Technical terms, clearly explained

The EU AI Act (Regulation (EU) 2024/1689) is the European Union’s comprehensive law on artificial intelligence. It regulates AI systems according to their risk, imposing the strictest obligations on high-risk uses and banning a narrow set of practices outright. It applies far beyond the EU’s borders wherever an AI system’s output is used in the Union.

Under the EU AI Act, an AI system is a machine-based system that operates with some autonomy and, from the input it receives, infers how to generate outputs such as predictions, content, recommendations, or decisions that can influence its environment. The emphasis on inference distinguishes genuine AI from ordinary deterministic software.

High-risk AI systems are those the EU AI Act considers capable of significantly harming health, safety, or fundamental rights — for example AI used in employment, education, essential services, or law enforcement. They are not banned, but must meet extensive requirements (risk management, data governance, documentation, human oversight, robustness) and pass a conformity assessment before reaching the market.

AI literacy under Article 4 of the EU AI Act is the obligation of providers and deployers to take measures that support the AI literacy of their staff and others acting on their behalf. Since the Digital Omnibus of July 2026, no specific level per person must be guaranteed; proportionate, role-based and evidenced measures are required.

A general-purpose AI model (GPAI) is an AI model — typically trained on very large amounts of data — that displays significant generality and can competently perform a wide range of tasks, regardless of how it is later put on the market or integrated into downstream systems. Large language models are the prime example. The EU AI Act sets dedicated obligations for GPAI providers, with stricter rules for models posing systemic risk.

Annex III is the part of the EU AI Act that lists the specific use cases considered high-risk. AI systems intended for the purposes it enumerates — in areas such as employment, education, essential services, law enforcement, and migration — are presumptively high-risk and subject to the Act’s full set of obligations. For most organisations, checking their AI against Annex III is the decisive step in risk classification.

Prohibited AI practices are the uses of AI that the EU AI Act bans outright under Article 5 because they pose an unacceptable risk to fundamental rights and safety — such as manipulative or deceptive techniques that distort behaviour, exploitation of vulnerabilities, social scoring, and certain uses of real-time remote biometric identification. No safeguard or documentation can make a prohibited practice lawful.

A conformity assessment is the procedure by which a high-risk AI system is checked against the EU AI Act’s requirements before it is placed on the market. For most high-risk systems it is a self-assessment by the provider against the requirements; for some it involves an independent notified body. A system that passes is documented in an EU declaration of conformity and may bear the CE marking.

Under the EU AI Act, the provider is whoever develops an AI system, or has it developed, and places it on the market or puts it into service under its own name. The deployer is whoever uses an AI system under its own authority in a professional capacity. The role determines the obligations, and a deployer can become a provider through its own changes.

The fundamental rights impact assessment (FRIA) is an obligation under Article 27 of the AI Act for certain deployers of high-risk AI: before first use, they assess how the system may affect the fundamental rights of the people concerned and notify the market surveillance authority of the results.

Article 50 of the AI Act requires providers and deployers of certain AI systems to disclose: people must learn that they are interacting with an AI, AI-generated content must be marked in machine-readable form, and deepfakes and certain AI texts on matters of public interest must be disclosed. The obligations apply regardless of risk class.

An AI regulatory sandbox is a controlled framework set up by an authority in which companies can develop, train and test innovative AI systems under supervision for a limited time. Under Article 57 of the AI Act, every EU member state must establish at least one such sandbox.

Post-market monitoring is the obligation of providers of high-risk AI to observe their systems’ performance systematically after placing them on the market, collecting and analysing data on it. Under Article 73 of the AI Act this includes reporting serious incidents to the market surveillance authority within short, staggered deadlines.

Human oversight requires high-risk AI systems to be built and operated so that people can effectively monitor them in use, interpret their outputs correctly, override them and stop them. Providers must enable oversight by design (Art. 14 AI Act); deployers must assign it to competent people (Art. 26).

AI governance is the organisational framework through which a company steers its use of AI: responsibilities, policies, an inventory of systems, risk classification, approval of new applications, monitoring and documentation. It brings the requirements of the AI Act, data protection, information security and supervision together in one process.

An AI inventory is the complete, maintained list of all AI systems a company develops, offers or uses — with purpose, user group, data processed, vendor, role under the AI Act, risk class and responsible person. It is the basis for every further obligation, from the AI Act through data protection to supervision.

Harmonised standards are European standards developed at the Commission’s request and published in the Official Journal. Building a high-risk system to them gives a presumption of conformity with the corresponding AI Act requirements (Art. 40). Their delay was one reason the Digital Omnibus postponed the high-risk obligations.

ISO/IEC 42001 is the international standard for artificial intelligence management systems, published in December 2023. It describes how an organisation systematically steers the use and development of AI and is certifiable. It is structured like other ISO management systems, such as ISO/IEC 27001 for information security.

