Platform · Layer 1 of 3
Data LayerOne set of data.Every value with its source.
The Data Layer is designed for your existing systems, nothing is replaced. The result is one record per customer, part and order, and every value keeps its source. Nothing to structure up front, no data project.
The problem
Every system knows only its part. And every copy says something else.
The same customer has one name in the ERP and another in the CRM, the price list sits on the drive in five versions, and the last agreement is buried in someone's mailbox. AI on top of that can only guess.
What research and practice say
AI fails on the data first.
60%
Forecast
Through 2026, Gartner expects organisations to abandon 60% of AI projects that lack AI-ready data.
68%
68% of CEOs call integrated, enterprise-wide data architecture critical; 50% report disconnected, piecemeal technology.
50%
For German AI users, data preparation (50%) and integration (41%) are big cost items; licenses only 21% (Bitkom).
Our own analysis
In demi's analysis of 196 use cases
- 100%
196 of 196 join data from at least two systems, with a median of at least 4.
How it works
Six steps from scattered to connected.
Step 1 / 6
Your systems stay. Nothing is replaced.
The Data Layer is built for ERP, CRM, PLM, DMS, mailbox and drives. demi works with read-only rights and connects to every system; if an adapter is missing, we build it with you.
- items
- 2,290
- connections
- 2,539
- your systems
- 6
Example: fictional company · Hover over a dot to see what it is connected to.
- ERP
- CRM
- PLM / CAD
- DMS
- Mailbox
- Drives
Step 2 / 6
New stays new, with its source.
A new order, a new contact, the new version of a data sheet: what is new in the connected sources enters the Data Layer with its source.
- items
- 2,290
- connections
- 2,539
- your systems
- 6
Example: fictional company · Hover over a dot to see what it is connected to.
Nothing is replacedTechnical user, read-onlyNo migration- ERP
- CRM
- PLM / CAD
- DMS
- Mailbox
- Drives
Step 3 / 6
One record per thing.
Customer, part, order: what sits in several systems, demi merges into one record. Every value keeps its source, so you can check it.
- items
- 2,290
- connections
- 2,539
- your systems
- 6
Example: fictional company · Hover over a dot to see what it is connected to.
SyncLive- ERPOrder A-2024-118changed
- CRMStadtwerke Nord: new contactnew
- MailboxRequest about pump KP-500new
- DrivesData sheet KP-500, version 3changed
- ERP
- CRM
- PLM / CAD
- DMS
- Mailbox
- Drives
Step 4 / 6
Everything linked.
Customers, products, orders, plants and documents connect the way they do in your company. This is what it looks like for a fictional company.
- items
- 2,290
- connections
- 2,539
- your systems
- 6
Example: fictional company · Hover over a dot to see what it is connected to.
ERPStadtwerke NordCRMStadtwerke Nord GmbHMailboxSW NordCustomer
Stadtwerke Nord GmbH
- Customer no. 10442ERP
- Contact K. WeberCRM
- Last request KP-500Mailbox
1 record · 3 sources
- ERP
- CRM
- PLM / CAD
- DMS
- Mailbox
- Drives
Step 5 / 6
Open and standardised.
An open Data Layer with a standardised data model, on which you use AI and build your own software. Your data belongs to you: export at any time as JSON or through the API.
- items
- 2,290
- connections
- 2,539
- your systems
- 6
Example: fictional company · Hover over a dot to see what it is connected to.
- ERP
- CRM
- PLM / CAD
- DMS
- Mailbox
- Drives
Step 6 / 6
Writing back only after approval.
demi writes back only after approval. Every write is recorded.
- items
- 2,290
- connections
- 2,539
- your systems
- 6
Example: fictional company · Hover over a dot to see what it is connected to.
Customer
Stadtwerke Nord GmbH
- Customer no. 10442ERP
- Contact K. WeberCRM
- Last request KP-500Mailbox
1 record · 3 sources
Standardised data modelopen{"type": "Customer","name": "Stadtwerke Nord GmbH","customer_no": "10442","contact": "K. Weber","last_request": "KP-500","sources": ["ERP", "CRM", "Mailbox"]}ExportExportcustomer-10442.jsonAPIYour data belongs to you: export at any time as JSON or through the API.
- items
- 2,290
- connections
- 2,539
- your systems
- 6
Example: fictional company · Hover over a dot to see what it is connected to.
- ERP
- CRM
- PLM / CAD
- DMS
- Mailbox
- Drives
DemiDocs: the open data standard
Everything that comes into demi becomes DemiDocs: one model for everything, readable as Markdown and JSON.
Standardised
Every source in the same model
Normalised
Numbers, dates, units in one form
Harmonised
One record per customer, part and order
De-noised
Only what the document really says
Everything goes into any format: JSON, Markdown, PDF, DOCX, XLSX, PPTX, CSV, HTML and any other.
Your data stays yours: export at any time as JSON or through the API.
More about the data standardTwo worlds come together.
demi reads the world outside with the knowledge in your data.
The plate shows kinds of company data.
What you get
One set of data for the whole company.
≈ 1 day
until the first case runs
With read-only rights on your systems. You add more systems step by step, and demi writes back only after your sign-off.
0
systems replaced
demi sits on top of your data without replacing anything. No migration, no new system to maintain.
1
record per customer, part and order
Every value keeps its source, so you can check it.
≈ 1 day: from real starts.
One case
Customer orders from e-mail as checked order drafts
Today
Order entry retypes purchase orders from PDFs, and every customer uses a different layout. Differences from the quote in price, quantity or date only surface in production or at invoicing.
In the Data Layer
Data sources in this case: the order inbox, ERP and CRM. demi replaces nothing and links every order to customer, articles, prices and quote. Customer part numbers are mapped to your own article numbers.
What it brings
- No more retyping of purchase orders
- Price and quantity differences caught at entry
- Faster order confirmations
The three layers
Data first. Then rights. Then AI. For each use case.
- Layer 3Intelligence LayerAI, traceable
- Layer 2Permission LayerOrg chart as rights
- You are hereLayer 1Data LayerYour data, connectedYou are here
FAQ
What IT and the business want to know
Do our systems stay, and does demi write into them?
Your systems stay. demi writes back only after sign-off. The data layer is designed for ERP, CRM, documents and email. Nothing is replaced, and you can go back at any time.
What does our IT need?
Not much: for each system demi should read, one technical user with read rights, write rights only where demi should write back after sign-off, and single sign-on via your identity provider (OpenID Connect). For each task the AI sees only what the person may see. Your data belongs to you: export at any time as JSON or through the API.
How fast does the first case run with demi?
The first case runs in about a day, read-only on your systems; nothing is replaced. You add more systems step by step, and demi writes back only after your sign-off. If you want help at the start, our services cover it.
Who owns the data in the Data Layer?
Your data belongs to you: export at any time as JSON or through the API, and for at least 30 days after the contract ends. What applies to model training on your data is governed by the DPA (§ 3 para. 3).
On your case
See your data as one set of data.
In 20 minutes we show you the Data Layer on a case from your industry, such as an order from your inbox: linked to customer, item and price, every value with its source.
