Stand up a real data warehouse — in weeks, not months.
I build data warehouses for companies with a small data team or none at all. Fixed scope, fixed price, and you own everything when it's done — the warehouse, the pipeline code, the documentation — in your own cloud account, under your name.
Fixed-price projects
Scoped, flat-rate engagements with a known price and a known timeline. No open-ended hourly billing, no surprise invoices.
Data Architecture Review
A fixed-price assessment of what you actually have. I inventory your source systems, pipelines, models, and reporting layer, then tell you what's working, what's quietly costing you, and what to fix first.
What you get
- Written findings report covering sources, models, pipelines, reporting, and cost
- Prioritized roadmap with effort estimates for each fix
- A one-hour readout to walk through it with your team
How it works
- Kickoff call and read-only access to your systems
- Two to six weeks of review depending on tier, with a short check-in midway
- The full review fee is credited toward any build engagement started within 90 days
Essentials
From $3,500
Typically 2 weeks
Spreadsheets plus one or two systems, no real warehouse yet.
- Up to 3 source systems
- Up to 5 scheduled jobs or pipelines
Standard
From $6,500
Typically 3 weeks
An existing warehouse or BI layer that someone inherited.
- 4 to 8 source systems
- Up to 15 scheduled jobs or pipelines
Extended
From $12,000
Typically 4 to 6 weeks
Multiple warehouses, many pipelines, or several teams to interview.
- 9 or more source systems
- 15 or more scheduled jobs or pipelines
Not sure which tier fits? Tell me roughly how many systems and jobs you have and I'll place it during the free consultation. Early clients get design-partner pricing while the tooling is still being built out.
Book a ReviewWarehouse Foundation
For teams with a handful of sources and no warehouse.
A flat-rate build: source scan, dimensional model, generated schema and transformations, a starter reporting layer, and a spec in your own repo you can maintain. Sized for up to three source systems.
Starting at $9,500 Typically 2 to 3 weeks
Talk about a build →Pipeline Rescue
For the one fragile pipeline nobody else understands.
Stabilize, document, and add monitoring to a single critical pipeline so it stops being a single point of failure. One week, one pipeline, fixed price.
Starting at $2,500 Typically 1 week
Talk about a rescue →Add a Source System
For an existing warehouse that needs one more source.
Scan the new system, extend the spec and the model, build the pipeline into staging and the warehouse, and update the documentation. One source, fixed price.
Starting at $4,500 Typically 1 to 2 weeks
Talk about a new source →Built for companies without a data team
Mid-sized companies — typically under $500M in revenue — are the ones everybody else quotes six to eighteen months and six or seven figures for a data project. You've been putting off a real warehouse because every quote made the whole idea die on the spot. That's who this is for.
Small data team, or none
Data is owned by IT, finance, or whoever picked it up — alongside everything else on their plate.
Fixed scope, fixed price
You know the price and the timeline before we begin. No open-ended hourly billing.
Owned by you, permanently
Everything lives in your own cloud tenant under your name. Stop working with me and nothing stops running.
Two ways to work together
Fixed-scope engagements
Warehouse builds, source system integrations, ETL/ELT pipeline design, architecture review, and second opinions on warehouse proposals you've been quoted. I can own a project end to end or plug into an existing team. Currently taking two projects and able to start immediately.
Talk about an engagement →Tooling that shortens the build
I'm automating this end to end, with heavy use of agentic AI. Where it stands today: the source discovery scanner is finished and runs read-only on your own machine, and extraction into staging is generated from the spec. Modeling, reporting, and deployment automation are in progress and done by hand for now. You're told which is which — and early clients get design-partner pricing because of it.
Ask about design-partner pricing →How an engagement actually runs
-
1
Discovery
A read-only scanner runs on your own machine and documents what you actually have — source systems, tables, keys, current pain points.
-
2
Design
We design the warehouse together, and every decision is written into a single readable spec that lives in your own GitHub repo.
-
3
Build
The warehouse, pipelines, and documentation are all built from that spec — generated where the tooling is ready, by hand where it isn't yet.
-
4
Handoff
Everything is yours, in your own cloud tenant, permanently — the warehouse, the pipeline code, and the spec it was built from. Not a black box.
Services
Architecture Review
Improper data architecture causes chronic problems and quietly increases costs. A fixed-price review tells you exactly where you stand and what to fix first. See review pricing →
Warehouse Builds
Fixed-scope builds from a documented spec: dimensional model, pipelines, and a reporting layer, deployed in your own tenant. See the Warehouse Foundation package →
Source Integrations & Pipelines
New source systems added to an existing warehouse, and ETL/ELT pipeline design that turns messy source data into clean, documented, query-ready tables.
Part-Time Data Engineering
An extra set of hands for teams that have one: modernizations, new sources, pipeline work, or bridging a gap between hires.
Coaching & Training
Dimensional modeling, SQL, and architecture fundamentals for your team and the people who consume their data.
About
I'm Ryan Russell, founder and principal data architect at RFA Consulting, based in the Charlotte area. I've spent 20 years in IT and data engineering across utility, retail, and self-storage: building warehouses, integrating source systems, designing ETL/ELT pipelines, and running the data and IT teams that own them.
The stack I reach for most is SQL Server, Azure SQL, dbt, and Power BI, with Kimball dimensional modeling underneath. I've also worked in Snowflake, Databricks, Azure Data Factory, AWS, and Python where the job called for it.
Why this costs less than you think
Enterprise platforms like Microsoft Fabric start around $263/month before you've built anything, and climb into the thousands from there. A right-sized stack for most SMB workloads runs a fraction of that — the goal is a warehouse that fits your budget and your team, not the other way around.
The other half of the cost is labor. Writing every decision into one spec and generating the pipelines from it removes most of the hand-built hours that make traditional warehouse projects expensive. That saving is built into the fixed prices above, and it grows as more of the tooling comes online.
Let's talk about your data
Tell me a bit about what you're working with, and I'll follow up to schedule a free consultation.
Email sales@rfaconsulting.bizIf one of the fixed-price packages looks like a fit, mention it and roughly how many source systems you have. Currently taking two projects and able to start immediately.