A Data Protection Officer (DPO) is a designated person responsible for overseeing an organisation’s data protection compliance, advising on obligations, monitoring adherence to the GDPR, and acting as the contact point for individuals and the supervisory authority. The GDPR requires a DPO in defined cases — notably large-scale systematic monitoring or large-scale processing of special-category data.

Privacy by Design is the GDPR principle — framed as data protection by design and by default — that data protection must be built into systems and processes from the outset and as the default setting, rather than added afterward. It means embedding measures like data minimisation, pseudonymisation, and access control into the architecture, so that the most privacy-protective option is the one that operates automatically.

Data subject rights are the enforceable rights the GDPR grants individuals over their personal data: to be informed, to access their data, to have inaccuracies corrected, to have data erased in defined cases, to restrict or object to processing, and to data portability, plus protections around automated decision-making. Organisations must be able to honour these requests, usually within one month.

Consent under the GDPR is one of the six lawful bases for processing personal data. To be valid it must be freely given, specific, informed, and unambiguous, expressed through a clear affirmative action, and as easy to withdraw as to give. Pre-ticked boxes, bundled consent, and inactivity do not qualify, and the controller must be able to demonstrate that consent was obtained.

Special categories of personal data are types the GDPR treats as especially sensitive under Article 9 — data revealing racial or ethnic origin, political opinions, religious or philosophical beliefs, trade-union membership, and genetic, biometric, health, or sex-life and sexual-orientation data. Processing them is prohibited unless a specific Article 9 exception applies, such as explicit consent.

The GDPR (General Data Protection Regulation, Regulation (EU) 2016/679) is the European Union’s central data protection law. It governs how personal data is processed, grants individuals enforceable rights over their data, and binds any organisation that processes the data of people in the EU — wherever that organisation is based. Breaches can attract fines of up to 4% of global annual turnover.

The revFADP is Switzerland’s revised Federal Act on Data Protection, in force since September 2023. It modernised Swiss data protection law and aligned it closely with the GDPR — strengthening transparency, data-subject rights, and accountability — while keeping some distinctly Swiss features. It applies to private organisations and federal bodies processing personal data in or affecting Switzerland.

A Data Processing Agreement (DPA) is the contract required under the GDPR whenever a controller engages a processor to handle personal data on its behalf. It sets out the subject matter, duration, nature and purpose of the processing, the types of data and categories of individuals, and the processor’s obligations — including security, confidentiality, sub-processing, and assistance to the controller.

A Data Protection Impact Assessment (DPIA) is a structured process required under the GDPR before processing that is likely to result in a high risk to individuals’ rights and freedoms. It describes the processing, assesses its necessity and proportionality, identifies and evaluates the risks, and sets out measures to mitigate them — documenting that risk was considered before the processing began.

An adequacy decision is the formal finding that a third country offers a level of data protection equivalent to one's own. In the EU it is adopted by the European Commission under Article 45 GDPR, in Switzerland by the Federal Council under the revFADP. Where it exists, personal data may be transferred to that country without additional safeguards.

Standard contractual clauses (SCC) are contract templates issued by the European Commission that allow personal data to be transferred to countries without an adequacy decision. They bind the recipient to a European level of protection. Since the Schrems II judgment they must be combined with an assessment of the destination country's law; Switzerland recognises them with adaptations.

A record of processing activities is a structured overview of which personal data a company processes, for what purpose, about whom, for whom and for how long. It is in principle mandatory under Article 12 of the Swiss FADP and Article 30 GDPR; in Switzerland, companies with fewer than 250 employees are exempt if their processing carries no heightened risk.

Technical and organisational measures (TOMs) are the safeguards a company uses to ensure the security of personal data: confidentiality, availability, integrity and traceability. The Swiss FADP (Art. 8) and the GDPR (Art. 32) require measures appropriate to the risk and leave the concrete design open.

Under the Swiss FADP, a data security breach must be reported to the FDPIC as soon as possible if it is likely to result in a high risk for the persons concerned (Art. 24 FADP). Under the GDPR, the supervisory authority must be notified within 72 hours unless a risk is unlikely, and data subjects must be informed of high-risk breaches.

Pseudonymised data can only be attributed to a person with additional information kept separately; for the controller it remains personal data. Anonymised data can no longer be attributed to a person with reasonable effort and falls outside data protection law. The difference decides which obligations apply.

The Data Privacy Framework lets US companies self-certify with the US Department of Commerce and thereby receive personal data from the EU and Switzerland without additional safeguards. The EU has recognised it since July 2023, Switzerland the Swiss-US DPF since 15 September 2024 — in each case only for certified companies.

The privacy notice fulfils the duty to inform data subjects: who processes personal data, for what purpose, to whom it goes and whether it goes abroad. The Swiss FADP sets a minimum content in Art. 19; the GDPR requires considerably more in Arts. 13 and 14, such as legal bases, retention periods and data subjects’ rights.

