The Business Model as a Signal of Trust Cuadernos Lacre · Analysis · May 18, 2026 https://solo2.net/en/notebooks/articulos/the-business-model-as-a-trust-signal.html How a service is economically sustained determines what the user represents to that service. Analysis of models in circulation and the question that should be asked before adopting them. --- To be clear: A company that doesn't charge you, charges someone — and that someone wants something in return. Sometimes that something is you: your data, your attention, your contacts. Sometimes not. The question before adopting a professional service is not if it's free; it's where the money comes from for those who operate it. --- The sober question of the digital economy An adult conversation about digital services ends sooner or later with the same question: how does this company make money? The question is not offensive. Any service that exists must be funded somehow; if it's not clear how, it's worth taking a second look. The question is not whether the service makes money — it must, or it will disappear — but who pays for it and why. There is an intuition that the general reader often has formulated in the phrase “if the product is free, you are the product.” The phrase is famous, punchy, and, like most famous phrases, a simplification. There are free services whose model is not to sell you; there are paid services whose model also includes reselling you. But the intuition captures something important: when one of the parties — the user — is not the one who pays, you have to look at who does pay, and what they want in return. What “the product is you” really means The classic formula for the advertising model is as follows: The company offers a free service to millions of users. These users generate, by using the service, two valuable things: attention — which is sold to advertisers in the form of ad impressions — and data — behaviors, preferences, locations, relationships, history — which are used to segment who to show which ad to and for how much. The customer of the service, in a strict economic sense, is not the user. It is the advertiser. The user is the attention and the data being sold to the advertiser. This, in itself, is not illegitimate. It is what commercial television has done since the 1950s. The difference from television is one of magnitude: the TV station knew approximately what time slot its viewers watched; the contemporary digital platform knows with individual precision who looks at what, when, for how long, with whom, from where, in what likely mood. That precision changes the nature of the exchange: attention is still sold, but identity is sold as well. A brief mapping of models in circulation Without claiming to be exhaustive, it is worth listing the models that a professional encounters today in their daily practice. 1. Fixed periodic subscription. You pay every month or every year and gain access to the service; the operator's incentive is for the service to be satisfactory, or you will not renew. But "satisfactory" allows for two interpretations: that it provides real value or that it is difficult to leave. Netflix, for example, has deliberately bet on serialized content rather than movies. A movie ends; the viewer is free to compare it with any other platform. A series hooks you episode by episode: the incentive to renew is that the next chapter is on the other side of the payment. The shape of the product can push towards retention rather than satisfaction, without the subscription model changing. 2. Pay-per-use or transactional. You pay based on what you consume; the incentive is for each use to be valuable enough for you to return. 3. Freemium model. A free tier attracts users, a paid tier monetizes a fraction; transparent when both tiers have clear functions, ambiguous when the free tier is designed to be inconvenient enough to induce payment. 4. Advertising model. Free for the user, funded by advertisers; mixed alignment. 5. Data monetization model. The operator sells data segments about aggregated or not-so-aggregated behavior to third parties other than the user; frequent opacity. 6. Public funding. Rare in the tech industry, common in European broadcasting; peculiar incentives depending on the institutional regime of the operator. These models are not mutually exclusive. Most large providers combine two or three. Microsoft 365 is a subscription and, depending on the module, also exploits data for ad personalization on other properties. Google Workspace is a subscription for the business customer and exploits attention and data for its parallel free services. What counts is not the model declared on the front page, but the effective combination in practice. The alignment of incentives The reason why the model matters has a technical name: alignment of incentives. The operator's incentive determines where the product design pushes. 1. Operator charging a subscription. Has incentives for the customer to be satisfied — if not, they cancel; design tends to be optimized toward perceived real utility. 2. Operator monetizing attention. Has incentives for the customer to stay connected as long as possible; design tends to be optimized toward capturing attention, which is not the same as satisfaction, and is sometimes the opposite. 