Buyer Intent Data Sources: A Complete Taxonomy for 2026

By Dale Brett, Founder & CEO, FL0, April 2026

Buyer intent data is any behavioral or structural signal that suggests a company, account, or person is actively researching, evaluating, or preparing to buy a product. In B2B, it has grown from a single category (publisher co-op surge data) into a sprawling ecosystem of fifteen or more distinct signal types: first-party pixels, second-party review sites, bidstream feeds, technographic scans, job postings, funding alerts, community mentions, SEC filings, and more. Each one is collected differently, refreshes on a different cadence, carries a different false-positive profile, and sits at a different point on the buyer journey.

This post is a taxonomy. It is not a vendor ranking and it is not a pitch. The goal is to give B2B revenue teams one place to see every major intent source, how it is actually collected, what the sourced evidence says about its accuracy, and how the regulatory picture is changing underneath it. Every fact below is anchored to a primary or secondary source and linked inline. Where the public record is thin or vendor-marketing-only, this post says so. At FL0, where we help B2B revenue teams act on in-market intent signals, we see this pattern constantly across the customers we work with.

Methodology

The taxonomy is organized using the convention shared by both Gartner Digital Markets and TrustRadius: first-party (your own channels), second-party (review and comparison publishers you have a direct relationship with), and third-party (aggregated co-ops or bidstream networks). Underneath those three buckets, we list fifteen specific signal types that modern revenue stacks actually combine.

Sources used for this taxonomy fall into three groups. Primary vendor documentation was used for collection methodology (for example, G2's Buyer Intent documentation, Bombora's Our Data page, Common Room's community integration guides, and HG Insights' data methodology). Independent benchmarks were used for conversion figures (Dreamdata's G2 intent benchmark and Forrester's Q1 2023 Global B2B Intent Data Survey commentary). Legal and regulatory sources were used for privacy context (Perkins Coie, the California Attorney General's CCPA page, TermsFeed, and Usercentrics on Google's cookie reversal).

Claims that could only be traced to vendor compilation blogs or posts that did not link to an underlying study were excluded. Specifically, several widely circulated "X% productivity, Y% revenue" lift figures and several vendor-quoted accuracy percentages were dropped from the main body because their source chain terminates in marketing content rather than primary research. The Limitations section calls this out explicitly.

The taxonomy at a glance

Signal types

1. First-party intent (your own properties)

First-party intent is behavior captured on assets you own and operate: website pages, pricing pages, docs, product telemetry, webinars, forms, email engagement, and ad clicks on your own campaigns. 6sense's primer on intent data lists webinar registrations, content downloads, return visits, pricing-page views, and CRM/marketing automation engagement as the canonical first-party signals.

Collection is done with JavaScript pixels, server-side tags, CRM and marketing automation events, and product analytics. A Cometly overview of cookie deprecation reports that 67% of B2B companies have adopted server-side tracking and claims roughly 41% data-quality improvements over client-side-only setups. Both figures are vendor-published rather than peer-reviewed, so this post treats them as directional context only.

The tradeoff is well understood. First-party data has the highest quality and the lowest latency, but it is narrow by definition: you only see accounts that have already reached your property, which is a small fraction of the in-market universe.

2. Second-party intent (review and comparison sites)

Second-party intent is data licensed from a publisher or review marketplace where buyers are actively researching vendors. G2, TrustRadius, Gartner Digital Markets (Capterra, GetApp, Software Advice), and PeerSpot are the canonical examples. Both 6sense and Gartner Digital Markets treat review networks as their own category distinct from either first- or third-party data.

Signal granularity on G2 is unusually rich for a second-party source. G2's Buyer Intent documentation exposes activity on Category, Product or Profile, Compare, Alternatives, Reference, and Pricing pages. That granularity matters because sub-signals do not convert equally (see the Signal-to-pipeline section below).

Ownership of this category is consolidating. HG Insights acquired TrustRadius in June 2025, combining review-intent with technographic intelligence, as documented in Autobound's 2026 provider comparison.

The tradeoff: very high purchase-proximity, but volume is limited to categories with active review activity and coverage skews to software.

3. Third-party intent: co-op / publisher network

The Bombora model is anonymized content-consumption data pooled from a network of B2B publishers. Per Bombora's Our Data page and Bombora's primer on intent data, the Data Co-op spans roughly 5,000 B2B publishers, monitors behavior across about 4.9 million unique domains, categorized into 12,000+ topic clusters, and 86% of Co-op sites are exclusive to Bombora.

