Product passport from AI

From the outside, the product line appears disjointed. Marketplace. Industrial controllers. Engineer monitoring. Chat with translation. Video service. Device licenses. Typically, such a list indicates a company is trying to grab everything it can.

But if you look at the system's structure, something else emerges. Here, the person, the document, and the machine are citizens of the same state . The user has a role and a set of rights. The partition and file have an access level. The device has a token, a hardware ID, an owner, and a regular self-report. And all of this resides in a single database, under a single rights model, with a single action log.

They don't do that. A typical company assembles a zoo:

Four systems with four personality models can't answer a simple question: "Does this person have the right to view the archive of this machine?" — because the "person" lives in one, and the "machine" in another, and there's nothing to connect them with except an Excel export.

Here, this question is a single line of rights verification. And it is from here that a product emerges that doesn't yet exist.

While the platform is simply running, it quietly accumulates three things. None of them are features, none of them can be purchased, and none of them arise if the business is rented space.

Why is this more important than functions?

All three assets have the same economics: in the first month, they're worth nothing, but after three years, they're worth more than the platform itself . One hundred dialogues is a correspondence. One hundred thousand dialogues in fourteen languages is an objection map, which is used to formulate all commercial arguments. A year of telemetry is a report. Five years is knowledge of which node is failing at the 8400th hour, a knowledge that even the node manufacturer doesn't have.

And the most important feature: they can't be taken from someone else's platform . By working on a marketplace, you generate these assets—but for its owner. They receive a demand map, and you receive a sales report.

Now, the reason for this section. The above results in a product that is assembled almost entirely from prefabricated parts and that competitors don't have because they lack a unified personality model.

Each physical product gets a permanent page. The address is printed as a QR code on the case or nameplate. Point your phone at it and you'll be taken to the passport for that specific unit, in your language, with the content specifically for you.

One address - four different pages

The key point for which a unified driver's license model is needed is that there is only one QR code, but the content depends on who scanned it.

All of the above are existing mechanisms: a page with pricing and media, access levels, files with server rights verification, video, chat with automatic translation, device heartbeat, and auditing. The only new feature is linking the page to the serial number and QR code .

The "it will be convenient for the client" argument doesn't sell well. There are three real reasons, and they're all about money.

Three sources of income

• The service is no longer reactive. The service life threshold has been passed—the maintenance call is automatically cancelled, before a breakdown, not after.
• Spare parts are shipped directly. The "Order" button on the page for a specific part with a known configuration is a completed order from a reseller.
• The secondary market becomes yours. A product with a fully traceable history is worth more—and when resold, it brings the next owner directly to you.

And fourth, which isn't revenue but is more valuable: failure statistics for the entire fleet . Not based on warranty claims, but on actual operating hours. After two years of operation, it's clear which component is failing, at what time, and for which customers. This changes both the design and negotiations with the component supplier.

A section that's usually missing from presentations. If you were selling a product datasheet tomorrow, here's what you'd have to finish—and it's better to say it yourself than to hear it on the first technical call.

1. Serial Number Registry and QR Code Generation: Currently, the page is created manually. We need batch output: upload a list of serial numbers and receive pages and codes for printing. This is the core work; the rest is add-ons.
2. Ownership Transfer: The product is sold, donated, or retired. A mechanism for changing ownership is needed while preserving the history and without the previous owner losing access to it.
3. Working without internet access : The shop may not have a communications network. An offline copy of the manual and a delayed repair report submission are required.
4. Accepting payment for spare parts currently involves issuing payment details. For retail owners, a standard one-step payment is required.

To boil it all down to one thing: this isn't a website or a CMS. It's a layer where people, documents, and machines are described in the same way. The website is just one way to present it. A marketplace, a production portal, a database of engineering work, and a product data sheet aren't four products, but four perspectives on the same thing.

This leads to a practical rule for talking to a buyer: don't list features. Ask what they have that should have its own page—a product, a machine, an object, an employee, a contract, a product. Anything that can be named can be created. Everything created gets a language, rights, a history, and a communication channel.

And then those three assets start to accumulate—and with each passing month, the argument for leaving becomes weaker. Not because someone is holding on to them, but because there will be nothing left to carry them with.