Early access

All your data, one hub, on infrastructure you control

Data Hub pulls data from your business systems, files and sensors, cleans it on the way in, and makes it available to the people and tools that need it: colleagues, partners, reporting tools and AI. Everything stays with you, and nothing is shared unless you decided it may be.

  • Runs with you, works without internet
  • You decide what may be shared, and with whom
  • Every read of sensitive data is logged

Sources

  • Finance, CRM and business systems
  • Business Central and other APIs
  • Files: Excel, CSV, exports
  • Sensors and streaming data
Data Hub
  • Clean: correct, filter, mask
  • Classify: open → sensitive
  • Version: change without disruption

one store per department · everything that happens is logged

Recipients

  • Other systems via API
  • Power BI, Excel, reporting tools
  • Files and email on a schedule
  • AI assistants, within your limits

Data Hub · from source to recipient

How it fits together

Data Hub runs where your data is

Data Hub is the data layer. Run it on the Stackship platform, where it installs from the catalogue in a few clicks, or on a server you already have.

Data Hub: the data layer

Collects, cleans and shares. Keeps track of the sources, the rules, the versions and who read what.

Stackship: the platform underneath

Runs Data Hub as an app with database, sign-in and operations taken care of, with the same security, isolation and upgrades as everything else on the platform.

From source to shared data in five steps

The menu in the product follows the same path. You collect at the top and share at the bottom.

  1. 1

    01 · Connect

    Connect a system, a file location or a sensor stream. Choose which tables to fetch and how often. Data Hub can also keep history, so you see how a value changed even when the source system does not remember.

  2. 2

    02 · Clean

    Fix formats, remove test rows and duplicates, mask what should not be seen. You see the result of every step before anything is saved, and what you see is exactly what you get.

  3. 3

    03 · Combine

    Combine data from several systems into one view, for example customer, invoices and support cases together. Click and drag, no code. The view automatically inherits the strictest confidentiality level of anything in it.

  4. 4

    04 · Share

    Make the view available to other systems and tools with one click. Every change becomes a new version, so those already fetching data keep working while they switch at their own pace.

  5. 5

    05 · Deliver

    Send data onward on a schedule: to another system, as a file to a partner, or as an email attachment. Or let an AI assistant answer questions about the data, limited to what you have opened to it.

A look inside

See Data Hub work

From the first connection to data shared under control.

01 · Connect

Every source in one list, with the state of each

Business systems, file locations and sensor streams side by side. Each source shows when it was last fetched, how much arrived and whether anything failed. Logins to your systems are stored once and reused, so a password change is made in one place.

  • Finance systems, CRM, Business Central and Dynamics, databases, APIs, file servers and sensor streams
  • Files in Excel, CSV and other formats, from network drives, SharePoint or cloud storage
  • Every row can be traced back to the file or system it came from
Sources · connected systems and file imports
02 · Clean

What you see in the preview is exactly what gets saved

Cleaning is built as a chain of steps, and after each step you see what the data looks like. Amounts written as text become numbers, dates in different formats become dates, and a value that cannot be read is left empty rather than stopping everything. No surprises between what you approved and what landed.

  • Correct, filter, remove duplicates, calculate new columns, mask
  • Compare and summarise without writing a line of code
  • Nightly runs only touch what changed
Pipeline · step chain with live preview
03 · Combine

Combine data from every system without moving anything

Because all data lands in the same store, one view can connect the finance system to the CRM to a sensor stream. Click in the tables, drag the connections, tick the columns. A view that mixes confidentiality levels automatically gets the strictest one, so nobody can lower protection by mixing in something harmless.

  • Click and drag, the result shows immediately
  • A view can never change or overload the store, only read
  • Every shared view carries a description so the recipient knows what they get
View builder · customers × invoices × invoice lines
04 · Share

Change without anyone fetching data being affected

When a view is shared it is locked as a version. If you need to change it you publish a new version, and the old one keeps answering exactly as before. Data Hub works out whether the change affects recipients, and the old version is phased out with three months' notice that recipients get automatically.

  • Heavy views can be precomputed so answers come instantly, whatever the data volume
  • Refreshed on a schedule, manually, or automatically when the underlying data changed
  • Can also be opened directly in Power BI and other reporting tools
Endpoints · v2 stable, v1 being phased out
05 · Trust

Know when the data is wrong, not only when the fetch is

Set rules for what the data must contain: that a field is always filled in, that customer numbers are unique, that amounts are within reasonable limits, that identity numbers have the right format, that a source actually delivered today. The rules are checked after every fetch and alert even when the fetch itself looked successful.

