Glossary
Terms from development, design, promotion and AI — plain definitions, plus why each one actually matters on a project.
40 terms
In short
The Veltos.Tech glossary is a reference for digital development and promotion terminology: from MVP, headless CMS and Core Web Vitals through to GEO optimisation, RAG, end-to-end analytics and software import substitution. Each term gets a single clear definition, then a substantive explanation of how it works, when it matters and where people usually get it wrong.
Development
- Legacy codeLegacy code is working code that is expensive to change: it runs on an outdated stack, has no test coverage or has lost its authors, so every edit means reconstructing the logic from the source and risks breaking something unrelated.
- MicroservicesMicroservices are an architecture in which an application is split into several independent services, each with its own data and release cycle: they talk over the network through APIs, deploy separately and scale individually, but they demand mature infrastructure.
- MonolithA monolith is an application built and deployed as a single unit: all the code lives in one repository, runs in one process and works against one database, releasing as a whole, which buys simplicity at the cost of coupling.
- RefactoringRefactoring is changing the internal structure of code without changing its external behaviour: the functionality before and after is identical, what changes is readability, coupling and the cost of the next change, verified by tests and regression checks.
- Technical debtTechnical debt is the accumulated gap between how a system is built and how it should have been built: every shortcut taken to hit a deadline turns into a permanent surcharge on the cost and duration of everything built afterwards.
- APIAn API is a fixed set of rules by which one program requests data or actions from another: method addresses, request and response formats, authentication and rate limits, documented so that both sides can evolve independently.
- CI/CDCI/CD is a pipeline that on every commit builds the project automatically, runs tests and checks, and then ships the passing build to a staging environment or to production, removing manual deployment and the human error that comes with it.
- Core Web VitalsCore Web Vitals are Google’s three page experience metrics: LCP, the render of the largest element, should stay within 2.5 seconds, INP, the response to an interaction, within 200 milliseconds, and CLS, cumulative layout shift, within 0.1.
- Headless CMSA headless CMS is a content platform with no built-in templating layer: editors work in a familiar admin panel while a website, a mobile app and any other channel pull the same content over an API and render it themselves.
- MVPAn MVP is the first release of a product that carries exactly one user journey end to end, shipped in four to eight weeks so the demand hypothesis gets tested against real users and real payments rather than survey answers.
- PWAA PWA is a website that behaves like an app: it installs to the home screen straight from the browser, works offline through a service worker and can send push notifications, all without going through App Store or Google Play review.
- SSR vs CSRSSR and CSR are two ways of producing HTML: with server-side rendering the page is assembled on the server and arrives complete, while with client-side rendering the browser receives an empty shell and JavaScript draws the content after load.
Design
- Responsive designResponsive design is an approach where a single layout reorganises itself to the viewport: grid, typography and block order change at defined breakpoints instead of sending the visitor to a separate mobile version on its own subdomain.
- WireframeA wireframe is a rough screen schematic without colour, typography or imagery: it fixes which blocks exist, their priority and the transitions between screens, so the interface logic is agreed before expensive visual design starts.
- Design systemA design system is a set of tokens, components and usage rules that exists simultaneously in the design files and in code, fixing colours, typography, spacing and element states so that a new screen is assembled from existing parts rather than reinvented.
- Usability testingUsability testing is checking an interface with real people: participants from the target audience carry out realistic tasks while thinking aloud, and the team records where they stall, where they go wrong and what they fail to find.
Promotion
- Commercial ranking factorsCommercial ranking factors are the group of signals Yandex applies to commercial queries: visible prices, assortment breadth, contact and company details, delivery and payment terms, reviews and live chat, in other words evidence that buying here is actually possible.
- Schema markupSchema markup is structured data based on the Schema.org vocabulary, usually delivered as JSON-LD: it tells search engines and AI answer engines which part of the page is a product, a price, a review, an organisation or a question with its answer.
- Behavioural signalsBehavioural signals describe how users act in the results page and on the site: whether they click the snippet, whether they bounce back to search afterwards, how long they stay; Yandex uses them in ranking and penalises manipulation aggressively.
- Semantic coreA semantic core is an intent-grouped list of the search queries a site should be found for: collected from Wordstat, Webmaster and competitor tools, then clustered by results-page overlap so you know how many pages are needed and what each one covers.
