployment pipelines, data quality rules with an escalation path, handover to Power Platform Governance and Application Care.
Result. An operations manual and a handover after which your team can run the path itself — or a solid basis for an Application Care contract.
What you provide
What we need from you is stated up front — then it is already answered in the initial conversation instead of shifting the schedule later.
An honest boundary
For Dataverse data, Microsoft Fabric is the strategic path. It is still not the right answer in every project. We do not recommend Fabric when:
If one of these points applies, we say so in the initial conversation. If a Fabric architecture still makes sense afterwards, we plan the build-up of the necessary skills in the customer team as part of it — for us that belongs to Application Care and Power Platform training.
Scope boundaries
We do not take on legal assessment. We build and document access, logging and encryption configurations. Whether your processing is permissible under data protection law is assessed by your organisation or your legal advisers — we supply the technical basis for it.
We do not make license procurement decisions for you. We calculate the options against your volumes and lay them open. Procurement and contractual commitment remain your decision.
We do not take over Microsoft's product roadmap. Several mechanisms on this path are in motion: low-latency sync is being rolled out in stages, and the Delta Lake output of Azure Synapse Link is being retired. We plan against the documented state and name what is in preview. Dates and scope are set by Microsoft.
We do not run your capacity on the side. Capacity monitoring, pause plans and reacting to throttling are ongoing operational tasks. We design the plan during the project; anyone who does not want to run it themselves needs an Application Care contract for it.
We do not build workarounds for documented product limits. Where a limit exists — 2,000 tables, 200 MiB per record, one workspace per environment, no file content from notes — we say so instead of placing a construction next to it that nobody maintains.
Authority over business definitions stays with you. What a won opportunity is and which status counts towards the rate is decided by your business side. We build the logic that matches that decision and write it down.
The data path in detail
Which path Microsoft still offers for new links today, what is replicated and what is not, how the medallion layers are cut for CRM data, why your Dataverse permission model ends at the boundary to the lakehouse, and which capacity limits shape the architecture — we wrote that up as of 6 November 2026 on the basis of the Microsoft documentation: Dynamics 365 and Dataverse data in Microsoft Fabric.
Licensing models, prices, discounts and break-even points are on our page about Power BI licensing, pricing and discounts. This page here describes what we take on.
Questions and answers
Five results, each one acceptable on its own: a running link together with the decision record, a documented access model, gold tables per business domain, an SKU proposal with verifiable arithmetic, and an operations manual with handover. The documentation stays with you and is not tied to us.
We only say that once we have seen the environment — a number beforehand would be guesswork. It is determined by the number of tables, the state of your data model, whether Microsoft Entra ID groups are already in place, whether an existing Azure Synapse Link path has to be replaced, and who runs the access model afterwards.
A Dataverse environment with change tracking on the desired tables, a Fabric workspace or the decision about the capacity, Microsoft Entra ID groups for the readership, read access in the agreed scope, a business contact for definitions, and a role that runs the access model after the handover.
Fabric is billed as capacity per capacity unit, pay-as-you-go or reserved, and can be obtained through an Azure subscription or through a Cloud Solution Provider — we are a CSP partner. The current Microsoft prices, the full SKU table and the break-even calculation are on our page about Power BI licensing, pricing and discounts. We calculate our own work after the environment review.
An eligible existing organisation keeps its path but has to switch to CSV output or to Link to Fabric by December 2027. That replacement belongs in building block 01 and not in a separate project afterwards. Important when changing environments: eligibility is tied to the Dataverse organisation ID, not to the tenant.
Often yes — and then we say so. If you need CRM reports and not a data platform, Power BI Pro with a direct Dataverse connection covers many analyses, with the advantage that users' permissions apply in Dataverse instead of being rebuilt in Fabric. The scope section on this page names the other cases in which we do not recommend Fabric.
Not on the side. Capacity monitoring, pause plans and reacting to throttling are ongoing tasks. We design the plan during the project; running it is available as an Application Care contract or stays with your team — for which we lay the basis in the handover.
30-min initial conversation or 45-min architecture conversation
Bring your table count, your consumer count and your permission model in Dataverse. In 30 minutes we work out whether Link to Fabric or a direct Power BI connection is the right path — and what you need for it in terms of capacity and licenses. For more complex situations, go straight to a 45-minute architecture conversation.
Accompanying services
Engineering projects rarely stand alone — license logic, architecture clarification, quality gates, knowledge transfer, and follow-on operations usually run in parallel. Here are the most common accompanying services we add to Discovery Spikes, sprint fixed-price engagements, or Application Care contracts.
Up front · architecture
Before implementation: tenant structure, data model, security concept, integration mapping. The result is an architecture document any engineering team can pick up — including one other than us.
Up front · CSP
Which license bundles for which users, which add-on SKUs are needed, where you are over- or under-licensed. If you wish, you can source the licenses through us as a Microsoft Cloud Solution Provider — one invoice, one point of contact.
During · quality gate
Independent second opinion during a running implementation project — whether we are delivering it or another partner. CMMI-based quality gates, risk reviews, fixed price per gate.
During · adoption
Not the classic two-day workshop that's forgotten after a week — but a dynamic learning program over 4–6 weeks with kickoff training, application phases, and advanced sessions. Training matrix for roles and topics.
After · operations
After go-live: a predictable Application Care contract with monthly flat rate, SLA-based. Includes releases, hotfixes, extensions, tenant hardening — and continuous support instead of merely reacting to tickets.
After · knowledge
When the original developers are gone, the previous partner is no longer reachable, or the documentation is outdated — reverse engineering of the existing solution with a documented result: code map, data model, customization inventory.
