In early April, Guangdong issued two documents in succession in the field of data elements: one was the Several Measures for Guangdong Province to Deepen the Market-Oriented Reform of Data Element Allocation and Cultivate and Strengthen the Data Industry, jointly issued by 17 departments including the Provincial Government Services and Data Administration (April 3); the other was the Guangdong Province Administrative Measures for Authorized Operation of Public Data Resources (Yue Zheng Shu [2026] No. 9, April 7).
When many corporate managers see the terms 'data elements' and 'authorized operations,' their first reaction is that this is a matter for the government and big platforms and has little to do with a factory like mine with a few hundred people.
Don't scroll away just yet. Among the several measures, there is one sentence that I think every business owner who intends to run their business seriously should read:
Promote enterprises to implement digital transformation and establish a data resource management system covering R&D, production, operations, services, management, and industrial chain collaboration.
This sentence was written into a provincial policy document. It means two things: first, the establishment of a data resource management system by enterprises has changed from "whether to do it" to "must do it"; second, the door to data assets is opening to enterprises.
First, the big picture: data is starting to have property rights and can be traded
The overall framework of the document is to address the three fundamental issues of data as a factor of production:
Rights confirmation. Establish a data property rights registration system in accordance with national arrangements; registration certificates can serve as "trusted credentials" for data circulation, transactions, and financing guarantees. In plain terms, your data will have a "property ownership certificate" from now on.
Supply. Public data will be kept in a single province-wide "ledger" and governed under the "one datum, one source, one standard" approach; through authorized operation, operating institutions will be allowed to develop it in compliance. In the future, enterprises can obtain public data in compliance from the data exchanges in Guangzhou and Shenzhen and use it in business decision-making.
Application. The “Data Elements ×” initiative continues to be carried out, and carriers such as the “Bay Area Data Super” dataset innovation competition and the Government Intelligent Joint Innovation Workshop are all promoting data applications in real-world scenarios.
In June, Guangdong introduced the country's first "token economy" policy, pushing data supply in the AI era one step further. The direction of this package of measures is clear: data is no longer just something backed up on hard drives by IT departments; it is an asset that can be registered, circulated, used for financing, and recognized on the balance sheet.
Now on the enterprise side: for data to become an asset, it must first pass three hurdles
The policy has paved the way for "data assetization," but whether the data in enterprises' hands can get on board depends on how well they pass the three hurdles.
The first hurdle: whether it is in the system. For most SMEs we have seen, customer information lives on salespeople's personal phones and WeChat, BOMs in Excel files on engineers' computers, inventory relies on a whiteboard in the warehouse, and cost accounting relies on a last-minute push by finance at month-end. If these data have not entered the business system, they cannot be called "resources," let alone "assets"—they do not even qualify to be counted and have their ownership confirmed.
The second hurdle: whether standards are in place. In several measures, 'one data item, one source, one standard' appears repeatedly. The same customer is called Company A in the sales system, A Co., Ltd. in the financial ledger, and A Factory on warehouse delivery notes; even if such data are all entered into the system, they cannot be aggregated and analyzed, let alone have their value assessed. Master data governance—ensuring that core data such as customers, materials, suppliers, and BOM have a single source and unified standards—is the first foundation stone of data assets.
The third hurdle: can it be kept under control? Covering R&D, production, operations, services, management, and industrial chain collaboration—this sentence in the policy actually outlines a complete business data map. Data must be governed throughout the entire process from collection and storage to use and authorization, so that in the future, whether for data registration, data trading, or responding to regulation, it can hold up.
A practical judgment: ERP is the skeleton of an enterprise data resource management system
How can these three hurdles be overcome? They cannot be solved by specifically "building a data middle platform"; the correct approach for small and medium-sized enterprises is to start with the business systems themselves.
Take Odoo, which we are familiar with, as an example: customer and supplier master data are created and maintained in a unified file, and sales orders, purchase orders, inventory transactions, production work orders, and financial vouchers are naturally interrelated within the same system; each data field has a unique entry point and validation rules. The longer the system runs, the richer the enterprise's data assets become—whether later conducting business analysis, applying to include data assets on the balance sheet, or participating in “Data Elements ×” scenario applications, the foundation is already in place.
Conversely, if the business systems are fragmented—one for sales, one for production, and another for finance, with data aligned through manual transfer—then the debt of data governance will only keep piling up.
Suggested Action Sequence
Here are three practical suggestions:
- Start with an inventory. Spend one to two weeks getting a clear picture of the current state of the company's core business data: what is in systems, what is in Excel, what is in individuals' hands, and how the quality is. This list itself is the foundation for all subsequent actions.
- Establish the rules again. Set the rule that core master data (customers, materials, suppliers) has a single source of truth, and no longer allow data with "the same name but different codes" to continue multiplying.
- Then implement systems (or refactor existing systems). Make core business processes truly online, with data standards built into the processes rather than relying on manual standardization afterward.
The policy dividends from data elements will ultimately fall only to those enterprises that have a clear picture of their data assets. This is not a difficult problem, but it must be done early—for an asset like data, the earlier you accumulate it, the longer the compounding effect.
If you want to know whether your company's data foundation is sufficient to meet the threshold for 'assetization', please feel free to contact us for a quick assessment.
