Direct Answer
FEXLINK keeps emphasizing that it is "a company that produces data" — how is this different from vendors that sell sensors and meters? The difference lies at the end of the value chain — a sensor vendor delivers raw electrical signals (voltage, current, temperature), while a "data-producing" company takes the electrical signals through cleaning, object modeling and scenario analysis and turns them into decision-ready data with location, confidence and handling recommendations; the former sells equipment, the latter sells conclusions. From an electrical signal to decision-ready data, four steps lie in between: collection, modeling, analysis and recommendation — with every step, the data moves closer to "decidable".
A scene: with the same meter — a sensor vendor says "I gave you voltage and current data" — FEXLINK says "this transformer is at a 78% load factor, there are 2 risk points in the next two weeks, and we recommend checking the B-phase connection first" — from the same electrical signal, the latter became decidable data.
1. The Difference between a Sensor Vendor and a "Data-Producing" Company
The end point of a sensor vendor — delivering raw electrical signals: voltage, current and temperature — the equipment is sold and the responsibility ends; The end point of a "data-producing" company — delivering decision-ready data: location, confidence and handling recommendations — only when the conclusion arrives does the value arrive; The essence of the difference — the end of the value chain differs: the former stops at "data exists", the latter walks to "conclusions are usable"; The four-step path — electrical signal → data → intelligence → value: one processing layer per step — value sits at the end; The customer experience — buying equipment shows "meter readings", buying data shows "what to do" — different experiences, different values.
In one sentence: the difference is at the end of the value chain — selling equipment delivers raw electrical signals, producing data delivers decidable conclusions — the former sells meters, the latter sells conclusions.
A difference example: with the same meters installed at a park — one vendor marked "data collected", while FEXLINK gave "busbar B-phase connection heating, likely to alarm in a week". The same electrical signal — stopping at different ends — "collected" and "understood" are two different deliveries.
2. Step One: The Electrical Signal — Raw Material
What the electrical signal looks like — voltage, current, temperature and leakage: continuously changing physical quantities — raw material; The limitation of the electrical signal — no object and no scenario: a string of values without knowing which device it belongs to — far from decisions; The collection of the signal — meters and sensors: turning physical quantities into numbers — collection is the first step; The key points of collection — accurate and complete: signals credible and coverage without gaps — material quality decides product quality; The positioning of the signal — raw material: valuable but unprocessed — like ore to metal.
In one sentence: the electrical signal = raw values of voltage/current/temperature + without object or scenario — collection is the first step, and material quality decides product quality.
A signal example: the current value of a feeder is "132.5 A" — but which device and what state — the raw signal does not say. The electrical signal is "raw material" — a string of values does not answer "what device, what to do" — processing starts from this step.
3. Step Two: Data — With Location and Object
The processing of data — cleaning and modeling: denoising, filling gaps and tagging devices — the signal becomes data; Location and object — every value bound to a device: which transformer, which feeder and which connection — the data has an owner; Structured fields — timestamp, device object and measurement type: clear fields — machine-parseable; What the data looks like — "#2 transformer A-phase temperature 78 ℃ @ 10:32": object, time and value complete — usable data; The value of data — from "a string of values" to "a record": queryable and traceable — data is the basis of decisions.
In one sentence: data = the electrical signal after cleaning and modeling + with location and object + structured fields — "which device in what state", data starts to answer.
A data example — the same 78 ℃ — the raw signal only says "78" — the data record is "#2 transformer A-phase connection temperature 78 ℃". From the electrical signal to data, what is added is "ownership" — device, location and time — with ownership, the value gains meaning.
4. Step Three: Intelligence — With Confidence and Analysis
The processing of intelligence — scenario analysis: algorithms analyze the data — judging states and recognizing trends; Confidence — every judgment carries a credibility level: 90% confidence vs 50% doubt — the decision knows how much to trust; What the analysis looks like — "A-phase connection heating, 92% confidence, rising trend": with judgment and degree — decidable; The value of intelligence — from "what the data is" to "what the data means": state recognition and risk prediction — intelligence gives conclusions; The boundary of intelligence — mark doubt when confidence is low: not forcing a conclusion — credibility is the premise of decisions.
In one sentence: intelligence = algorithms analyzing the scenario + judgments with confidence + trend recognition — "what the data means", intelligence starts to answer.
An intelligence example: a busbar temperature rose slowly for three days — the algorithm judged "poor contact, 90% probability, rising trend". From data to intelligence, what is added is "judgment" — normal or abnormal, how credible — with judgment, the data gains a viewpoint.
5. Step Four: Value — With Handling Recommendations
The processing of value — decision output: from analysis to recommendation — telling the user what to do; The handling recommendation — "recommend checking the B-phase connection first, within a week": clear action and deadline — directly executable; What the value looks like — "high hazard risk, clear handling recommendation, prioritized": decision-ready data — the conclusion arrives; The landing of value — operational action: work orders, inspections and maintenance arrangements — data-driven action; Verifying value — the user handles by the recommendation and the hazard is resolved: the conclusion effective — the value loop closes.