An automated individual decision is a decision based solely on automated processing that has a legal effect on the data subject or significantly affects them, such as a fully automated credit rejection. The Swiss FADP requires information and, on request, human review (Art. 21); the GDPR governs it in Art. 22.

A transfer impact assessment is the documented check of whether safeguards such as standard contractual clauses actually work for a transfer to a third country without an adequacy decision, in particular in view of government access rights in the recipient state. It goes back to the Court of Justice’s Schrems II judgment of 2020.

The BöB is the Swiss federal act on public procurement. It governs how federal bodies award contracts: which procedures are permitted, which principles apply, and what legal remedies suppliers have. Cantonal and municipal awards are governed not by the BöB but by the IVöB.

The IVöB is the Swiss intercantonal agreement on public procurement. It governs awards by cantons, municipalities and bodies attached to them. Each canton accedes individually, so the state of implementation is not the same everywhere.

simap.ch is the joint information system for public procurement in Switzerland. Federal, cantonal and municipal bodies publish their tenders and awards there. For any bid, the awarding body’s official documents always remain authoritative.

Thresholds are contract values above which a particular award procedure is prescribed. Above the treaty thresholds from the WTO Government Procurement Agreement, the strictest rules and the widest market access apply. The values are revised periodically.

In the open procedure a contract is publicly tendered and any interested company may submit a bid directly. It is the most accessible procedure type and is generally required above the applicable thresholds.

In the direct award procedure a public body awards a contract directly, without a tender. This is permitted below certain thresholds and in exhaustively regulated exceptional cases — for instance where only one supplier can deliver the service.

Suitability criteria test whether a supplier is fundamentally capable of performing the contract — through references, qualifications, and financial and technical capability. They are checked before bid assessment and are a yes-no decision.

Award criteria compare the admitted bids against one another and determine who wins the contract. Typical criteria are price, quality, the experience of key personnel and the proposed approach — each with a weighting announced in advance.

The specification is the part of the tender documents describing the service required. It sets out what is to be delivered, to what quality and under what conditions — and is therefore the basis for costing and the bid.

A procurement appeal is the legal remedy against decisions of an awarding body, such as an exclusion or an award. Deadlines are short and stated in the appeal instructions; an appeal does not automatically have suspensive effect.

A family office is an organisation that runs the wealth and administrative affairs of one or several wealthy families. A single family office serves one family, a multi family office serves several. The core of the work is operations and governance, not only investment.

The ultimate beneficial owner, or UBO, is the natural person to whom an asset, company or structure is ultimately attributable, or who exercises control over it. The relevant shareholding thresholds differ considerably from one jurisdiction to another.

KYC stands for know your customer and denotes the duty to establish a client’s identity, determine the beneficial owner, and understand the background of the business relationship — at onboarding and on an ongoing basis thereafter.

The Swiss Anti-Money Laundering Act, the GwG, obliges financial intermediaries to due diligence: identifying the contracting party, determining the beneficial owner, making special enquiries where risk is elevated, documenting all enquiries traceably, and reporting on reasonable suspicion.

A politically exposed person is someone who holds or has held a prominent public function, together with their close family members and close associates. The status is not an allegation but triggers enhanced due diligence.

Look-through consolidation aggregates assets through the ownership layers rather than showing only directly held positions. With intermediate companies, that is the difference between a correct figure and a merely plausible one.

Tax-lot accounting tracks purchase lots of a position individually and assigns specific lots to a sale. The realized gain follows from that assignment. Common methods are FIFO, LIFO, HIFO and the average method.

A periodic review is the recurring check of an existing business relationship: are the details still current, has the structure changed, is the risk rating still appropriate. The rhythm follows the risk.

The TJPG is the Swiss Federal Act on the Transparency of Legal Entities and the Identification of Beneficial Owners. In force since 1 October 2026, it requires Swiss companies and foreign companies with a Swiss nexus to identify their beneficial owners and report them to a non-public federal register.

A transparency register is a state register of the beneficial owners of companies and similar entities. It shows authorities and obliged entities under anti-money laundering law who actually controls a structure. Germany, Austria and Liechtenstein have run such registers for years; Switzerland has done so since 1 October 2026.

A controlling person (Kontrollinhaber) is the natural person who ultimately controls an operating company: primarily through at least 25 percent of capital or votes, otherwise by other means, and as a fallback the managing director. The term comes from Swiss anti-money laundering practice and is recorded on Form K.

Form A is the form commonly used in Switzerland on which the contracting party of a bank or financial intermediary declares in writing who is the beneficial owner of the assets deposited. It originates from the Swiss banks’ due diligence agreement (CDB/VSB) and has siblings for controlling persons, trusts, foundations and insurance wrappers.

Under Swiss anti-money laundering law, a domiciliary company (Sitzgesellschaft) is a legal entity, foundation, trust or similar arrangement that conducts no trading, manufacturing or other commercially run business. For such a company the beneficial owner of the assets is established, typically on Form A, rather than the controlling person.