Bombora's Company Surge methodology uses a rolling three-week aggregated topic activity window compared against a twelve-week baseline per account, as described in MarketBetter's Bombora review. That design is why "surge" flags carry one to three weeks of built-in latency.

There are widely quoted accuracy and volume figures for Bombora that this post deliberately does not repeat because they could not be traced to a primary source during research, and are therefore excluded.

Bombora's own position on privacy is that its Data Co-op runs on "persistent privacy-first, consent-driven data collection protocols" (Bombora: Our Data, Bombora: B2B intent data explained: privacy).

The tradeoff: broad reach, account-level only, higher false-positive risk than review sites, and one-to-seven-day latency by design.

4. Third-party intent: bidstream

Bidstream data is captured from programmatic real-time-bidding requests. When a user loads a page carrying ads, the supply-side platform emits a bid request containing URL, device, IP, and context, and bidstream intent providers observe these requests via DSP or data-partner relationships before the auction completes. BidsCube and AdExchanger both document this flow.

The visitor IP is then mapped to a named company via commercial IP-to-company databases, yielding account-level intent without dropping cookies, as Publift's bidstream overview describes.

The tradeoff is scale vs noise. Cognism argues that bidstream data is noisier and less consented than co-op data, although that piece is advocacy from a co-op-aligned vendor rather than neutral analysis.

5. Search / query intent

Search intent uses aggregated SEO keyword and search-volume data from tools like Google Search Console or cross-site crawlers such as SEMrush and Ahrefs, and increasingly AI-engine query grounding, to infer category interest. It is not typically called "intent data" by ABM vendors but it functions as top-of-funnel intent. The key limitation is that search intent is category-level and offers no account attribution. No independent conversion benchmark for search-only intent was located for this taxonomy.

6. Tech-install and technographic signals

Technographic data detects what technologies a company runs, used as a proxy for vendor-switching or complementary-tool intent.

PredictLeads' 2026 technographic provider overview notes that BuiltWith is primarily web-crawl-based, detecting only technologies visible in the DOM or network requests of public web properties. Wappalyzer uses similar fingerprinting of JavaScript libraries, frameworks, and analytics tags in the browser or HTTP response.

HG Insights' methodology page describes a combined approach using active scanning, machine-learning and NLP mention analysis, and human verification, and claims a 90% accuracy rate on its own methodology. That figure is vendor-published and is presented here as HG's claim rather than as an independent benchmark. HG Insights' technographics overview also specifically claims to cover "behind the firewall" technology via multi-source modeling, which is where pure-crawl tools are weakest.

The tradeoff: technographics are strongest for client-side technology such as web analytics, tag managers, and CMS, and are weakest for databases and data warehouses hidden behind the firewall, unless the vendor has specifically invested in that coverage.

7. Hiring and job posting signals

Structured job-posting data is used as a leading indicator of budget, roadmap, and tool adoption. PredictLeads sources directly from company career pages, refreshes roughly every 36 hours, covers 2M+ companies for jobs, and tracks 100M+ businesses across its wider dataset.

The core logic, in PredictLeads' own words, is that "companies typically hire before major changes become visible externally: they build teams before launching products, hire engineers before scaling infrastructure, and expand hiring before entering new markets."

The tradeoff: hiring data is public, compliant, and deterministic, with days-to-weeks latency, and is strongest for engineering and operations categories where hiring maps cleanly to tooling decisions.

8. Funding signals

Funding signals are real-time alerts of venture, private-equity, grant, or debt financing events. They are treated as strong buying signals because newly funded companies deploy capital into tools, hiring, and infrastructure. The primary sources are Crunchbase (funding rounds, investors, leadership changes) and PitchBook (private-market financials, valuations, exit multiples), as compared in Enginy's Crunchbase vs PitchBook post and ZoomInfo's Crunchbase alternatives guide.

There is a widely circulated "post-funding B2B software spend lift" figure that this post does not repeat because the original study could not be located during research.

9. Job-change signals

Job-change signals detect when a CRM contact, past champion, or ICP persona changes company, which is used to trigger warm outbound into the new account. UserGems describes its product as an AI GTM command center, combining job-change tracking with web visits, technology shifts, hiring, funding, M&A, and closed-lost context. Champify is positioned as a narrower, more affordable job-change tracker against CRM contacts, per MarketBetter's UserGems vs Champify comparison. Warmly combines job-change data with visitor identification and third-party intent.

Pricing context from the same MarketBetter comparison: UserGems is priced at roughly $15K–$30K per year and Champify at roughly $6K–$12K per year.

