Web Intent (Contact)

Web Intent (Contact-Level)

Contact-level web intent signals identify the specific individual who is actively researching a topic — not just the company, but the actual person.

How It Works

Person-level intent data comes from a different tracking mechanism than company-level: cookies.

When content is consumed on ad-supported sites:

  1. If the visitor's browser has a cookie already tied to a known identity (through prior form fills, logins, or other identification events), the system associates the activity with a specific contact
  2. The cookie connects the content consumption to a real person in the B2B identity graph
  3. The system resolves the cookie to a full contact record: name, job title, company, email, phone, LinkedIn URL

The output becomes: "This specific person is consuming content about Topic Y."

Key Distinction: With company-level intent, contacts shown alongside are chosen based on ICP matching (job title, department). With contact-level intent, the contact was directly tracked via cookie-based identity — you are seeing the actual individual who consumed the content.

Confidence & Limitations

Cookie-based identity resolution is probabilistic:

  • Cookies can be shared, cleared, or misattributed
  • When confidence is high enough, the platform attaches a specific contact to the topic signal
  • Match rates vary by geography and industry (US coverage is strongest)

Data Source

Powered by 5x5 Data / Intentsify Market Pulse combined with the 5x5 B2B Contact Database (254M+ US contacts, 387M+ international).

  • 36,937 B2B topics across 16 categories and 170 subcategories
  • Daily refresh cadence
  • 641M+ total contact records for resolution

Schema

FieldTypeDescription
up_idstringUnique person identifier (5x5 format: 5x5-{hash}) — the join key across all files
first_namestringContact first name
last_namestringContact last name
business_emailstringVerified business email
job_titlestringCurrent job title
seniority_levelstringSeniority (CXO, VP, Director, Manager, etc.)
departmentstringDepartment (Sales, Engineering, Marketing, etc.)
mobile_phonestringMobile phone number
linkedin_urlstringLinkedIn profile URL
company_namestringCurrent employer
company_domainstringCompany website domain
primary_industrystringIndustry classification
company_revenuestringRevenue range
company_employee_countstringEmployee count range
topic_idstringTopic identifier (format: b2b_{number})
topic_namestringHuman-readable topic name
categorystringTop-level topic category
subcategorystringDrill-down category
site_visitor_hashstringHashed cookie/device identifier

Resolution Chain

The full resolution works in three steps:

1. Cookie/Device → Site Visitor Hash
      ↓
2. Site Visitor Hash → UP_ID (5x5 Person ID)
      ↓
3. UP_ID → Full Contact Record
   (name, email, title, company, phone, LinkedIn)
      +
   UP_ID → Intent Topic
   (topic name, category, subcategory)

The up_id field is the universal join key — it connects intent signals to the full B2B contact database.

Example Resolved Signal

{
  "up_id": "5x5-74c35dc98464ff195c11bdab54317151",
  "first_name": "Chrissy",
  "last_name": "Silverman",
  "job_title": "Senior Vice President, Digital Director",
  "company_name": "Bank of America",
  "company_domain": "bankofamerica.com",
  "industry": "Banking",
  "seniority_level": "VP",
  "department": "Information Technology",
  "linkedin_url": "linkedin.com/in/chrissy-silverman-861061b",
  "topic_name": "Cloud Computing",
  "category": "Technology",
  "subcategory": "Cloud Infrastructure"
}

Delivery

MethodDescription
GCS/S3 BucketDaily flat file delivery (Parquet). Intent signals + contact database delivered as separate files, joined on up_id
WebhooksReal-time push of resolved intent signals
APIQuery on demand via REST

Use Cases

  • Hyper-targeted Outreach: Know the exact person researching your category and reach out while intent is hot
  • ABM Campaigns: Build audiences of real individuals showing buying behavior, not just accounts
  • Sales Intelligence: Arm reps with "Chrissy at Bank of America is researching Cloud Computing this week"
  • Lead Scoring: Enrich your CRM with person-level intent signals to prioritize follow-ups
  • Competitive Monitoring: Get alerted when specific contacts at target accounts research competitor products

Companion Databases

Contact-level web intent is powered by joining three datasets — all delivered alongside the intent signals:

B2B Contact Database (the big one)

  • 254M+ US contacts and 387M+ international contacts (641M+ total)
  • Fields: first/last name, business email, programmatic emails, personal emails, job title, seniority level, department, mobile phone, direct number, LinkedIn URL, work history, education history, company name, domain, industry, revenue, employee count
  • Delivered as compressed CSV files across 256+ shards (~80GB+ for US alone)
  • The up_id field in this database is the same up_id in the intent signals — this is how you resolve a topic signal to an actual person
  • Refreshed regularly: 5.9M new profiles, 48M new emails, and 45M new phone numbers added in the most recent update

Company / Firmographic Database

  • Millions of company records with full firmographic detail
  • Fields: company name, domain, phone, industry, SIC/NAICS, address, LinkedIn URL, revenue range, employee count, description, related domains
  • Join on company_domain or cc_id (company-level 5x5 identifier)

IP2Company Database

  • Maps IP addresses to company identities
  • Powers the upstream company-level resolution before cookie-based person resolution kicks in

Topic Taxonomy

  • 36,937 B2B topics with category, subcategory, description, and product flag
  • Delivered as a single CSV alongside the intent data
  • Available in the S3 delivery bucket under current-taxonomy/

How the Files Fit Together

Intent Signals (daily, small)
  └── up_id ──→ B2B Contact Database (256+ shards, ~80GB)
  └── topic_id ──→ Topic Taxonomy (single CSV, ~7MB)
  └── company via contact ──→ Firmographic Database

Sizing note: The daily intent signal files are small (millions of rows but compact). The companion databases are massive bulk files (hundreds of GB total) that you ingest once and update incrementally. Plan your data infrastructure accordingly.