KYB, Know Your Business, is due diligence on a company as a counterparty: its existence, legal form, seat, governing bodies, authorised signatories, business activity and, above all, the natural persons who control it. KYB complements KYC, which concerns natural persons, and only ends at the beneficial owners.

Sanctions screening is the systematic comparison of people, companies and transactions against sanctions lists — in Switzerland primarily the lists maintained by SECO under the Embargo Act. Its purpose is to detect relationships with sanctioned persons before assets move, and to decide matches in a traceable way.

Adverse media screening is a targeted search for negative media reports about a person or company, for example on criminal proceedings, corruption, fraud or sanctions evasion. It adds risks that appear on no list to sanctions and PEP checks, and requires an assessment of source, currency and relevance.

Source of wealth describes the origin of a person’s total wealth: how it was built up over the years, for example through entrepreneurship, inheritance, employment income or investments. The review tests whether size and origin plausibly fit together, and it is mandatory for higher risks and politically exposed persons.

Source of funds describes the origin of the specific assets deposited into a business relationship or moved in a transaction. The review uses bank statements, contracts or settlement notes to establish exactly where these funds come from, and complements the question of where the total wealth originates.

Enhanced due diligence (EDD) means the additional enquiries required for business relationships and transactions with higher money laundering risk: deeper review of source of wealth and funds, the purpose of the relationship, approval by a senior function and closer ongoing monitoring. In Switzerland, Article 6 of the Anti-Money Laundering Act sets out this duty.

The risk-based approach is the founding principle of anti-money laundering under the FATF standards: due diligence is not applied uniformly but graduated according to the concrete risk of a relationship. Applying it means defining risk criteria, classifying each relationship and aligning review depth, approval and review frequency accordingly.

A self-regulatory organisation (SRO) is a private body recognised by FINMA that supervises financial intermediaries — and since 1 October 2026 also certain advisers — for compliance with the Swiss Anti-Money Laundering Act. It issues regulations on due diligence duties, has its members audited and can impose sanctions up to expulsion.

A suspicious activity report is a financial intermediary’s notification to the Money Laundering Reporting Office Switzerland (MROS) at the Federal Office of Police when it knows or has founded suspicion that assets stem from a felony or aggravated tax misdemeanour or serve terrorist financing. The duty to report is set out in Article 9 of the Anti-Money Laundering Act.

AMLA is the European Union’s authority for countering money laundering and terrorist financing, based in Frankfurt am Main. Established by Regulation (EU) 2024/1620, it coordinates national supervisors and financial intelligence units and is due to supervise selected cross-border financial institutions directly from 2028.

The AMLR is Regulation (EU) 2024/1624 on preventing the use of the financial system for money laundering and terrorist financing. It applies directly in all member states from 10 July 2027 and replaces many national rules with a single rulebook, covering customer due diligence, beneficial ownership and an EUR 10,000 cash payment limit, among other things.

An independent asset manager manages clients’ assets under a power of attorney while the assets are held by a custodian bank. In Switzerland, since the Financial Institutions Act (FinIA), such managers need a FINMA licence and are supervised on an ongoing basis by a supervisory organisation; the Anti-Money Laundering Act applies to them directly.

A trust is a common law arrangement in which a settlor transfers assets to a trustee, who holds and manages them for beneficiaries under the terms of the trust deed. A trust is not a legal entity. Switzerland recognises foreign trusts through the Hague Trust Convention, which has applied to it since 2007.

A family foundation is a foundation whose assets are tied to a family. Under Swiss law, Article 335 of the Civil Code allows only certain purposes, such as covering the costs of education, endowment or support of family members; pure maintenance foundations are not permitted. Since 2016 family foundations must be entered in the commercial register.

In a family office, a mandate is the unit of work for a family or client: it bundles the people, companies, foundations and trusts involved, the assets, documents, obligations and the agreed services. Access rights, deadlines, reports and invoices all hang off the mandate.

A custodian bank holds a client’s securities and accounts, executes transactions and provides statements. In the independent asset manager model, assets sit with the custodian while the manager decides under a limited power of attorney. Family offices often work with several custodians at once.

Non-bankable assets are assets not held in a bank custody account and not valued continuously by a bank: real estate, direct participations in companies, private loans, art, collections, vehicles or self-custodied crypto assets. They need their own valuation with date and method to flow into an overall view.

Reconciliation is the systematic comparison of one’s own records with external sources, typically custodian statements: positions, cash, transactions and prices. Differences, known as breaks, are detected, ranked by severity and resolved before figures flow into reports, invoices or tax documents.

An audit trail is a complete, chronological and tamper-proof record of who viewed, entered, changed, approved or deleted which data, and when. It makes decisions and controls provable and is the basis of every audit under anti-money laundering and data protection obligations.

The four-eyes principle requires critical operations to be checked or approved by two independent people: payments, master data changes, accepting higher-risk relationships or releasing reports. It reduces errors and abuse, but only works if the second check is genuine and documented.