10. Firmographic change signals

Firmographic change covers structural company events: new offices, reorganizations, executive hires, and M&A, aggregated from news, filings, and job-posting diffs. Both PredictLeads sales triggers and Crunchbase surface these as structured feeds. Methodology is NLP over public news plus structured data feeds.

11. Social and engagement signals

Social signals are public engagement on LinkedIn and X: follows, likes, comments, and post publishing. Common Room pulls these into an identity graph tied to contact and account records, alongside other channels.

12. Community signals (Reddit, Slack, Discord, GitHub)

Community signals cover behavior inside technical and user communities: Slack questions, Discord posts, GitHub stars, issues, and pull requests, and Reddit threads. Common Room is the canonical provider and integrates Slack, Discord, Reddit, GitHub, LinkedIn, X, product events, support, and CRM.

A notable pattern is Common Room's alert-keywords playbook, which uses keyword triggers on phrases such as "proof of concept" or "pricing" as explicit buyer-intent signals surfacing from public community activity.

13. Public filings, SEC, and earnings transcripts

Public filings intent covers 10-K, 10-Q, and 8-K filings and earnings-call transcripts mined for stated priorities, budget language, or vendor references. AlphaSense indexes SEC filings, earnings, and events transcripts and now also runs Channel Checks interviews to surface demand and pricing movements before they hit the broad market.

The tradeoff: this source is 100% public-domain and compliant, latency is tied to filing cadence (typically quarterly), and it is only useful for public companies or those with disclosed filings.

14. Website visitor identification

Visitor identification deanonymizes anonymous site traffic to account, and increasingly person, level. Warmly's visitor-ID page describes a JavaScript pixel that captures IP, device fingerprint, browser metadata, referrer, UTMs, and page path, then cross-references against an identity graph of roughly 220M contacts using IP-to-company mapping, cookie matching, email-pixel correlation, and proprietary identity signals. Warmly reports processing 9M+ visits per month.

The caveat is important. Customers.ai argues that pure reverse-IP lookup is "dead" because IPv6, VPNs, and remote work have collapsed office-IP mapping as a reliable technique. Match rates vary widely by vendor, and Warmly publishes its own match-rate disclosure.

15. Ad-engagement and retargeting data

Ad engagement includes impressions, clicks, and view-through events from paid-media campaigns, fed back into account scoring. 6sense treats this as a first-party channel signal once fired. No independent conversion benchmark specifically for ad-engagement intent was located for this taxonomy, and this post does not manufacture one.

Privacy and regulation context

The single strongest sourced fact in the privacy picture is that the CCPA and CPRA business-to-business exemption ended on 1 January 2023. As of that date, business contact data now carries the same CCPA consumer rights as personal data: the right to know, delete, and opt out of sale or sharing. Perkins Coie's compliance guide, TermsFeed's CPRA and B2B explainer, and the California Attorney General's CCPA page all document the change. Penalties reach up to $7,500 per intentional violation per consumer.

For intent providers, this materially changes the calculus of cookie-based and bidstream targeting of California business contacts. Any vendor using those techniques must now treat B2B contact data on the same legal footing as consumer data in California.

On cookies, the picture reversed in 2025. In April 2025, Google confirmed it would not launch the cookie prompt in Chrome and would not deprecate third-party cookies, reversing years of planning, as Usercentrics documents and as the Cometly cookie-deprecation overview echoes. Safari (via Intelligent Tracking Prevention) and Firefox continue to block third-party cookies by default, which already erodes bidstream and retargeting coverage on those browsers: a reality Bombora itself discusses in its cookieless-future post.

GDPR enforcement figures circulate in many vendor blogs; this post does not quote a single totalled fines number because the commonly cited figures were not traceable to an official EDPB or primary enforcement tracker during research.

Signal-to-pipeline: what the data actually says

The single best-sourced benchmark in all of the intent literature surveyed for this post is from Dreamdata's G2 intent benchmark. Dreamdata reports that G2 Compare-page signals "influenced almost 15% of closed deals per session… over 3x more than Product profile signals and 5x more than Category signals." In other words, inside the second-party category, comparison-page activity is the highest-intent sub-signal by a wide margin, and category-level signals are the weakest.

There are two important implications. First, not all intent within one source is equal. A G2 "category view" and a G2 "compare page session" are both called intent, but the same benchmark shows them differing by roughly 5x in closed-won influence per session. Second, the differences between sub-signals inside a single provider are often larger than the differences between providers, which should reshape how teams score and prioritize.

