A Fabric pilot that ends with something you keep
AI initiatives stall without clean, unified data, and most organizations discover it after they’ve already bought the AI. Winmill puts Fabric to work on your own data, at a fixed price, with documents you keep at every step.
What Fabric and Foundry actually are
Microsoft Fabric brings your organization’s data into one platform: pipelines, warehousing, real-time analytics, and Power BI, all working from shared storage in OneLake. Microsoft Foundry is where AI applications get built on top of that data. If your teams already live in Microsoft 365, Teams, and Azure, this is the shortest path from scattered data to answers.
And if what you actually want is AI that answers questions from your own documents, that’s a RAG implementation, and it’s the most common thing we build on this stack. Not a chat bubble: a research tool, where every answer cites the document it came from and your users can open it.
Four steps, each one ends with documents you keep
You shouldn’t have to fund a transformation program to find out whether Fabric fits. Every step below produces artifacts you can put in front of your leadership, and each one qualifies the next.
Fabric Fit Check
Free. Three questions. Under two minutes.
Tell us what you’re trying to improve, what Microsoft tools you run today, and what your data environment looks like. You get an immediate read on the screen, a short written fit summary by email, and the option to book a 30-minute architecture walkthrough.
Fabric Pilot
Fixed price. Dates set at kickoff.
One source. One data product. You keep the environment either way. In scope: a Fabric workspace stood up in your own Azure tenancy, one real data source connected end to end, and one governed data product your business users can open on day one: a working semantic model and a Power BI report on your real data.
What you keep: the working environment in your tenancy, the report, a written target architecture, a written capacity and cost model including how your existing Azure credits apply, and a fixed price proposal for the full build. Out of scope, stated plainly: multiple source systems, historical migration, production hardening, and go-live.
Not a data platform, but a pile of documents nobody has time to read? That is a RAG implementation, and it runs on the same four steps with its own scope. See AI on Your Data.
Fabric Foundation Build
Scoped from the pilot. Eight to twelve weeks typical.
The full data estate: multiple sources, ingestion pipelines, medallion architecture, access model, semantic layer, and a report suite. You receive the production environment, its documentation, knowledge transfer for your team, and a load and performance report. Fixed price where scope allows; otherwise estimated with a stated not to exceed ceiling.
Managed and Optimize
Monthly retainer.
Capacity monitoring and cost optimization, pipeline reliability, and new data products on a cadence. Every month you get a written report covering capacity utilization, cost against budget, pipeline health, and what changed. You always know what you’re paying for.
Holding unspent Azure credit?
A surprising number of Fabric conversations start with credits that are quietly expiring. We architect so your existing Azure credit actually applies, and we know which models are and are not available in Foundry before you commit to a design that assumes otherwise.
Quick Microsoft Fabric fit check
Answer three questions about your environment and you’ll get an immediate read on the screen, then a short written fit summary by email. If Fabric looks like a fit, you can book a 30-minute architecture walkthrough with Eddie on the spot.
Rather talk first? Book a 30-minute architecture walkthrough with Eddie Hudson, who leads our Emerging Technology practice.
Common questions
Will Fabric replace our existing warehouse or data lake?
Not on day one, and it doesn’t have to. OneLake connects to existing sources through shortcuts, so you can adopt Fabric incrementally and move workloads when moving them earns its place. The pilot’s target architecture document maps exactly this.
Do we need to move all our data into OneLake?
No. Shortcuts let Fabric read data where it already lives. The capacity and cost model you receive in the pilot covers both patterns so you can compare them with real numbers.
How does this fit with Teams and Microsoft 365?
Reports and data products publish into the tools your teams already use, including Power BI in Teams. If your organization runs on Microsoft 365 and Azure today, you’re already most of the way there.
What does the pilot cost?
The pilot is a fixed price engagement, and we tell you the number before any commitment. Book the 30-minute architecture walkthrough and you’ll have the number in the first conversation.
What documents do we actually receive?
The fit check produces a short written fit summary. The pilot produces four documents: a target architecture, a capacity and cost model, the Power BI report itself, and a fixed price proposal for the full build. The foundation build adds full documentation and a load and performance report. Everything is written to be forwarded inside your organization.
Put Winmill’s AI and Azure team to work
Book a 30-minute architecture walkthrough with Eddie Hudson to talk through your data environment and leave with a clear next step.