KYC remediation is the systematic updating of an existing client book to current due diligence standards: obtaining missing documents, updating outdated information, reassessing risk classifications and closing gaps. Triggers include new regulations, audit findings, acquiring a book or years of neglected reviews.

Professional secrecy obliges certain professions to keep secrets entrusted to them. In Switzerland, Art. 321 of the Criminal Code protects it, among others for lawyers, notaries, auditors bound to confidentiality under the Code of Obligations, doctors and their auxiliaries. For lawyers, Art. 13 of the Lawyers Act (BGFA) sets out the duty without time limit.

Auxiliaries are people who help a professional bound by secrecy in their work and learn of secrets in doing so, such as assistants, secretaries or trainees. Under Art. 321 of the Swiss Criminal Code they are themselves bound by professional secrecy. Whether and when external providers such as IT or cloud vendors count as auxiliaries is disputed and must be assessed case by case.

The engagement agreement governs the relationship between client and law firm, fiduciary or adviser. In Switzerland it is usually a simple mandate under Art. 394 ff. of the Code of Obligations: careful performance is owed, not a particular result, and either side may end it at any time. Terms of engagement add fees, communication and data protection.

A conflict of interest exists when a professional cannot represent a client’s interests independently because their own interests or those of another client stand in the way. For lawyers, Art. 12(c) of the Swiss Lawyers Act requires avoiding any such conflict. Firms therefore run a systematic conflict check before accepting every mandate.

A Minimum Viable Product (MVP) is the simplest version of a product that delivers real value to early users and lets a team test its core assumptions with the least effort. It is not a half-finished product but a focused one — built to learn whether the core idea works before investing in everything around it.

Next.js is a React framework for building web applications and websites. It adds server-side rendering, static generation, file-based routing, API routes, and performance optimisations on top of React, making it a standard choice for production web apps and SaaS products that need both rich interactivity and strong SEO.

Supabase is an open-source backend-as-a-service platform built around PostgreSQL. It provides a hosted database, authentication, instant APIs, file storage, and serverless functions, with row-level security as a core feature. It lets development teams build the backend of a SaaS product quickly while retaining the power and portability of a standard Postgres database.

A Progressive Web App (PWA) is a web application that uses modern browser capabilities to behave like a native app — it can be installed to the home screen, work offline, send push notifications, and load quickly — while still being a website accessible through a URL. PWAs give SaaS products an app-like experience without separate native builds for each platform.

Data residency refers to the geographic location where data is physically stored and processed. For European customers it usually means a requirement that personal data remain within the EU or a specific country. For SaaS providers, offering EU data residency — hosting data in regions like Frankfurt — simplifies GDPR compliance and is increasingly demanded by privacy-conscious customers.

SaaS — Software as a Service — is a model in which software is delivered over the internet as a subscription service rather than installed and run on the customer’s own machines. The provider hosts, maintains, secures, and updates the application centrally; customers access it through a browser and pay recurring fees. It is the dominant model for modern business software.

Multi-tenancy is a software architecture in which a single running instance of an application serves multiple customers (tenants), with each tenant’s data logically isolated from the others. It is the foundation of most SaaS products, making them economical to operate and update, while placing strict demands on data isolation and security.

Row-Level Security (RLS) is a database feature that enforces access-control rules at the level of individual rows, so that a given user or tenant can only see or modify the rows they are permitted to — even if a query forgets to filter for them. In multi-tenant SaaS it is the strongest line of defence against cross-tenant data leakage.

An API is a defined interface through which one program can use another’s functions. It fixes which requests are permitted and which responses to expect — making it a contract, not merely a technical connection.

A webhook is a notification a system sends of its own accord to a registered address when an event occurs. It reverses the direction: instead of repeatedly asking, the receiving system is told as soon as something happens.

Idempotency means that performing the same operation several times leaves the same state as performing it once. In distributed systems it is not a nicety but the precondition for a retry being safe.

CI/CD is the automation of integration and delivery: every change is built and checked automatically, and approved builds go live automatically. The purpose is to find errors early and make releases unremarkable.

Observability is the ability to infer a system’s internal state from the signals it emits. It goes beyond classic monitoring: monitoring answers questions defined in advance, whereas observability lets you ask the unexpected ones a real incident raises.

A service worker is a script the browser runs in the background, separately from the web page. It sits between the application and the network, can intercept requests, cache responses and receive push messages — making it the technical foundation for offline capability and notifications in Progressive Web Apps.

Offline-first is an architecture principle in which an application works with data on the device first and synchronises with the server in the background. A missing connection is therefore not an error but a normal state. The core challenge lies in synchronisation and conflict resolution, not in storage.

Vendor lock-in describes dependence on a supplier that makes switching so costly or risky that it effectively does not happen. Causes include proprietary data formats, missing export options, deeply integrated special features and contracts without an agreed exit. With SaaS, the central question is what happens to your data on termination.