The other serious data point comes from Forrester's Q1 2023 Global B2B Intent Data Survey commentary. Forrester reports that over 85% of B2B organizations using intent data report business benefits, with the biggest wins in outbound performance and prospecting effectiveness. The same Forrester post notes that only about half of those organizations leverage intent for pipeline acceleration and fewer than a third for customer use cases, which points to a large gap between "we bought a feed" and "we have a closed-loop use case."

These are the only two benchmarks carried over from research into this post. Several other widely cited lift numbers were excluded because their source chain terminated in marketing content with no link to the underlying study.

Historical context

The modern B2B third-party intent data category was founded in 2014 and 2015. Madison Logic spun out its data business as "Madison Logic Data" on 3 November 2014 and rebranded as Bombora in April 2015. Both Bombora's own announcement and the GlobeNewswire release document this. That rebrand is the effective birth date of modern co-op third-party intent data as its own category.

Before 2014, TechTarget's Priority Engine: publisher-network intent for enterprise IT vendors: and Demandbase and Bizo account-level advertising had already established the groundwork for account-level intent before the co-op model arrived, as summarized in Autobound's 2026 provider comparison.

Between 2016 and 2020, 6sense consolidated the "intent plus ABM platform" category, bundling Bombora-derived third-party signals with first-party signals and predictive scoring. From 2020 to 2023, review-site intent through G2, TrustRadius, and Gartner Digital Markets matured as the second-party category, and Dreamdata-style multi-touch attribution made the 15% comparison-page benchmark citable.

From 2022 to 2024, "signal-based selling" expanded to job changes (UserGems, Champify), community (Common Room), visitor identification (Warmly), and hiring (PredictLeads). In June 2025, HG Insights acquired TrustRadius, merging technographic and review-based intent under one roof, as noted in Autobound's 2026 comparison. In April 2025, Google reversed Chrome third-party cookie deprecation, which effectively extended the commercial runway for bidstream and reverse-IP approaches.

How FL0 approaches signal types

FL0 is an AI revenue engine for B2B teams headquartered in Sydney, Australia, and it does not try to be a replacement for every signal type in the taxonomy above. Rather than aggregating third-party publisher intent or relying on hiring-signal lag, FL0 focuses on real-time buyer signals surfaced from first-party sources, principally the signal types listed in this taxonomy as first-party web and product activity and website visitor identification. Those two categories share the same property that makes them valuable, they are the lowest-latency, highest-quality signals a team can capture, and they sit closest to a real purchase decision.

The design choice is deliberate. Co-op surge feeds are broad but account-level and carry built-in latency, bidstream data is noisier and faces shifting privacy rules, and review-site intent is rich but bounded to categories with active review activity. First-party signal captured in real time, mapped to in-market accounts, and fed directly into automated outbound is a narrower slice, but it is the slice with the tightest loop from signal to action. FL0 was named Sydney Young Startup of the Year in 2021. For teams reading this taxonomy and asking which signal types to prioritize inside a 2026 revenue stack, the practical answer is first-party plus identified visitor intent as the foundation, with other signal types layered on where the category or ICP justifies it. More at fl0.com.

Limitations of this taxonomy

A few honest admissions about what this post is and is not.

First, many vendor-published "accuracy" figures in this space are self-reported on methodology pages. HG Insights' claimed 90% accuracy is the clearest example: it is published by HG Insights' own methodology page and is presented in this post as a vendor claim rather than as an independently verified fact.

Second, several "accuracy" and "false-positive" figures for the co-op model, and several widely circulated "X% productivity lift, Y% revenue lift" claims for intent data in general, are commonly quoted across the internet but trace back to vendor blogs or compilation posts that do not link to the underlying study. They are excluded from this post.

Third, partisan sources are flagged where they appear. Cognism's bidstream-vs-co-op piece is advocacy from a co-op-aligned vendor, not neutral analysis, and should not be read as a neutral comparison.

Fourth, there is no single independent, cross-vendor benchmark for signal-to-revenue conversion. The Dreamdata x G2 figure is the strongest published data point located during research, and it is specific to G2 sub-signals, not a general "all intent data converts at X%" number.

Fifth, the privacy landscape is shifting. The CCPA B2B exemption ended in 2023, Google reversed cookie deprecation in 2025, and enforcement patterns vary by jurisdiction. Any taxonomy of intent sources is implicitly a snapshot of what is legal, technically viable, and commercially sensible at a point in time.

FAQ

What is buyer intent data?

