Last week a client asked me how the device connectivity rate is calculated.
Their factory has more than 300 pieces of equipment. In the fixed asset ledger, every one has its original value, depreciation life, and workshop location—not a single field missing. But when asked which ones could collect data, the equipment engineer flipped through the records for a long time and said about forty-something. For the rest, he said, “Whether they even have a network port isn't certain—you'd have to open them up to check.”
This question used to be a bonus item. In the future, it may become a required question.
Of the eight indicators, only one falls in the workshop.
At 3 p.m. on November 17, the Guangdong Provincial Government portal posted the "Digital Guangdong Construction '15th Five-Year Plan'," with the document number Yue Fu [2026] No. 85, and the date of formulation written as October 31. The main text of the plan is a 17-page PDF attachment; Box 1 lists eight main indicators for the "15th Five-Year Plan" period:
The value-added of core industries of the digital economy as a proportion of GDP, from 16.5% to 18% (16.5% marked as 2024 data); the number of valid invention patents in core industries of the digital economy from 484,000 to 650,000; valid datasets open for public data resources from 103,000 to 130,000; operating revenue of internet and related service industry enterprises above designated size from 798.32 billion yuan to 1.1 trillion yuan; the share of revenue from new cultural business forms from 40.3% to 48%; the number of 5G (including 5G-A) base stations per 10,000 people from 34.7 to 51.
For the remaining two items, the 2025 column is all “—”:
One is the penetration rate of digital production equipment in industrial enterprises above designated size, 72% by 2030.
The other is the penetration rate of applications such as next-generation intelligent terminals and intelligent agents, which will exceed 90% by 2030.
The “—” symbol deserves a closer look. It does not mean zero; it means that there was previously no provincial-level statistical basis, and baseline figures cannot be produced. Among the eight items, only these two start from blank, and these two happen to be what enterprises need to fill in, not what the government needs to build.
In the “Indicator Attribute” column, all eight items are written as “expected.” Expected indicators are not included in performance assessments, so the government will not be held accountable. But expected indicators will go into statistical reports; when the reports are issued, it is enterprises that fill them out.
The denominator is in the fixed asset ledger, and the numerator is in the gateway list
How this indicator is calculated is not explained in the main text of the plan. The PDF contains only a special column table, with no indicator definition, no numerator and denominator definitions, and no statistical frequency. As of November 18, neither the portal of the Provincial Government Services and Data Administration nor the portal of the Provincial Department of Industry and Information Technology has a separate indicator explanation document. Therefore, the following is inferred from the conventional interpretation of the three characters “普及率”, not the original text of the document; this is stated first.
The denominator is the total number of production equipment. In most enterprises, this number is in the fixed asset ledger and can be exported from the financial system.
The numerator is the number of digital production devices. What counts as “digital”? There are at least three interpretations: equipped with sensors, capable of connecting to the internet and uploading data, and data that has passed through a system. Calculated under these three definitions, the results can differ by a factor of two.
In the project, we tried to match these two sides up, and the sticking point was not technology but coding. The equipment ledger was built by finance, with one asset number per piece of equipment, ordered by purchase batch; workstations were built by production, with one process number per workstation, ordered by process route. The two numbering systems each follow their own path, so the same piece of equipment looks completely different in the two tables.
There was a client making precision parts, with more than 400 machines; the two sides tried hard to reconcile them, and in the end they merged them on a composite key of 'asset number + operation number.' After three weeks of reconciliation, more than 20 still could not be matched—that batch had been replaced wholesale during a technical upgrade; Finance processed them as scrapped and newly added, while Production simply changed the workstation names.
In Odoo, these two sets of data are already in two tables: fixed assets are in account.asset, and equipment maintenance is in maintenance.equipment. After default installation, the two tables are not linked. To run this metric, you need to add four columns to the equipment card: control mode (manual, semi-automatic, fully automatic, CNC), communication protocol (whether Modbus, OPC UA, Profinet are available), data collection point (whether a collector is installed and what model it is), and data start date (from which date continuous upload records begin).
Only when all four columns are filled in are the numerator and denominator on the same table; only then is the penetration rate calculated, not estimated.
There is a pitfall to mention upfront. Some companies, seeking speed, directly count “buying equipment with a screen” as digital equipment. A screen is an HMI, not a data interface. If reviewers really check, what they pull is the gateway’s backend records, not the equipment manual.
In the same plan, there are also two more direct statements
Chapter Four discusses the digital-intelligent economy. The original text of Article (2) reads: “Deepen industrial digital transformation and intelligent upgrading, promote the ‘small orders, quick response’ flexible smart manufacturing model, and cultivate smart factories in a tiered manner. Build a national-level manufacturing digital transformation promotion center and cultivate digital-intelligent transformation service providers. Build industry-general knowledge and knowledge corpora in the industrial field, and create industrial large models and agents.”
The term “small orders, quick response” used to be common in the apparel, luggage, and home furnishings industries, and now it has been written into provincial-level plans. For ERP practitioners, it corresponds to a whole set of things—make-to-order production, small-batch line changeovers, batch traceability, and operation flow—not something that can be solved by adding a report; it is a configuration of the production model itself. In Odoo, the MTO route, outsourced operations, and lot/serial number traceability are all designed for this scenario.
Article (3) also has a sentence: “Carry out the ‘Hundred Chains Resilience Enhancement’ action and promote the use of secure and controllable key basic software, industrial software, and other such applications.”
The emphasis of this statement is on domestic substitution. However, the plan does not specify whether ERP counts as "key basic software" or "industrial software," nor does it provide a replacement list or timetable. The direction is clear, but the path is left blank.
It should be noted that none of these three sentences comes with supporting funds or an application channel. The real funding in this round still comes from those existing lines under the Provincial Department of Industry and Information Technology: technological upgrading, the digital transformation city pilot, and the Guangdong Province Manufacturing Empowerment Reward Special Fund Plan line (as of November 18, no officially issued version had yet been found). The plan is a five-year framework; the money is in the annual special programs. If you see a direction mentioned in the plan, do not assume it is a project you can apply for next year.
There is one more point worth noting. The plan proposes to "build a national-level manufacturing digital transformation promotion center." If this center comes to fruition, service providers will have an additional point of contact. However, as of November 18, no construction plan, application notice, or public announcement of the contractor had been found.
Three things you can do this week
1. First, pin down the denominator. Export the fixed asset ledger, select production equipment (excluding auxiliary equipment outside of office, transportation, and testing), group by workshop, and add two columns, "Control Mode" and "Communication Protocol". Up to 400 units can be completed in a week; for over a thousand units, batch it by workshop—don't expect to finish it all at once.
II. Pick one production line for a collection pilot and run it continuously for a full 30 days. The metric calls for penetration rate, but any inspection looks at whether there is a real data stream. Thirty days of upload records are more useful than an equipment list.
III. Prepare only one equipment list. Before December 31, the last batch of the Guangzhou city pilot (declared on November 22) needs detailed equipment investment information, and the calculation of the equipment purchase amount for the industrial orderly transfer fund also has the same deadline. The two have different scopes/definitions but share the same source of basic data; if they are prepared separately as two documents, they will definitely not match in the end.
The number 72%, five years from now, is a cell on a statistical report. Whether it can be filled in or not depends on whether, this year, you have started managing equipment as a data source rather than as an asset.
