The data layer: one version of the truth
The single place where your data is correct and can be asked questions
One number, not three versions in three spreadsheets
The question is not whether you have the data. It is how many different answers you get when you ask for the same number.
What you get
The same number everywhere
The report, the app and the automation read from the same place, so they cannot contradict each other.
History is not lost
It keeps what an order or a price looked like at the time, not only how it looks now.
New questions without a new project
Once the data is structured, a new question is a query rather than a rewrite.
Automations have something to stand on
Flows that make decisions need trustworthy data. Without it, you only automate the mistake faster.
How it works
1. The model
We establish which entities really exist in your business: customer, order, product — and what each one precisely means.
- PostgreSQL
2. Feeding it
Data flows in from existing systems, by API where possible and by import where not.
- n8n
- REST
3. The rules
Constraints live in the database, not in each application. What the business forbids, the data refuses.
- PostgreSQL
4. History
Changes are retained, with who and when. Nothing silently overwrites what came before.
- PostgreSQL
Where it works well
- Businesses with data in several systems that disagree
- Reports rebuilt by hand every month
- Automations that need historical context
- Situations where who changed what and when must be visible
Where it does not
- It does not repair wrong data — it only makes it consistently wrong, visibly
- A single well-kept source does not need a layer on top of it
- It is not a warehouse for large-scale analytics — it is the operational foundation
What we use
- PostgreSQL
- n8n
Frequently asked questions
- Does this mean replacing our current systems?
- No. The data layer sits beside them and reads from them. The systems stay where they are, with their usual users.
- Our data is messy. Do we start by cleaning it?
- Only as far as the first use case needs. Cleaning everything before doing anything else is the project that never finishes.
- Who has access to the data?
- You decide, by role. The database can be hosted on your side. Our access can be limited or revoked at any time.
- Why PostgreSQL and not something else?
- It is free, mature, and any developer can take it over. We do not tie you to a technology only we understand.
- Do we need this before any automation?
- Not always. A simple flow does not need it. It becomes necessary once automations start making decisions from the data.