Single sign-on (SSO) lets users sign in once with a central identity provider and gain access to several applications. Two standards dominate in B2B SaaS: SAML 2.0, common in enterprises, and OpenID Connect (OIDC) built on OAuth 2.0, common in modern applications. For enterprise customers SSO is often a buying requirement.

RPO (recovery point objective) is the maximum tolerable data loss, measured as the time span before an outage. RTO (recovery time objective) is the maximum tolerable time until operation is restored. Both targets determine how backups, replication and contingency plans must be built, and they belong in customer contracts.

Feature flags are switches in code that turn features on or off at runtime without redeploying. They separate shipping code from releasing a feature and enable gradual rollouts, tests with individual customers, fast switch-off when errors occur and features per plan or tenant.

A penetration test is an authorised, simulated attack on an application or infrastructure in which specialists find vulnerabilities before attackers do. Unlike automated scans, a pentest also examines logic and authorisation flaws, such as whether one tenant can access another’s data. The result is a report with prioritised findings.

Rate limiting caps how many requests a user, tenant, key or IP address may make to an application within a period. It protects against abuse, brute-force attacks, overload and cost explosions, for example with AI features. Exceeded limits are usually answered with HTTP status 429.

Secrets management is the secure handling of credentials such as API keys, database passwords, signing keys and tokens: where they are stored, who has access, how they reach applications, how they are rotated and how leaks are detected. Ground rule: secrets never belong in code and never in the browser.

Monthly Recurring Revenue (MRR) is the predictable revenue a subscription business earns each month, normalised to a monthly figure. It is the central health metric for SaaS, capturing the recurring income from active subscriptions and forming the basis for tracking growth, churn, and expansion.

Churn is the rate at which customers or recurring revenue are lost over a period. Customer churn measures the share of customers who cancel; revenue churn measures the recurring revenue lost. It is one of the most important SaaS metrics, because high churn quietly undermines growth no matter how many new customers are acquired.

Net Revenue Retention (NRR) measures how much recurring revenue a cohort of existing customers generates now compared with a year ago, including expansion and after subtracting contraction and churn — but excluding new customers. An NRR above 100% means the existing base grows in value on its own; it is a key indicator of SaaS health.

Customer Acquisition Cost (CAC) is the total cost of acquiring a new customer — typically all sales and marketing spend in a period divided by the number of customers gained. It measures the efficiency of growth, and is most meaningful when compared against the value a customer generates over their lifetime.

Customer Lifetime Value (LTV) is the total recurring revenue — or gross profit — a business can expect from a customer over the entire duration of the relationship. It depends on how much a customer pays, the margin on it, and how long they stay, and it is the value side of the equation that justifies acquisition cost.

ARR is the annualised recurring revenue from active subscriptions. It counts only reliably recurring income — one-off revenue such as setup fees, training or project work explicitly does not belong in it, because it says nothing about the reliability of the revenue base.

CAC payback period is how long it takes for a customer to earn back the cost of acquiring them. It is calculated from contribution margin, not revenue — and it determines how much capital growth ties up.

Freemium is a pricing model with a permanently free tier and paid extensions. It only works when the free tier delivers genuine value and at the same time has a natural limit that leads to an upgrade.

Product-market fit is the state in which a product meets a need so well that demand and retention arise on their own. It is not a milestone with a date but a state that holds per segment and can be lost again.

Dunning is the automated process following a failed subscription payment: staggered retry attempts, accompanied by notifications to the customer. It recovers revenue that would otherwise show up as churn, although the customer never intended to cancel at all.

Cohort analysis groups users by when they started and tracks each group over time. It shows whether the product is improving — which an overall average systematically hides, because new arrivals pull it up.

The QR-bill is the Swiss standard for invoices with a payment part. A Swiss QR Code carries all payment details in structured form, so payments are captured without retyping. Since 30 September 2022 it has fully replaced the orange and red payment slips; the technical rules are set out in the Swiss Payment Standards.

A Large Language Model (LLM) is an AI model trained on vast amounts of text to understand and generate human language. It predicts likely continuations of text and can be applied to tasks such as writing, summarising, answering questions, and reasoning. LLMs power most modern generative-AI products and are typically accessed as general-purpose models via an API.

Retrieval-Augmented Generation (RAG) is a technique that improves a language model’s answers by retrieving relevant information from a knowledge source at query time and supplying it to the model as context. Instead of relying solely on what the model learned in training, RAG grounds responses in current, specific, trustworthy data — reducing hallucination and enabling answers from private content.

Prompt engineering is the deliberate wording and structuring of input to a language model in order to obtain more reliable results. It covers a clear task, context, examples and format requirements — and in production systems is closer to software engineering than to writing.

An embedding is the representation of a text, image or other content as a numeric vector in which similar things sit close together. That makes similarity computable — the basis of semantic search and of RAG.

