The domain data model
The objects your company is made of (customer, order, project, employee, asset), named once, defined once, governed in one place. Not a star schema; not a warehouse design pattern. A model your operators recognize.
Service: AI-Ready Data Foundation
FOUR TO EIGHT WEEKS · FIXED SCOPE
Data modeling, ETL, pipelines, AI data readiness.
The data foundation your AI bets actually need.
A domain data model, cleaned canonical sources, and the pipelines that keep them current. The layer that makes Shugyo (and any other agent you run) honest.
Book a fit call
Download the data readiness checklist
Starts from EUR 15k. Four to eight weeks. Engagement closes when your dashboards stop disagreeing.
01
WHO THIS IS FOR
You have tried the AI thing, and the data was the reason it did not work.
- 01 Two dashboards in your company disagree about the same number, and both sides defend their version.
- 02 The CRM and the ERP cannot agree on what a customer is.
- 03 A pilot with Claude, ChatGPT, or an internal copilot stalled because the data was not trustworthy.
- 04 A reorg, an acquisition, or a new business line has left your data layer in a state nobody can describe.
If two of four ring true, the data layer is the blocker, and you are paying for it whether you have named it or not.
02
WHAT YOU GET
A foundation, not a report.
Cleaned canonical sources
For each object, the source of truth is named, reconciled, and documented. Where two systems disagree, the rule for which wins is written down, not folklore.
The pipelines
Data modeling, ETL, ingestion, and reconciliation pipelines that keep the canonical sources current. Built to standard tools; documented so your team can change them.
The runbook
A short operational document: what to do when the data drifts, when a new source appears, when the model needs an extension. Not a vendor lock-in; a way to keep the model honest after we leave.
03
HOW THE ENGAGEMENT RUNS
Six weeks, in three phases.
-
Weeks 1-2With your team
Model.
We work with operations, finance, and engineering to define the domain data model. Every object is named once. Every disagreement is surfaced and resolved with the owner in the room.
-
Weeks 3-5Our synthesis
Cleanse and pipe.
We build the ingestion, the reconciliation, and the pipelines that bring the canonical sources current. Existing tools where they fit; clean code where they do not.
-
Week 6With your team
Hand off.
Runbook delivered. Your team takes operational ownership; we stay on a small retainer for thirty days if you want, then off.
04
THE GRADUATION PATH
What sits on top of the foundation.
You leave with the model, the pipelines, and the runbook. They are yours. If you never work with us again, they are still load-bearing; your dashboards stop disagreeing, your reorg survives the data hangover, your next AI pilot has a foundation under it.
If we do continue, the foundation is what Shugyo runs on. The semantic layer that gives your company sight is only as honest as the data underneath it. With the foundation in place, the rest is a matter of weeks, not quarters.
05
THE MAP
Most AI pilots do not fail at the model. They fail at the data. The model was just where it became obvious.
EmpoweredHouse · AI-Ready Data Foundation
06
PRICING
Starts from EUR 15k.
What changes the number: the count of source systems (more sources, more reconciliation), the regulatory weight (data residency, PII handling), the speed (parallel teams vs. sequential pace).
Source systems
More sources, more reconciliation.
Regulatory weight
Data residency and PII handling.
Speed
Parallel teams vs. sequential pace.
We will quote the engagement in the fit call. If your data layer is fundamentally healthy and we cannot justify the spend, we will say so.
07
THE FIT CALL · 30 MIN
Find out whether the data layer is the right place to start.
A fit call is thirty minutes. We will ask three questions about your sources, your disagreements, and your owner. By the end, you will know whether this engagement is the one.