An answer you can act on
A readiness assessment that names the actual bottleneck — model design, refresh window, source sprawl, or genuine volume — and states plainly whether a capacity would fix it.
Fabric architecture & implementation
A Microsoft Fabric Consultant for Power BI Teams
Lakehouse design, medallion pipelines, and Direct Lake semantic models built by a senior US-based engineer — starting with an honest assessment of whether Fabric is the right answer for you at all. Prepaid hours, no contracts.
Representative dashboard — sample data
Advice before architecture
The first question is whether you need it
Fabric is a genuinely capable platform and a genuinely expensive habit to acquire early. Plenty of teams arrive convinced they need a lakehouse when what they actually have is a slow semantic model and four overlapping dataflows. Fixing that costs a fraction of a capacity and solves the same complaint.
So we start with an assessment, not a sales pitch. If Fabric is the right answer — growing volumes, many sources, refresh windows that no longer fit, or several teams needing the same curated data — we design and build it. If it is not, we say so and fix what is actually slow.
What we do in Fabric
- Readiness assessment
- Lakehouse & medallion design
- Data Factory pipelines
- Direct Lake semantic models
- Capacity sizing
- Dataflow sprawl cleanup
Try us risk-free for 4 hours
Buy a block of prepaid Flex hours and, if you're unhappy within the first 4 hours, we refund you in full — no questions asked. Hours are deducted only for work you've reviewed and approved, there's no project minimum, and your banked hours never expire.
Why teams bring us in
Senior architecture without an enterprise engagement
You send the request; a senior engineer assesses, designs, and builds one domain end to end as a working reference.
Medallion architecture that holds
Bronze, silver, and gold layers with clear ownership, so you can trace any figure back to the raw source and reprocess without re-extracting from systems you may not control.
Direct Lake where it earns its place
Large models served from lakehouse tables without an import refresh window, applied where the data volume justifies it rather than everywhere because the feature exists.
Capacity sized honestly
Capacity is billed by Microsoft, not by us, so we have no reason to oversize it. We measure your workload, recommend a SKU, and revisit it once real usage data exists.
Handover, not dependency
Documented patterns your team extends, plus banked hours to keep a senior reviewer available. The goal is that you stop needing us, not that you keep paying a retainer.
What is Microsoft Fabric, in practical terms?
Fabric is a single SaaS analytics platform that puts data engineering, storage, and Power BI on one foundation. Storage is OneLake, a tenant-wide data lake in open Delta format. Compute is bought as a capacity rather than per workload, and Power BI sits on top as the reporting layer it always was.
Microsoft's own Microsoft Fabric overview (opens in new tab) is the best neutral starting point if you are still deciding. The practical shift for a Power BI team is that data preparation moves out of dataflows and into pipelines and notebooks writing to the lakehouse, and semantic models can read those tables directly instead of importing a copy. It is also where the ETL layer of a retired BI suite lands when its warehouse is being retired too, which is why our Pentaho to Power BI migration page hands Kettle re-platforming to this work instead of folding it into the report rebuild. That is a real architectural improvement at scale and unnecessary overhead below it.
When does Fabric make sense — and when does it not?
The dividing line is rarely company size. It is whether your current pain is caused by volume and source sprawl, or by design. Here is the assessment we run before recommending anything.
| What you are experiencing | Usual real cause | Right response |
|---|---|---|
| Reports are slow to open | Model design and DAX, not infrastructure | Fix the model — a capacity will not help |
| Refresh no longer fits the window | Full reloads of large fact tables | Incremental refresh first; Fabric if volume keeps growing |
| Dozens of overlapping dataflows | No shared curated layer | Lakehouse with a medallion structure |
| Several teams rebuilding the same tables | No governed single source | OneLake plus shared gold-layer tables |
| Models too large to import comfortably | Genuine data volume | Direct Lake semantic models on Fabric |
| One department wants better reporting | Scope, not scale | A well-built Power BI model on Pro licensing |
How does a Fabric engagement actually run?
In phases, and never as a single cutover. Assessment first, then a designed target architecture, then one data domain built end to end — ingestion, medallion layers, semantic model, and a report proving the numbers match today's. That reference becomes the pattern for everything that follows.
Building one domain properly matters more than covering all of them quickly. It surfaces the awkward questions early: how late-arriving data is handled, who owns the silver layer definitions, what happens when a source system changes shape. Where Direct Lake is the right fit, Microsoft's Direct Lake documentation (opens in new tab) describes the mode and its constraints, several of which affect model design and should be understood before you commit. Recent work in this shape includes a Fabric medallion architecture for a national healthcare nonprofit.