  • Tolerance per rule, so a source that always misses a fraction does not alert every night
  • A rule can stop further processing, so broken data never moves on as if all were well
  • History per rule: three weeks of nightly alerts is a different signal than one from yesterday
Data quality · rules with status and history
06 · Govern

The confidentiality level decides what data may become

Every source is labelled with a level, from open to sensitive. For each level you decide whether it may be shared at all, what sign-in is required, whether reads are logged and whether an AI assistant may read it. Tighten a level afterwards and Data Hub shows exactly which shares are now too weakly protected.

  • Four levels: open, internal, confidential, sensitive
  • A log of every read of sensitive data, denied attempts included, with AI reads labelled
  • GDPR erasure: search for a person, delete or anonymise, and keep a receipt with the case number
Classification · levels and what each one allows

Everything between the source and the recipient

Collect, keep history, clean, share, deliver and let AI in, in one product.

Connect what you already have

Finance and business systems, Business Central and Dynamics, databases, APIs, file servers, SharePoint and sensor streams. A new source in minutes, without an integration project.

History the source system never had

Let Data Hub keep a picture of the data at every fetch, and a value that changes in the source becomes a timeline with you. See how stock levels, prices or the case queue developed over time.

Clean on the way in

Fix formats, filter, remove duplicates and mask in steps whose result you see immediately. What you approve is what gets saved.

Share under control

Make a view available to other systems with one click. Versions let you change without affecting recipients, and access keys show exactly who fetches what.

Deliver where the data should go

Send data on a schedule to another system, as an Excel or CSV file to a partner, or as an email attachment. Everything, or only what is new since last time.

AI on your terms

Connect an AI assistant that can answer questions about your data. What it may see is set per confidentiality level, and every AI read of sensitive data lands in the log.

Built for organisations that answer for their data

Six things Data Hub does not compromise on.

Your infrastructure

Data Hub runs with you, on Stackship or on your own server. No cloud account, no data sent home to us, no internet required. The help is built into the product.

The confidentiality level governs everything

Whether data may be shared, what sign-in is required, whether reads are logged and whether AI may read it follows from the level. A view always gets the strictest level of anything it contains.

Erasure you can show

Search for a person by identity number, email or customer number across the whole store, delete or anonymise, and keep a receipt with case number, who did it and what was removed. The receipt cannot be deleted.

Three logs, three questions

Did last night's fetch go through? Who changed this setting, and what was it before? Who read the salary table last month? Each question has a log that answers it.

Secure from the ground up

Passwords to your systems are stored encrypted and never shown again. Views that users build can only read, never change or slow down the store. Permissions are checked on every request.

No lock-in

The store is an ordinary database that Power BI, Excel and other tools can read directly. Settings can be exported and moved, and you can take your data and leave at any time.

Made for regulated environments

Classification, logs and erasure receipts are part of the product, not an add-on.

GDPRNIS2On-premises / air-gapped
Pricing

One platform fee, then per workspace

Priced per year, excluding VAT. Nothing per source, user, row or request.

Data Hub platform

190 000 kr/year

One workspace included, with unlimited sources, users and shares.

Additional workspace

90 000 kr/year

Its own separate store with its own members: a department, a subsidiary, or a test environment.

Five workspaces

460 000 kr/year

Platform plus four additional workspaces, with one of them on the house.

  • Support as for every Stackship service: Basic included, Standard 8 000 kr/month, Enterprise quoted separately
  • On Stackship you install it yourself from the catalogue in a few clicks
  • Without Stackship, installation is added as a one-off 40 000 kr

Questions

The things data owners and IT ask first.

With you, in a database you control. Each workspace gets its own separate store, so departments or subsidiaries are kept apart and can be backed up, moved or deleted individually.

Early access

Get Data Hub early access

Data Hub is available now as early access: a full installation with you, with the team close by while you connect your first sources.

Who it's for

  • Public-sector bodies and regulated industries with data they must account for
  • Organisations with data spread across business systems, file servers and databases
  • Teams that want to share data with colleagues, partners and AI without losing control of it
  • Organisations already running Stackship, or planning to

What you get

  • Data Hub on your own infrastructure or your Stackship platform
  • Help connecting your first sources and sharing your first views
  • A direct line to the team building it, so your feedback shapes the roadmap
  • Early-access pricing, per workspace and per year

Data Hub is in active development. Early-access scope and onboarding are agreed per organisation.

SEE IT LIVE

Request early access

Tell us what you’d like to explore and leave your email, and we’ll be in touch about early access.

No commitment, just a conversation and a walkthrough.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.