Companies hold back from data sharing because they fear losing control. They do not always know who will use the data, for what purpose, for how long, or whether it will be passed on. Inside a company, this leads to disputes over access, quality, and responsibility. Between companies, the risk feels higher: loss of competitive advantage, leaks, use outside the agreed scope, or conflict with a partner.
More business decisions now depend on data from partners. A supplier needs a forecast. A service team needs machine data. A manufacturer needs quality information about a component. A customer asks about a batch, CO2 footprint, or delivery date.
The answer is not to send all data to everyone. A safer path is sovereign data exchange, where access is given under clear rules.
Data sovereignty means a company keeps control over its data even when it shares that data with a partner. In simple terms: you decide who can use the data, for what purpose, for how long, and under which conditions.
You should know:
This direction is visible in Manufacturing-X, a German government-funded initiative for industry. Its goal is to build an open data ecosystem for manufacturing. Factory-X is one of its lighthouse projects and focuses on sovereign data exchange in industry.
For manufacturing companies, data sharing between businesses will increasingly involve quality, traceability, service, claims, CO2 footprint, energy, maintenance, and supply planning.
Several barriers still block or limit data exchange because companies are afraid of losing control. Below are the main reasons companies do not want to share data, together with practical ways to reduce that fear.
Data can reveal more than a company wants to show:
That is why data access should be limited to the business purpose. A supplier does not need to see the full production plan. In many cases, a demand forecast by week, product family, and allowed change range is enough.
What to do:
Fear of losing data inside the company is not irrational. Data breaches can cost millions of dollars, and many incidents are caused not only by external attacks, but also by errors, misuse, and poor control over where data goes.
Blocking data does not remove the risk. It often moves the risk into places the company controls poorly:
A safer setup gives people controlled access, keeps an activity history, and defines how the data may be used.
Many companies see only two options: full access or no access. That is too narrow.
| Situation | Data that is usually enough | What does not need to be shown |
|---|---|---|
| Supplier claim | defect type, batch, date, photo | prices, margins, customer list |
| Remote machine service | alarms, cycles, technical parameters | production plan, operator data |
| Product CO2 footprint | result for the batch or product | recipe, energy cost |
| Risk of delay | order status, planned date | full plant schedule |
The rule is simple: share the smallest data scope that lets the partner solve the problem.
The hardest moment comes after the data has been shared. The company wants confidence that the partner will not use it outside the agreed purpose.
Every B2B data exchange should have a short rules card:
Without these rules, data sharing can feel like giving up control.
GDPR, trade secrets, NDAs, and the Data Act do not mean every data exchange must be blocked. They mean the company must define the purpose, scope, legal basis, safeguards, and responsibility.
The Data Act, Regulation (EU) 2023/2854, deals with fair access to and use of data. It matters, among other areas, for data from connected products and related services. That includes companies working with machines, industrial equipment, and data generated during operation. The rules generally apply from September 12, 2025.
Regulations do not require companies to hand over everything. They require order.
Sometimes a company does not share data because it does not trust its own data.
Common warning signs include:
Poor data quality is one of the main barriers to data sharing in manufacturing. If the data is incomplete or inconsistent, sharing it will expose the problem faster.
Good starting points in production include MES, OEE, CMMS, SPC, traceability, planning, and reporting. When machine data, spreadsheets, and business systems all tell different versions of the process, the first task is to organize the data in one controlled environment.
Data exchange stops when one side is expected to give data while the other side gets the benefit.
Good data sharing helps both companies:
If the benefit is not visible, data will be treated as risk without a return.
Start with one process and one partner, not a large program.
Plan:
A simple data risk split may look like this:
| Level | Examples | Rule |
|---|---|---|
| Low risk | status, trend, aggregation | can be shown more broadly |
| Medium risk | batches, failures, claims | access for selected roles |
| High risk | prices, customer data, know-how | only with a strong basis |
A company has a quality issue with a component. The supplier needs data, but the manufacturer does not want to reveal too much.
A reasonable scope includes:
Outside the scope:
Access rules:
The partner gets what is needed, while you keep control.
Get 5 chapters of the book for free!
Join the newsletter and gain access to 40% of the book
”15 Steps to Buying an Information System”
Choose one area where your team loses time on emails, spreadsheets, and explaining why reports do not match.
A good first test is small:
Companies do not have to choose between blocking data and losing control. They can share less data, describe it better, secure it properly, and connect it to a specific business purpose.
The most important step is to decide which data is needed, who needs it, why it is needed, and under which rules it will be shared.
Most often, they fear losing control. They worry about leaks, misuse, loss of competitive advantage, and unclear responsibility.
It is a form of data exchange where the company defines who may use the data, for what purpose, for how long, and under which conditions.
It can be safe if the company limits the data scope, assigns an owner, grants access only to selected people, and keeps a history of use.
Start with low-risk or medium-risk data: order statuses, quality indicators, claims, failure history, or batch data.
Classify data by risk, set time-limited access, prepare a rules card, keep an activity history, and avoid sending files when a view or report is enough.