What does it cost?
Our work is prepaid Flex hours that never expire: Starter 20h at $125/hr ($2,500), Growth 40h at $115/hr ($4,600), or Power 80h at $100/hr ($8,000). An assessment and architecture design typically fits well inside a Starter block. Fabric capacity itself is billed separately by Microsoft.
We deliberately do not quote capacity prices here, because SKU pricing, regions, and reservation options change and a stale number would be worse than none. What we will do is size the capacity against your measured workload and show you Microsoft's current pricing during the assessment. On our side there are no project minimums and no retainer — see how prepaid Flex hours work, or our Power BI consulting page for advisory work that stops short of a platform change.
Who is this for?
Data and analytics leads at organisations already invested in Power BI who are weighing Fabric, mid-market teams whose refresh windows have run out, and nonprofits or healthcare organisations consolidating fragmented reporting. If the real problem is a slow or untrusted report estate, start with a Power BI health check or our troubleshooting work instead.
The ideal fit is a team with a real architecture decision to make and nobody senior enough in-house to make it confidently — and the willingness to hear that the answer might be no. If you would rather strengthen the model layer first, our Power BI data modeling and DAX page covers that work. And if the reason a capacity is on the table is Copilot, note that the capacity only makes the feature available — the answers still depend on the metadata in your models, which is what our Power BI Copilot consultant page covers. Contact is form-only — send your request and we'll reply with a scope estimate.
What a Fabric engagement covers
- Readiness & cost assessment
- Lakehouse & medallion design
- Pipelines & ingestion patterns
- Direct Lake semantic models
- Documented handover to your team
3–5
business days, typical
Prepaid
hours, never expire
You
own everything
Fabric capacity is billed by Microsoft; our hours are the only cost from us. See Flex pricing →
How it works
Assess, design, then build one domain properly
- 1
Assessment
We review your sources, models, refresh windows, and workspace estate, then tell you plainly whether Fabric solves your problem — and what it would cost in capacity as well as hours.
- 2
Architecture & first domain
A senior US-based engineer designs the lakehouse and medallion structure, builds one domain end to end, and proves the reported numbers match what you have today.
- 3
Handover & extend
You get documented patterns your team can repeat, and banked hours to keep a senior reviewer on call as you roll out the remaining domains at your own pace.
Representative perspective
We had approval to buy a capacity and were told to expect a six-week engagement. The assessment showed our problem was the model, not the platform, and it was fixed in a week for a fraction of the budget.Analytics lead, mid-market services company
Representative perspective
Eleven overlapping dataflows and three teams maintaining versions of the same table. Moving to a shared curated layer meant we finally argued about decisions instead of about whose number was right.Data manager, nonprofit organisation
Find out whether Fabric is your answer before you buy a capacity
Tell us your sources, your refresh pain, and what you are being asked to decide. We'll scope the assessment and tell you the hours before any work begins — drawn from prepaid Flex hours that never expire, with no project minimums and no obligation.
FAQ
Fabric questions, answered honestly
Often not, and we will say so. If your reporting runs on a handful of sources inside a well-built semantic model, Fabric adds cost and administration without solving a problem. It earns its place when data volume, source count, or refresh windows have outgrown what Power BI alone handles well.
Capacity is billed by Microsoft, not by us, and pricing depends on the SKU and region you choose plus whether you reserve or pay as you go. We size the capacity against your actual workload and point you at Microsoft's current pricing rather than quoting a number that may be out of date.
Yes. Fabric is the same platform family, so existing workspaces, reports, and semantic models continue to run. What changes is where the data is prepared and stored underneath. Migration is incremental, one data domain at a time, rather than a single cutover weekend.
Three layers in the lakehouse: bronze holds raw source data as it arrived, silver holds cleaned and conformed tables, gold holds the business-ready tables your reports use. The value is traceability, because you can always show where a number came from and reprocess without re-extracting.
Yes, and that is the common arrangement. We design the architecture, build the first domain end to end as a working reference, and hand over documented patterns your team extends. Prepaid hours make it easy to keep us available for review without a retainer.
Book a free demo
Get senior Power BI help, on demand
Tell us the task. We scope it, you approve, and a senior consultant delivers — typically in 3–5 business days. Prepaid hours that never expire.
- Prepaid hours from $100/hour — and they never expire
- Senior-led delivery — 20+ years on the Microsoft platform
- Typical 3–5 business day turnaround
- Full refund if you’re unhappy in your first 4 hours
The fastest way to reach us is the form — tell us the task and we'll reply within one business day.