3. Operator monetizing data. Has incentives to collect as much data as possible; design tends to maximize collection, asking for broad permissions and keeping information that would not be necessary for the strict operation of the service. There is no necessary malice in any of the three. There are incentives. Incentives produce, over hundreds of design decisions, systematically different products, even if their declared purposes are identical. The question when the data belongs to a third party For the professional, the question of the model takes on an additional layer. Lawyers, doctors, psychologists, tax advisors, journalists dealing with sources do not process only their own data; they process the data of third parties who entrusted it to them. When a professional service decides to adopt a digital tool whose funding includes data exploitation, it is not deciding on its own data: it is deciding on data that someone else entrusted to it for a specific purpose. That specific purpose — handling your case, treating your illness, preparing your tax return — rarely includes “and funding the operation of the tool provider.” The GDPR reflects this requirement in its Article 6: any processing of personal data requires an explicit legal basis, and the chosen basis binds the processing to the declared purpose. Processing a client's data with a basis other than that which covers the entrusted purpose, even through an intermediary technology provider, opens a compliance problem that the data protection authority can examine. The price of the free service, in this case, is paid by the professional's client without knowing it. Cases in the recent short term In recent years, the industry has produced particularly clear examples of the shift in incentive when the funding model changes. Reddit, a community platform for fifteen years funded by light advertising and donations, decided in 2023 to close free access to its API after announcing its IPO. Third-party applications that the community used to access the service disappeared in July of that year. The IPO, completed in March 2024, required a business model defensible to shareholders; the defensible model included aggressive monetization of the data produced by the community for fifteen years. The community had been generating a financial asset that it only found out about when the company decided to charge for it. The case is illustrative not because Reddit did something necessarily wrong, but because it clearly exposes the pattern. A service sustains itself for years with a certain model and communicates a certain implicit “pact” with its users. When financial requirements change — venture capital exhaustion, IPO requirements, acquisition by another company — the implicit pact is rewritten without negotiation with the users. The question “how does this company make money today?” is then completed with “how will it have to earn it in three years, and what does that force it to do with my data?”. The institutional exception There are services whose funding does not rest on the standard commercial model. Public radio and television in many European countries — BBC, RTVE, RAI, ZDF — are funded by a license fee or public budget that detaches them from the advertising incentive. Their problems are different: political dependence, pressure from the government of the day. But the nature of the incentive is different. On the digital level, non-profit foundation projects (Mozilla for years, Wikipedia, Signal Foundation) operate on an analogous model: donations, subsidies, non-commercial exploitation of the user. The sustainability of these models is more fragile; the alignment of incentives, clearer. A formulation that should not be overlooked: a private company with a subscription model is, in terms of alignment, more similar to the institutional model than the advertising model, even though it is private. It charges the user to serve the user. When that link remains pure — the user's money and only the user's money — incentives reasonably coincide with the user's interest. For the professional reader Five questions that should be asked before adopting a digital service for professional data, especially if that data belongs not to the professional but to their clients, patients, or representatives: 1. From what source, exactly, does the operator's income come? One source, two, a mix? 2. If the income includes advertising or data monetization, what data is monetized, on what legal basis, and does the declared purpose cover the third-party data that the professional will put into the service? 3. What is the operator's financial situation three to five years from now? Is it in the venture capital phase, the profitability phase, near an IPO, in the process of acquisition? 4. If the operator's financial requirements changed, which parts of the pact with the user would be at stake? What would happen to the data already submitted? 5. Are there alternatives with a more aligned model — pure subscription, free software, self-hosting, European alternative — whose real cost, compared to the assessed risk, justifies the change? The question is not ideological: it is operational. A professional who entrusts third-party data to a service whose economy relies on the exploitation of that data makes a decision that is ethically difficult to explain to the client. The fact that the decision is common does not make it correct. The fact that the tool is popular does not solve the question about the legal basis. --- “If you don't understand the business, be suspicious” is a popular phrase that has been circulating the internet for a decade. As with most popular phrases, it has its part of truth and its part of simplification. The analytical version of the same idea is this: the business model of any service determines, with considerable reliability and over time, what that service will do with the data you entrust to it. It won't always be bad. It won't always be worse. But it will be different depending on the model. And a professional who chooses tools for data that is not their own must understand that “different” before choosing. Sources and further reading - Regulation (EU) 2016/679, Article 6 — legal bases for the processing of personal data. The declared purpose binds the processing and limits its further use. - Regulation (EU) 2016/679, Article 28 — processor. Regime applicable to technology providers that process data on behalf of the professional controller. - Zuboff, S. — The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power (PublicAffairs, 2019). Systematic analysis of the business model based on the capture and exploitation of behavior. - Reddit, Inc. — Form S-1 filed with the SEC on February 22, 2024, registering the March 2024 IPO. Public documentation of the model change. - Regulation (EU) 2022/1925 on Digital Markets (DMA) — commercial transparency obligations for designated digital gatekeepers, applicable since May 2023. --- Cuadernos Lacre · A publication of Menzuri Gestión S.L. · written by R.Eugenio · edited by the team of Solo2. https://solo2.net/en/notebooks/