Indonesia's One Data Law Demands Interoperability: Lessons for Company System Integration
November 13, 2026 · 6 min read · Etzal Group Team
On Thursday, 5 November 2026, Indonesia's Minister of Communication and Digital Affairs, Meutya Hafid, presented the One Data Indonesia Law (Undang-Undang Satu Data Indonesia) at the office of the National Development Planning Ministry, Bappenas, in Central Jakarta. According to the ministry's press release, carried by the government news service InfoPublik on 6 November 2026, the law makes interoperability the main precondition, on the principle that data stays at its source.
The obligations described there are addressed to ministries and agencies, not private companies. But the requirements the government has set for itself are an unusually compact summary of what makes data integration work. A company whose ERP, CRM and warehouse systems do not talk to each other can use the Indonesia One Data law as a checklist.
What was said
The press release makes several points:
- Data exchange between agencies requires shared definitions, metadata rules, reference codes, security and access control, and traceability of data use.
- Data stays at its source: each ministry and agency remains responsible for the quality, updating and validity of its data. Integration happens through systems that communicate with each other, not by centralising every database.
- The Government Service Liaison System (Sistem Penghubung Layanan Pemerintah, or SPLP) is supported by the National Data Centre and the Government Intranet.
- Each agency must appoint a responsible officer and determine its role as a controller or processor of personal data.
Meutya Hafid is quoted as saying that common standards are very important, and that the law's success will be measured by real benefit to the public.
Our source is a single press release. It does not give the law's number or the date it was passed, so we do not state them. Implementing rules and technical detail on the SPLP are also not covered there.
Who is directly affected
If your company supplies systems to government agencies, or exchanges data with them routinely, this direction matters to you even though the duty sits with the agency. An agency that must meet requirements on definitions, metadata and traceability may reasonably ask the same of its partners. Confirm the requirements that apply with the agency you serve, because our source does not detail them.
For other companies, the value of this news lies in the lesson. What follows is our view.
The five requirements as a company checklist
1. Shared definitions
Most integration projects fail not on technology but because two systems use the same word for different things. "Customer" in the sales system may mean anyone who ever asked for a quote. In finance, a customer is a party with an invoice. "This month's sales" can mean orders received, goods shipped, or invoices issued.
Write definitions for the twenty terms that appear most often in management reports, and have unit heads sign them. It is dull work, and it decides the outcome more than anything else.
2. Metadata
Every data set needs a description: who owns it, which system it comes from, when it was last updated, and which units it uses. Without that, analysts spend their time guessing whether a weight column holds kilograms or tonnes.
3. Reference codes
Agree one list of codes for things used across systems: item codes, customer codes, branch codes, account codes and region codes. If the warehouse and finance use different item codes, somebody will maintain a mapping table in a personal spreadsheet, and that table will be wrong at the worst moment.
4. Security and access control
Integration widens the reach of data. Connections between systems should use dedicated service accounts with the smallest rights possible, not a borrowed administrator account. Decide which data may flow where, personal data above all.
5. Traceability
Record who or which system took data, when, and for what. When figures in two reports differ, this trail lets you find the cause in an hour and not a week.
Data stays at its source: worth copying, with conditions
Many companies instinctively copy all data into one place. The government's approach in this law is different: each owner keeps its data and answers for its quality, while systems request data from each other when needed.
| Data stays at source | Everything copied to one place | |
|---|---|---|
| Responsibility for quality | Clear, sits with the owner | Often blurred once data is copied |
| Freshness | Always current | Depends on the copy schedule |
| Load on source systems | Grows with every request | Light, only during copying |
| Analysis across data sets | Harder and slower | Easier |
| Dependence | Source systems must always be up | Reports still run when a source is down |
Neither wins outright. For operational data that must be right at this moment, such as stock and credit limits, asking the source directly is usually correct. For historical analysis, a central copy still makes sense. The part worth copying without conditions is the first half of the principle: every data set has one owner who answers for its quality, wherever the copies sit.
A responsible person with a name
The press release says each agency must appoint a responsible officer and determine its role as controller or processor of personal data. The translation for a company is simple: every main data set has one person's name next to it, and for personal data the company knows when it acts as controller and when as processor for someone else. That role has legal consequences, so ask your legal adviser for a view.
What you can do this week
- List the systems that hold customer, item and transaction data, with the owner of each data set.
- Pick the management report whose figures are argued over most, and trace the definition of each figure back to its source system.
- Compare item codes and customer codes between your two largest systems. Count how many do not match.
- Check the accounts used by system-to-system connections and their access rights.
- If you serve government agencies, ask your contact how they are preparing and what they will ask of partners.
The minister's measure of success also suits an integration project inside a company: not how many systems are connected, but the real benefit to the people who use them.
Etzal Group does integration work of this kind in its enterprise software practice. To discuss your systems, contact us.
Sources
- InfoPublik (Komdigi press release), 6 November 2026: One Data Indonesia, Meutya Hafid: the public need not submit data to the government repeatedly (in Indonesian)
Frequently asked questions
Does the One Data Indonesia Law bind private companies?
The obligations described in the ministry's press release are addressed to government agencies. For private companies this article treats them as design lessons, not legal duties.
What are the five requirements for data exchange?
Shared definitions, metadata rules, reference codes, security and access control, and traceability of data use.
What does the principle that data stays at its source mean?
Each agency stays responsible for the quality, updating and validity of its data. Integration happens through systems that talk to each other, not by centralising every database.
Related service
Website Development
Fast, SEO-ready websites that turn visitors into customers.
