Methodology

This page explains how we research and build the content on this site, so readers can judge how much weight to give it. It is not a legal document; it is a working description of our process.

Source hierarchy

When we describe a rule, a risk, or a market practice, we try to trace it to the most authoritative source available, in this order:

• Official regulators and law: FINMA, the Swiss National Bank, and federal legislation such as the Consumer Credit Act (KKG) and anti-money-laundering law (AMLA/GwG).
• Primary platform documents: terms and conditions, prospectuses, and other material published directly by a platform, read rather than paraphrased from marketing pages.
• Reputable secondary sources: established financial press and industry bodies, used mainly for context rather than as the basis of a factual claim.

If a claim cannot be traced to at least one of these, we either drop it or clearly label it as our own general observation.

How comparisons and checklists are built

Our platform checklist and any comparison tables are built from criteria that are checkable by any reader: whether a document is published, whether a licence or registration is claimed and where it can be verified, how a fee or default is defined in the small print. We do not score or rank named platforms against each other, and we do not accept payment to include or exclude a platform.

Illustrative numbers

Where we use a number to explain a concept, such as a default rate or a return, it is always an invented example, clearly flagged as illustrative (for example, “assume, for illustration, a 3% default rate”). We never present an illustrative figure as current market data, and we never attribute a specific figure to a named regulator or platform unless it is a fact we can cite.

Update cadence

Articles carry a “last checked” date in the source box near the end of the page. Sophie Martin reviews regulatory references and links on a rolling basis and updates this date whenever a page is re-verified, even if the wording does not change. Rules and platform practices can change faster than we can review every page, so a “last checked” date should be treated as a floor, not a guarantee of current accuracy.

How the risk screener works

The risk screener uses a simplified expected-loss model: it combines a few inputs a reader provides, such as an assumed default rate and an assumed recovery rate, into a rough illustrative loss figure. It is a teaching tool, not a forecast, and it does not use real platform performance data. All inputs are processed in the reader’s own browser and are never transmitted to us or stored; see our privacy policy for details.