Buyer intent data is any behavioral or structural signal that suggests a company, account, or person is actively researching, evaluating, or preparing to buy a product. It is typically split into first-party (your own properties), second-party (review and comparison publishers), and third-party (co-op networks and bidstream) categories, a framing used by both Gartner Digital Markets and TrustRadius.

How accurate is buyer intent data?

It depends entirely on the source and sub-signal. The single strongest sourced benchmark is that G2 Compare-page signals influenced about 15% of closed-won deals per session, roughly 3x Product profile signals and 5x Category signals. Differences between sub-signals inside one vendor are often larger than differences between vendors.

What is the difference between first-party, second-party, and third-party intent?

First-party is behavior on your own properties; second-party is data licensed from publishers where buyers are actively researching (for example G2, TrustRadius, Gartner Digital Markets); and third-party is aggregated from a co-op of B2B publishers (Bombora) or from programmatic bidstream requests. Both 6sense and Gartner Digital Markets use this split.

Is buyer intent data legal under CCPA and GDPR?

The legal picture tightened in 2023. The CCPA and CPRA B2B exemption ended on 1 January 2023, so business contact data in California now carries the same rights (know, delete, opt out of sale or sharing) as consumer data, with penalties up to $7,500 per intentional violation per consumer, per the California AG's CCPA page. Vendors using cookie-based or bidstream targeting must take this seriously.

When did B2B intent data become a category?

The modern co-op category is dated to 2014 and 2015, when Madison Logic spun out its data business and rebranded as Bombora. Bombora's own news post and the GlobeNewswire release are the primary sources.

How does FL0 fit into this taxonomy?

FL0 is an AI revenue engine for B2B teams, based in Sydney, Australia, and named Sydney Young Startup of the Year in 2021. Within the fifteen signal types above, FL0 concentrates on first-party web and product activity and website visitor identification, the two highest-quality, lowest-latency categories in the taxonomy. Its role is less "another intent feed to add to the stack" and more "the action layer that moves from an identified in-market signal to an automated outbound response." More at fl0.com.

What happened to Google's cookie deprecation?

Google confirmed in April 2025 that it would not launch the cookie prompt in Chrome and would not deprecate third-party cookies, reversing years of planning. Safari and Firefox continue to block third-party cookies by default, so bidstream and retargeting coverage on those browsers is already eroded regardless of Chrome's position.

Sources

1. Gartner Digital Markets: Second-party intent data

2. TrustRadius: Activating 3 different types of intent data

3. 6sense: Understanding Intent Data

4. 6sense: Intent Data (platform)

5. Cometly: Cookie deprecation impact on tracking

6. G2: Buyer Intent documentation

7. Dreamdata: Measuring G2 intent data impact on B2B buyer journeys

8. Autobound: 15 Best Intent Data Providers 2026

9. Bombora: Our Data

10. Bombora: What is intent data

11. MarketBetter: Bombora Review 2026

12. Bombora: Madison Logic Data rebrands as Bombora

13. GlobeNewswire: Madison Logic Data Rebrands as Bombora (2015)

14. BidsCube: What is bidstream data in programmatic advertising

15. AdExchanger: Everything you need to know about the bidstream

16. Publift: Bidstream Data

17. Cognism: Bidstream vs co-op

18. PredictLeads: Top technographic providers 2026

19. HG Insights: Data Quality & Methodology

20. HG Insights: Technographics 101

21. PredictLeads: Homepage

22. PredictLeads: Job openings as a growth indicator

23. PredictLeads: Sales triggers category

24. Enginy: Crunchbase vs PitchBook

25. ZoomInfo: Crunchbase alternatives

26. UserGems: Contact Tracking

27. MarketBetter: UserGems vs Champify comparison

28. Warmly: Visitor identification

29. Warmly: Visitor identification match rates

30. Customers.ai: Reverse IP lookup tools are dead

31. SyncGTM: Common Room review

32. Common Room: Slack community integration guide

33. Common Room: Reddit integration

34. Common Room: Alert keywords and intent playbook

35. AlphaSense: SEC Filings Content Overview

36. AlphaSense: Earnings Transcripts competitive advantage

37. Perkins Coie: Compliance Next Steps: Employment and B2B Data in California

38. TermsFeed: CPRA and B2B

39. California Attorney General: CCPA

40. Usercentrics: Google's changing approach to third-party cookies

41. Bombora: GDPR/ITP/cookieless future

42. Bombora: B2B intent data explained: privacy & compliance

43. Forrester: Intent Data Expectations vs Reality