One API call returns 15+ enrichment fields: buyer personas, named entities, sentiment analysis, technology stacks, competitive intelligence, and more. Add expanded_categories=1 to unlock the full data layer.
Website categorization assigns industry labels. Enrichment data goes deeper: it extracts the audience, the sentiment, the entities, the technology, the competitive landscape, and the intent behind every page. It transforms a URL from a simple address into a structured intelligence object containing dozens of actionable attributes.
While categorization answers "what is this site?", enrichment data answers the questions that actually drive decisions: Who visits this site? What companies are mentioned? Is the tone positive or negative? What technologies does this business use? Who are their competitors? What buyer persona does this content speak to?
Enrichment fields are returned alongside standard IAB v3, IAB v2, and Web Filtering categories. Set expanded_categories=1 in your API call to receive the full intelligence package with every request.
Each field includes structured data, confidence scores, and contextual explanations. Parse what you need, ignore what you don't.
Identifies the most likely audience segments visiting a site based on products, language, pricing, and design. Over 2,000 personas from tech enthusiasts to enterprise procurement professionals, each with confidence scores. Perfect for ad targeting and personalization. View all 2,000+ personas.
Detects organizations, products, people, locations, and technologies mentioned on the page. Each entity is classified by type and includes contextual evidence. Enables knowledge graphs, relationship mapping, and competitive monitoring across large URL sets.
Evaluates how each entity is portrayed in context: positive, neutral, or negative. Returns the entity name, type classification, sentiment label, and a contextual explanation. Critical for brand monitoring, reputation tracking, and media analysis at scale.
Overall page sentiment with a numeric score from -100 to +100, confidence percentage, positive/neutral/negative distribution, detected tones (professional, persuasive, casual), and a one-sentence summary. Returned in a separate sentiment_analysis field.
Identifies 4,000+ front-end and back-end technologies: CMS platforms, analytics tools, ad networks, JavaScript frameworks, hosting providers, payment processors, and more. Enables technology-based B2B targeting, competitive analysis, and migration opportunity detection.
Extracts high-level subjects discussed on the page with contextual descriptions. More granular than category labels: identifies specific stories, product features, promotions, and messaging angles. Powers content recommendation and competitive topic tracking.
Important keywords and search terms relevant to the page content. Informs SEO analysis, PPC campaign planning, content gap detection, and semantic clustering. Each keyword represents actual vocabulary used on the site, not inferred terms.
Identifies domains similar in industry, content, and market position. Ideal for competitive landscape analysis, market research, lookalike audience building, and discovering businesses operating in the same space that you may not have found through manual research.
Lists direct competitors with explanations of the competitive relationship. Based on market analysis, product overlap, and content positioning. Distinct from similar domains: returns company names with context rather than just domain addresses.
Estimated demographic profile of the target audience including age ranges, professional roles, education levels, and interest segments. Derived from content analysis, not tracking data. Useful for media planning, market sizing, and audience-product fit assessment.
Identifies brands mentioned on the page and provides reputation analysis: positive, neutral, or negative. Focuses specifically on brand portrayal in context rather than overall page sentiment. Powers brand safety scoring and competitive brand monitoring.
Comprehensive security scan detecting malware, phishing attempts, social engineering, deceptive practices, and other threats. Returns a boolean flag plus detailed threat indicators. Essential for brand safety enforcement, ad verification, and compliance workflows.
Identifies the primary language of the page content with support for 100+ languages. Enables multilingual content routing, international expansion planning, and regional content analysis. High-confidence detection even on pages with mixed-language content.
Extracts the primary legal entity or organization behind the website, including registered addresses when available from footers and contact pages. Supports compliance verification, entity resolution, deduplication, and business validation.
Categorical tags derived from page content, similar to blog post tags. Fine-grained topic markers that enable filtering, grouping, search interfaces, and content clustering at scales where manual tagging would be impossible.
Enrichment data requires no separate API endpoint, no additional authentication, and no extra configuration. Add expanded_categories=1 to any standard categorization request, and the full enrichment package is included in the response alongside your IAB, IPTC, and Web Filtering results.
The response payload grows from a few kilobytes to a rich JSON object containing all 15+ fields. Parse only the fields relevant to your use case, or consume the full package. Every field includes structured data types, confidence scores where applicable, and contextual explanations.