A vector database stores embeddings and returns the entries most similar to a query. It is optimised for similarity search in high-dimensional spaces — a task classic indexes were not built for.

A hallucination is an output from a language model that is plausibly worded but factually wrong or invented. It arises not from a fault but from how the model works: it produces probable text, not verified statements.

The context window is the maximum amount of text a language model can process in one request — input and output together. It is measured in tokens and limits how much context a model has available at once.

Fine-tuning is further training a pre-trained model on your own examples to make behaviour, style or format more reliable. It is suited to teaching form — not to keeping knowledge current.

An AI agent is a system that connects a language model to tools and works toward a goal over several steps — planning, calling functions, evaluating results and deciding the next step, rather than only answering.

LLM evaluation is the systematic checking of a language model’s outputs against defined expectations. It replaces intuition with a repeatable measurement and is the precondition for judging any change to an AI system at all.

Prompt injection is an attack on LLM applications in which injected instructions get the model to ignore its actual instructions — directly via user input or indirectly via content the model processes, such as web pages, documents or emails. OWASP lists it as the top risk for LLM applications.

Guardrails are control layers around a language model that check inputs and outputs and limit behaviour: topic filters, checks for personal data, schema validation, grounding against sources or approvals before actions. They make LLM features more predictable but replace neither secure architecture nor evaluation.

Tool calling, also function calling, is a language model’s ability to generate structured calls based on described tools — a database query, a search or an API action — which the application executes and whose result goes back to the model. It is the technical foundation of AI agents.

The Model Context Protocol (MCP) is an open standard that lets AI applications access external tools, data and templates in a uniform way. An MCP server exposes a system’s functions, and an MCP client in the AI application uses them. Anthropic published the protocol in late 2024; many vendors now support it.

Chunking is splitting documents into smaller sections before they are embedded and indexed for a RAG system. The size and logic of the sections determine whether search finds the right passage and whether the language model gets enough context for a correct answer. Poor chunking is one of the most common causes of weak RAG quality.

A token is the unit in which language models process text: a word, part of a word, a punctuation mark or a whitespace fragment. Context windows, output length, latency and the cost of LLM APIs are measured in tokens, split into input and output tokens. German text needs more tokens than English text of the same content in many tokenizers.

Topical authority is the degree to which a website is recognised by search engines as a comprehensive, trustworthy source on a particular subject. It is built by covering a topic deeply and broadly through interlinked content rather than chasing isolated keywords, and it increasingly determines how well a site ranks across an entire subject area.

Programmatic SEO is the practice of generating large numbers of targeted, high-quality landing pages at scale from structured data and templates — each page serving a specific search query. Done well, it captures long-tail search demand efficiently; done badly, it produces thin, duplicative pages that search engines penalise.

Core Web Vitals are a set of measurable metrics Google uses to assess the real-world user experience of a web page — focusing on loading performance, interactivity, and visual stability. They are a confirmed ranking signal, so good Core Web Vitals matter for both user satisfaction and search visibility.

Schema markup is structured data added to a web page — using a shared vocabulary from Schema.org — that describes the page’s content to search engines in a machine-readable way. It helps search engines understand the content precisely and can enable rich results such as FAQs, breadcrumbs, and ratings in the search listing.

E-E-A-T stands for Experience, Expertise, Authoritativeness, and Trust — the qualities Google’s search quality guidelines use to judge the credibility of content and its creators. It is not a direct ranking score but a framework that shapes how Google assesses quality, and it matters most for topics that affect people’s wellbeing, finances, or safety.

hreflang is an annotation by which a site tells search engines which language or regional versions of a page exist. It prevents language variants being treated as duplicates and steers which version is served to which audience.

A canonical tag names the authoritative URL of a page. It consolidates the ranking signals where the same content is reachable at several addresses, and stops a search engine indexing the wrong variant or treating several variants as competing duplicates.

Crawl budget is the extent to which a search engine fetches a site in a given period. It only becomes a bottleneck on large sites — and then usually not because of volume but because it is spent on worthless URLs.

GEO means optimising content to be cited in the answers of AI systems. Unlike classic SEO it does not target a place in a list but being used as a source in a generated answer.

llms.txt is a text file in a site’s root that gives AI systems an ordered overview of its most important content. It is an emerging proposal rather than a binding standard — and complements robots.txt and the sitemap rather than replacing them.

Internal linking means the links between pages of the same site. It determines which pages get found, how authority is distributed, and which thematic relationships search engines recognise — and it is the most underrated SEO lever.

Search intent is the goal behind a query: does someone want to know something, reach a particular site, orient themselves before buying, or act directly? Serving the wrong intent means not ranking, however good the content.

AI Overviews are AI-generated summaries that Google shows above the classic results for certain queries, with links to the sources the answer is based on. For websites they shift the question from “rank one” to “cited as a source”. According to Google, a page must be indexed and eligible for snippets to be included.

