Profound Glossary

Last updated: April 19, 2026

This document overviews key metrics and terms relevant to answer engine optimization and the Profound platform.

Key Metrics

This table overviews key metrics that will inform on your brand's performance in answer engines.

Metric	Definition	Example
Visibility	A binary metric that tracks whether your brand did or did not appear in a particular AI Answer—if your brand is present, you gain visibility for that response.	McDonald's is mentioned in 50 responses out of a total 100 responses tracked. The visibility score will be 50% (50/100).
Citation Share	A metric that tracks how often a specific citation is used across all citations.	www.mcdonalds.com/careers is used 10 times to answer questions in AI search. There are a total of 1,000 citations, meaning www.mcdonalds.com/careers has a citation share of 1% (10/1,000).
Share of Voice (SoV)	A comparative metric indicating how frequently your brand or product is mentioned versus all other brands.	McDonald's is mentioned in 50 responses out of a total 100 responses tracked. There are a total of 500 companies mentioned across all 100 responses. The share of voice will be 10% (50/500).
Aggregated Sentiment Score	The averaged sentiment value compiled from all relevant AI responses regarding a specific brand or competitor. It's the ratio of positive sentiment occurrences to all sentiment points/occurences.	McDonald's has 75 positive mentions and 25 negative mentions across all responses. The Sentiment score will be 75% (75/100).

Core Terms

Core Terms are terms you will encounter as you optimize your brand's performance and navigate the Profound platform.

AI Answer Engines

AI-driven services (e.g., ChatGPT, Bard, Bing Chat) that produce responses to users’ prompts in real time. Profound aggregates and analyzes responses from these engines to measure brand presence.

Answer Variation

Minor or significant alterations in the text, structure, or tone of responses generated by AI Answer Engines when presented with similar or slightly tweaked prompts.

Bot Tracker

A tool within Profound that detects and logs visits from AI Answer Engine bots/crawlers to your site in real time.

Brand

An entity (e.g., a company or product line) whose presence, mention, or sentiment is monitored by Profound.

Citations

When AI models reference or link to specific webpages, articles, or resources in their responses. In Profound, these references are recorded as “Citations.”

Citation Analysis Page

Analysis page in Profound that interprets the volume, distribution, and sources of citations across multiple AI Answer Engines.

Citation Tracking

The ongoing process of monitoring and recording citations as they appear in AI-generated responses over time.

Competitors

Other companies or brands that you monitor alongside your own. Profound tracks their presence in AI responses for comparison.

Configuration (Configs)

The initial setup process in the Profound Platform where you define your brand, competitors, topics, and other customizations.

Data Cleansing

The automated or manual process by which Profound refines raw, unstructured AI responses into actionable data (e.g., brand mentions, competitor presence, relevant citations).

Date Range (Time Period)

The specific timeframe over which prompts are run and AI responses are aggregated.

Domains

The high-level websites (e.g., example.com) behind each citation pulled by the AI Answer Engines.

Executions

The number of times that Profound analyzes a prompt's responses within ChatGPT, Perplexity, Google AIO, and Microsoft Copilot.

Exports

Downloadable data sets capturing prompts, AI responses, mention volumes, sentiment scores, and more.

Export Filters

A set of options within the Exports modal that refines which subset of data you download (e.g., only a certain topic, date range, or specific brand mentions).

HTML Rendering

The process by which a web page’s code is executed and displayed (i.e., the final experience a human or bot sees).

Indexing

The method AI bots use to scan, store, and retrieve information from your site for future use in AI Answers.

Mentions

References to a brand or product within an AI Answer. A single response can contain multiple mentions of the same brand.

Mention Frequency

The total number of times a particular brand is mentioned across all analyzed responses over a given Date Range (Time Period).

Node Chart

A visual representation in Profound that links Topics, Domains, and the volume of Citations.

Nodes

The individual elements (domains or topics) depicted in the Node Chart.

Pages

Specific URLs cited by AI Answer Engines, distinct from the broader Domains.

Platforms

In the context of Profound, these are the AI-based search or Q&A systems (e.g., ChatGPT, Bing Chat, Bard) from which Profound pulls data.

Profound Platform

Profound is the overarching system through which you track brand presence and competitive insights across various AI Answer Engines.

Prompt

An individual query submitted to an AI Answer Engine. Profound uses two types of prompts:

Visibility prompts: Track whether your brand appears in AI responses. These can be customized to match your specific use cases and target queries.
Sentiment prompts: Automatically generated prompts that measure how your brand is described when it appears. These follow a structured format (e.g., "Evaluate [your brand] on [topic]") and are always branded, meaning they specifically name your company and competitors.

Sentiment prompts feed data into the Sentiment tab in Answer Engine Insights and count toward your total prompt limit. They can be edited, customized, or deleted if you prefer to allocate that prompt budget elsewhere.

Slight changes in phrasing can affect how answers are generated.

Prompt Generation Model

The proprietary tool or algorithm Profound uses to systematically vary phrasing and context in queries, ensuring a wide range of AI responses.

Prompt Volume

The total number of prompts sent to AI Answer Engines within a specified Date Range (Time Period).

Responses

The complete set of AI-generated answers from the prompts, analyzed for brand presence, sentiment, and Citations.

Search Variation

The fluctuation in search or query outcomes based on differences in wording, region, and context.

Sentiment

A model’s evaluation of how positively or negatively a brand is viewed based on AI responses.

Topic

Key concepts or areas of focus (e.g., product lines, features, or marketing themes) that you want to rank for in AI Answers.

Topic Analysis Page

A specialized section of the Profound Platform dedicated to exploring performance related to a single Topic.