Skip to main content
A reference for the vocabulary used across the app and these docs. If a term is used anywhere in Videntic, it should be here. Alphabetical.
AI search — The category of products (ChatGPT, Perplexity, Gemini, Claude, Google AI Overviews) where users ask questions conversationally and get synthesized answers instead of a list of links. AI Visibility Score — The equal-weight mean of each tracked provider’s brand text-mention rate, expressed from 0–100. Providers with no responses are excluded. See AI Visibility Score. Analysis — One end-to-end run of Videntic’s pipeline on a website: generate prompts, simulate across models, score responses, audit the site, produce metrics. A report is produced from every analysis. Audit (GEO audit) — The on-site analysis that scores every tracked page against the GEO‑16 framework. Complements the off-site visibility metrics by telling you why you are (or aren’t) being cited. Brand document — A file you upload to Videntic (brand guidelines, tone-of-voice doc, writing sample) that Sira uses as context to stay on-brand in its responses and drafts. Cadence — How often Videntic runs an analysis on a given website: daily, weekly, biweekly, monthly, or quarterly. Set per website in Settings → Reporting. Citation — A mention of your brand or domain in an LLM’s response. Videntic counts a citation when the mention is attributional (not dismissive). Citation Position — How early your brand appears in an LLM response that cites you. Lower is better — position 1 is the first mention. See Citation Position. Competitor (tracked) — A domain classified as a direct business rival in your competitor list. Only tracked competitors count toward your Share of Voice. Content gap — A prompt where your brand is not cited but your competitors are. Ranked by opportunity, with a recommended content type and suggested angles. See Content gaps. Content Creator — One of Sira’s four specialists. Writes blog drafts, product pages, and other content artifacts. Available on Pro and Max plans. Crawl — The initial read of your website that Videntic performs after you add it. Builds a content index used for GEO‑16 scoring and Sira context. Different from an analysis. Data Analyst — One of Sira’s four specialists. Answers numeric and analytical questions about your metrics, competitors, and trends. Draft — A blog post generated by Sira but not yet published. Drafts live in Content → Drafts and can be edited, regenerated, or pushed to a CMS as an unpublished article. GEO (Generative Engine Optimization) — The practice of making your content citable by LLMs. The GEO equivalent of SEO for traditional search engines. GEO Auditor — One of Sira’s four specialists. Walks a single page through the GEO‑16 framework with specific, actionable fixes. GEO‑16 — Videntic’s 16-pillar framework for evaluating a page’s readiness to be cited by LLMs. Grouped into Technical, Content, Authority & Trust, and UX themes. See the GEO audit for how it’s surfaced in the app. Grace period — The short window (about 7 days) after a payment failure during which your full plan remains active. Gives you time to update your payment method without losing access. LLM — Large Language Model. The underlying AI that powers ChatGPT, Perplexity, Gemini, Claude, and Google AI Overviews. Videntic simulates against the major LLMs in every analysis. Pillar — One of the sixteen categories in the GEO‑16 framework. Each pillar has a score, a status (passing, borderline, failing), and a list of specific issues found. Plan — The subscription level of a website: Basic, Pro, Max, or Enterprise. Each plan sets a different prompt library size, number of competitor slots, analysis frequency, and feature access. See Plans. Prompt — A single customer-style query stored in your prompt library and used to test LLM responses. In a Videntic analysis, each active prompt is run against every LLM included in your plan. Your plan caps how many prompts you can keep active at once. Prompt library — The set of prompts Videntic runs against the LLMs on every analysis for a website. Managed per website; your plan caps how many prompts can be active at a time. Recipient — An email address that receives scheduled or on-demand reports for a website. Configured per website in Settings → Reporting. Report — The PDF summary produced from an analysis, delivered by email to configured recipients. See Reports. Share of Voice (SOV) — For each provider, your mentions divided by the combined mentions of you and tracked competitors; the headline is the equal-weight mean of non-empty provider pools. See Share of Voice. Sira — Videntic’s AI agent. Routes your request to the right specialist (Data Analyst, Content Creator, Strategy Advisor, or GEO Auditor) and returns an answer, a draft, or an audit. Specialist — One of the four roles Sira routes to. Each specialist is tuned for a specific kind of task. Strategy Advisor — One of Sira’s four specialists. Answers planning and prioritization questions — what to work on first, how to respond to competitive moves, how to frame results for leadership. Task — A long-running Sira job (e.g. draft a full blog post, audit an entire site). Tasks run in the background; you get an email when they finish. Tier (visibility tier) — The human-readable category label for your AI Visibility Score: Building Visibility, Gaining Traction, Well Positioned, or Category Leader. (For subscription levels, see Plan above.) UGC — User-generated content. A classification for competitor candidates — Reddit, Quora, Stack Exchange — that’s tracked for intel but excluded from Share of Voice. Verification — The one-time proof-of-ownership step after adding a website. Ensures you can only track sites you control. Website — A tracked domain in Videntic. Each website has its own analyses, audits, competitors, reports, and settings.

Core metrics

How the measured concepts work together.

What is Videntic?

The high-level primer.

FAQ

Common questions that use this vocabulary.
Last modified on July 17, 2026