- End-to-end analyticsEnd-to-end analytics is the system that links an ad click to booked revenue: data from ad accounts, call tracking, the website and the CRM is stitched together by visitor identifier so payback can be calculated per channel, campaign and keyword.
- AEOAEO is optimising for direct answers: a page is built so a search engine or voice assistant can extract a ready response to a specific question, whether that is a short paragraph under a heading, a list of steps, a table or a marked-up FAQ block.
- GEOGEO is optimising content so generative engines such as AI Overviews, Yandex Neuro, GigaChat and Perplexity quote it, and it layers on top of SEO rather than replacing it, because Yandex neural search keeps no separate index of its own.
- SEOSEO is the work on a site and its surroundings that earns organic positions in Yandex and Google: technical foundation, keyword coverage, content, commercial and behavioural signals and links, with the first durable results arriving after three to six months.
- UTM tagsUTM tags are parameters appended to a link (utm_source, utm_medium, utm_campaign, utm_content, utm_term) that pass the traffic source to analytics; without them paid, placement and email traffic collapses into one undifferentiated referral bucket.
AI and ML
- Prompt engineeringPrompt engineering is the design of instructions for a language model: role, context, task, output format and constraints are written so the result is reproducible, and quality is judged against a set of test cases rather than by impression.
- EmbeddingsEmbeddings are numeric vectors a model produces from text, images or products so that items close in meaning end up close in the vector space, which is what makes semantic search, recommendations and duplicate detection possible.
- AI agentAn AI agent is a program in which a language model does not merely answer but acts: it receives a goal, chooses its own tools such as search, APIs, a database or sending mail, works through steps in a loop and stops when the task is done or a limit is hit.
- LLMAn LLM is a large language model: a neural network trained on vast amounts of text to predict the next token, which is why it can write, summarise and answer, yet it stores no facts the way a database does and can be confidently wrong.
- RAGRAG is a pattern where a language model first retrieves relevant fragments from your own knowledge base and answers from them, which delivers current data and a source citation without any fine-tuning of the model itself.
Business and process
- Russian personal data law (152-FZ)152-FZ is the Russian personal data law: it requires that data on Russian citizens be collected in databases located in Russia, that the regulator be notified of processing, that consent be obtained separately, and that any breach be reported within 24 hours.
- Outsourcing vs in-houseOutsourcing and in-house are two ways to cover a development need: contract an external team at an agreed hourly rate, or build a staff team where salaries come with payroll taxes, recruitment, equipment and management on top.
- Software import substitutionSoftware import substitution means moving information systems onto Russian products listed in the state software registry: for public sector customers and critical infrastructure operators it is a legal requirement, and for other businesses it is a condition of tender access and a way to remove vendor abandonment risk.
- Product backlogA product backlog is a priority-ordered list of everything that could be built into a product: it is not a dated plan, it is continuously revised, and one person, the product owner, is accountable for the order of it.
- Technical specificationA technical specification is the document that fixes what exactly will be built and on what grounds the work counts as accepted: goals, roles, scenarios, screen-level requirements, integrations, non-functional requirements and an explicit out-of-scope section.
- Unit economicsUnit economics is the profitability calculation for a single business unit, usually a customer, an order or a subscription, comparing the revenue a customer brings over their lifetime (LTV) with the cost of acquiring them (CAC), with a ratio of three or more treated as sustainable.
- DiscoveryDiscovery is a one to three week pre-project phase that frames the problem, studies users and competitors, describes scenarios and constraints, and produces a prototype, a specification and an estimate with a justified range instead of a number pulled from the air.
- Fixed price vs time and materialsFixed price and time and materials are two contract models: the first fixes scope, deadline and price with the risk premium sitting on the contractor, while the second pays for hours actually worked at an agreed rate and leaves scope control with the client.
- ROIROI is the return on investment metric: the difference between income and cost divided by cost, expressed as a percentage, with the marketing variant ROMI counting only the advertising budget on the cost side.
- SLAAn SLA is a service level agreement: it fixes system availability as a percentage, response and resolution times by incident severity, the hours support operates, and the compensation owed when the provider fails to meet those numbers.