Indexing is the step in which a search engine processes a crawled page and adds it to its index, from which search results are generated. Only indexed pages can rank or be cited in AI answers such as AI Overviews. Crawled does not mean indexed: Google decides per URL whether to include it.

A zero-click search is a query answered directly on the results page so that the user clicks no website — through featured snippets, knowledge panels, maps or AI summaries. For businesses it shifts the value of visibility: presence and citation become more important, raw click counts less telling.

IndexNow is an open protocol that lets websites actively inform search engines about new, changed or deleted URLs instead of waiting for the next crawl. It was introduced in 2021 by Microsoft Bing and Yandex and is supported by several search engines. Google does not use IndexNow so far; there, the sitemap and Search Console remain the route.

Entity SEO means optimising so that search engines and language models recognise a company, person or product as a distinct entity and connect it correctly with topics, places and other entities. The means are consistent information, structured data with unique identifiers, references to official profiles and topically clear content.

AI crawlers are bots from AI providers that fetch web pages — for training models, for the search indexes of AI assistants or for fetches on a user’s behalf. Well-known examples are OpenAI’s GPTBot and OAI-SearchBot, Anthropic’s ClaudeBot, PerplexityBot and CCBot. They are controlled via robots.txt, separately by purpose.

An XML sitemap is a machine-readable list of a website’s URLs that helps search engines discover and crawl them. A single sitemap may contain at most 50,000 URLs or 50 MB uncompressed; larger sites use a sitemap index. It should contain only indexable, canonical URLs with an accurate modification date.

robots.txt is a text file in a website’s root directory telling crawlers which areas they may fetch. It controls crawling, not indexing: a blocked URL can still appear in the index if it is linked. The protocol has been standardised as RFC 9309 since 2022.

NAP stands for a business’s name, address and phone number. NAP consistency means these details are identical on your own website, in the Google business profile, in directories such as local.ch and search.ch and in structured data. Consistent details help search engines identify a business unambiguously and show it locally.

The Google business profile, formerly Google My Business, is a business’s free listing in Google Maps and local search. It shows name, address, opening hours, category, photos and reviews. For local searches, a complete, verified and accurate profile is one of the most important factors.

HERMES 2022 is the project management method of the Swiss federal administration, widely used in Switzerland for IT and business projects. It is an open, freely available standard structured around phases, scenarios, roles, and defined deliverables, designed to be scalable and combinable with agile approaches.

Scrum is an agile framework for developing and delivering complex products in short, fixed-length iterations called sprints. It defines a small set of roles, events, and artifacts that help a team work iteratively, inspect progress frequently, and adapt — favouring empirical, incremental delivery over detailed upfront planning.

Kanban is an agile method for managing work by visualising it on a board, limiting the amount of work in progress, and managing the flow of tasks to completion. Unlike Scrum’s fixed iterations, Kanban is continuous and evolutionary — it focuses on smooth, steady flow and incremental improvement rather than time-boxed sprints.

OKR — Objectives and Key Results — is a goal-setting framework that pairs an ambitious, qualitative Objective with a few measurable Key Results that define what achieving it looks like. Set on a regular cadence and kept transparent across an organisation, OKRs align effort toward outcomes rather than tasks.

Professional Services Automation (PSA) is software that supports the core operations of service-based businesses — managing projects, resources, time tracking, and billing in one integrated system. It connects the work delivered to the revenue it generates, giving consultancies and agencies visibility into utilisation, profitability, and project health.

Stakeholder management is the systematic identification, assessment and involvement of everyone who can influence a project or is affected by it. Its absence is the most common cause of project failure.

A risk register is the documented collection of identified project risks with an assessment, a measure and a named owner. Its value lies not in the list but in the entries being reviewed regularly.

A definition of done is a jointly agreed list of conditions a piece of work must meet to count as complete. It removes the most common friction in teams: that done means different things to different people.

A milestone is a defined point in a project at which a verifiable state must have been reached. It has no duration and no effort — its purpose is the decision about whether the project continues as planned.

A user story describes a requirement from the perspective of the person who benefits, usually in the pattern “As a role, I want a goal, so that a benefit”. Acceptance criteria define when the story is fulfilled and make it testable. Together they are agile teams’ tool for keeping requirements small, understandable and testable.

Requirements engineering is the systematic discipline of eliciting, documenting, validating and managing requirements for a system over the course of a project. It connects the perspectives of business, users and technology and reduces the most common project risk: building a system that works correctly but does the wrong thing. A widespread standard is the IREB CPRE certification.

A RACI matrix assigns for each task or decision who carries it out (Responsible), who is answerable and decides (Accountable), who is consulted (Consulted) and who is informed (Informed). It makes responsibilities in projects and processes visible and prevents tasks from getting lost between roles.

A change request is a formal request to change a project’s agreed scope, schedule or content. It describes the desired change, its rationale and its impact on cost, time, quality and risk, and is decided by an authorised body. It keeps a project steerable even as requirements change.