In one sentence: value = analysis plus handling recommendations + clear action and deadline + directly executable — "what to do", value gives the answer.
A value example — intelligence says "connection overheating risk is high" — value says "recommend checking the B-phase connection today; without handling, an alarm may occur in 3 days". From intelligence to value, what is added is "action" — recommendation, deadline and priority — with action, the conclusion lands.
6. Edge-Cloud: Why It Must Be Done This Way
The role of the edge — sensing and collection: the electrical signal becomes numbers at the source — the edge is the start of the data; The role of the edge layer — edge processing: nearby cleaning, analysis and fast response — the edge makes the data "immediately usable"; The role of the cloud — global analysis: analysis across devices and scenarios — the cloud makes the data "globally decidable"; The synergy of the three — edge-cloud integrated: collection, processing and analysis divided — missing any layer breaks a link; Why it must be integrated — selling only meters means only the edge: the signal never reaches conclusions — with the full edge-cloud chain, the electrical signal becomes decision-ready data.
In one sentence: edge-cloud = the edge collects the signal + the edge layer processes nearby + the cloud analyzes globally — selling only meters means only the edge, and the signal never reaches conclusions.
An edge-cloud example — the edge gateway analyzes and alarms quickly on site, the cloud analyzes trends across stations — impossible for a vendor selling only meters. Edge-cloud is the "production line" of data-producing — collection, processing and analysis each doing their job — conclusions only come out of the full chain.
7. How Acceptance Distinguishes "Bought Hardware" from "Bought Data"
The question of acceptance — ask three questions: does the data have an object? Does the judgment have confidence? Does the conclusion have a recommendation? — the three questions tell; Has an object — every piece of data bound to a device and location: not a string of ownerless values — traceable; Has confidence — every judgment marked with a credibility level: know how much to trust — referable; Has a recommendation — the conclusion gives action and deadline: not "for reference only" — executable; The yardstick of acceptance — all three "yes": what was bought is data — all three "no": what was bought is only hardware.
In one sentence: the three acceptance questions = does the data have an object + does the judgment have confidence + does the conclusion have a recommendation — all three yes, you bought data.
An acceptance example — a project's acceptance: data with device tags, judgments with confidence and conclusions with handling recommendations — the three questions passed. Acceptance should not only look at "how many meters installed" — the three questions ask about the delivery — data, judgment and recommendation all in place, that is buying data.
8. Back to FEXLINK: The Step from an Electrical Signal to Decision-Ready Data
The FEXLINK positioning and product family: the full edge-cloud chain processing electrical signals into decision-ready data — selling conclusions, not meters. The acceptance checklist: ① the signal layer — meters and sensors collecting accurately and completely: voltage, current, temperature and leakage → ② the data layer — cleaning and modeling, binding object and location: timestamp, device and measurement fields complete → ③ the intelligence layer — algorithm analysis with confidence marking: state recognition and trend prediction → ④ the value layer — handling recommendations and priorities: action and deadline clear → ⑤ the three acceptance questions — has an object, has confidence, has a recommendation — the four steps complete, the electrical signal becomes decision-ready data.
One summary: from an electrical signal to decision-ready data = collection (signal) + modeling (object) + analysis (confidence) + recommendation (action) — a sensor vendor stops at the first step, a data-producing company walks to the fourth step. The difference is at the end of the value chain — the former sells equipment, the latter sells conclusions — with the four steps complete, the data enters decisions.
One reminder: during selection, do not be carried away by the "smart hardware" label — ask three questions: does the data carry an object? Does the judgment carry confidence? Does the conclusion carry a recommendation? — with the three questions passed, you bought data; without them, you only bought a meter. "A company that produces data" speaks of the end of the value chain — the four steps from the signal to the conclusion — every step walked, that is truly producing data.
Scope and Limitations
- The conclusions in this article are intended for LV distribution and electrical-safety sites. Criteria are compiled from public codes and FEXLINK product documentation, and product parameters follow the latest selection handbook and product knowledge base.
- Quantitative criteria (protection zone, discharge and residual voltage, earth resistance, thresholds and grading) hold under the preconditions stated in this article. Where the site voltage level, earthing arrangement, environmental category or load profile differs, the figures must be recalculated.
- Specific product models and parameters are subject to the latest selection handbook and product knowledge base; this article does not constitute selection or procurement advice.
- Standard numbers cited here refer to the currently effective editions; when a standard is revised, the related quantitative conclusions must be re-checked.
- Deployment and implementation for specific projects must be carried out by qualified engineers based on site survey and design drawings.
Frequently Asked Questions
What Does "A Company That Produces Data" Actually Mean
FEXLINK keeps emphasizing that it is "a company that produces data" — how is this different from vendors that sell sensors and meters