Enrichment fields unlock use cases that simple categorization cannot support. Here are the most common implementations across our enterprise clients.
Match ad creatives to audience personas and page sentiment. A luxury brand can target pages with "affluent professional" personas and positive sentiment, while avoiding negative-sentiment content. Enrichment data enables targeting precision that category-level matching cannot achieve.
Monitor competitor websites at scale using entity extraction, technology detection, and similar-companies data. Track when competitors adopt new technologies, shift messaging, or get mentioned alongside your brand. Build automated competitive dashboards from structured API output.
Score and qualify leads using technology stack detection, company entity extraction, and buyer persona data. A SaaS company selling to enterprise can filter for leads whose websites show enterprise tech stacks and "CTO" or "IT Director" buyer personas with high confidence scores.
Go beyond category-level blocking with entity sentiment and content sentiment analysis. Block ad placements on pages where your brand is mentioned negatively, where competitors are discussed favorably, or where the overall sentiment conflicts with your brand values.
Build recommendation engines that match content to users through topic similarity, shared entities, and persona alignment rather than broad category overlap. A user reading about "cloud security" gets recommended articles that share entities like AWS, Azure, and zero-trust, not just articles in the "Technology" category.
Analyze thousands of websites to map market landscapes using technology adoption data, competitor clusters, audience demographics, and content themes. Identify emerging trends by tracking topic extraction across industry verticals over time.
Our enrichment pipeline is built for high-volume processing with parallel thread support.
Understanding who visits a website is the most valuable intelligence for marketing, sales, and advertising teams. Our buyer persona detection analyzes page content, messaging, products, and design to identify likely audience segments with calibrated confidence scores.
Personas span three dimensions: professional roles (CTOs, marketers, procurement managers), interest-based segments (tech enthusiasts, fitness buffs, automotive fans), and behavioral profiles (comparison shoppers, impulse buyers, B2B decision-makers). Each persona includes a confidence score from 0 to 1 indicating the strength of the signal.
Explore our Web Personas Atlas for an example of how enrichment-driven audience analysis maps the internet's persona landscape across millions of domains.
Most sentiment APIs give you a single score for an entire page. Ours returns two distinct layers that serve fundamentally different purposes.
Content sentiment captures the overall emotional tone of the page: is it promotional, critical, neutral, or mixed? It includes a numeric score (-100 to +100), confidence percentage, tone labels like "professional" or "persuasive", and a distribution showing what percentage of the content is positive, neutral, or negative.
Entity sentiment evaluates how each named entity is portrayed in context. A news article might have neutral overall sentiment but contain strongly negative coverage of a specific company. Entity sentiment catches this. It returns per-entity sentiment labels, classifications, and contextual explanations — critical for brand monitoring, media analysis, and reputation scoring at scale.
Together, these two layers let you build nuanced scoring: is this a positive article about our competitor? Is our brand mentioned in a neutral context on a negative page? Category filters can't answer these questions. Sentiment fields can.
Every website reveals its technology choices in its source code, headers, and behavior. Our detection library identifies over 4,000 technologies across CMS platforms, e-commerce engines, analytics suites, advertising networks, JavaScript frameworks, hosting providers, payment processors, marketing automation tools, security solutions, and customer service platforms.
Combined with the similar-companies and similar-domains fields, technology detection transforms your API from a categorization tool into a competitive intelligence engine. Feed in a list of URLs and receive back: what technologies each business uses, who their competitors are, and which companies operate in similar spaces.
Sales teams use this to qualify leads by tech stack fit. Marketing teams track competitor technology adoption. Investment analysts map technology landscapes across entire market verticals. All from one API call per URL.
Every enrichment field returns structured, typed data rather than free text. Personas come with float confidence scores. Sentiment comes with integer scores, string labels, and distribution objects. Entities come with type classifications and contextual evidence.
This structure means your downstream systems can apply quality thresholds appropriate to each use case. A brand safety system might require entity sentiment confidence above 0.8 before blocking a placement. A content recommendation engine might include personas with scores as low as 0.3 to broaden discovery. You control the precision-recall tradeoff.
Our enrichment models are continuously trained on fresh data, validated against human-labeled datasets, and refined through customer feedback. Technology detection libraries are updated weekly to track new frameworks, platforms, and tools as they emerge across the internet.
Try the enrichment fields live in our demo, or check our API documentation for integration details and code examples in Python, JavaScript, PHP, and more.