30-second answer: Intent data is signal data that indicates a B2B account is researching a problem your product solves. The vocabulary breaks into source classes (first-party, second-party, third-party, predictive), signal types (content consumption, search, technographic shift, funding, hiring), processing concepts (surge, decay, baseline, signal merge), and operating terms (in-market, signal-based selling, account-level identification, intent score). This glossary defines the 22 terms a B2B revenue team needs to read intent vendor documentation and run an in-market account play.
See intent data combined with account orchestration inside Abmatic AI, book a demo.
First-party intent is signal data captured on properties the vendor owns: their website, their content portal, their product, their events. It is the highest-trust class because the activity happened with the vendor itself. First-party signals carry full attribution context and timestamps, and account resolution is straightforward because the vendor controls the form, the cookie, and the IP capture. See first-party intent data.
Second-party intent is signal data shared by a partner under a data-sharing agreement. The most common case is a publisher sharing reader-level activity with a vendor whose category is being researched. Second-party is rarer than first or third-party but valuable because the partner's audience is relevant by selection.
Third-party intent is signal data derived from research activity outside the vendor's properties, aggregated across publisher networks, review sites, and content syndicators. It is the widest net but the lowest trust because the inference is at the account level rather than observed directly.
Predictive intent applies machine learning to combine first-party, third-party, and firmographic signals into a forward-looking propensity score. Predictive scores compress noisy signal streams into a single rank-orderable number. See predictive intent data.
Content consumption tracks which articles, whitepapers, webinars, and videos an account reads or watches. Topic clusters around buying-stage content (RFPs, comparisons, pricing) signal late-stage research; awareness-stage content signals early research.
Search signals are queries an account ran on search engines or category review sites. Branded searches for competitor products are a particularly high-trust late-stage signal.
A technographic shift is a measurable change in an account's technology stack, such as a new CRM going live or a security tool being deprecated. Stack shifts often precede category purchases.
A funding signal is the announcement of a new round of capital raised by an account. Funded accounts often expand spend in the 60-to-90 day window after announcement.
A hiring signal tracks job posts at an account. Hiring for a role that uses your product (a head of revenue operations, a security engineer, a paid media manager) signals a budget allocation has been approved.
Surge describes a multi-day increase in topic-level intent activity at an account, measured against the account's own historical baseline. Surge models distinguish in-market spikes from steady-state research noise.
Baseline is the historical activity rate for an account, used as the comparison floor for surge calculation. Without baseline, every account looks active or inactive in absolute terms; with baseline, surge is the fact.
Decay applies a half-life to intent signals so that recent activity weighs more than older activity. Most intent platforms decay signals over 14 to 30 days because B2B research velocity is high.
Signal merge combines first-party, third-party, predictive, and firmographic signals into one account-level score, resolving duplicates, conflicts, and decay. See signal merge.
An intent score is the single number a platform produces per account per topic, summarizing surge, decay, and baseline into a rank-orderable signal. Most platforms expose 0-to-100 scores or letter grades.
Account-level identification resolves anonymous web traffic and aggregated third-party activity to specific named companies. Without it, intent data cannot route to sales because the account is not known.
Reverse IP lookup resolves a website visitor's IP address to the company that owns it, enabling first-party identification of anonymous traffic at the account level. See reverse IP lookup.
Identity resolution stitches signals from cookies, IPs, form submissions, and CRM into a single account or person record across systems. It is the backbone of cookieless intent attribution.
An in-market account is one whose intent signals exceed a threshold indicating active research in the category. In-market is a label, not a guarantee. See identify in-market accounts.
Signal-based selling triggers outbound from real-time buying signals rather than working static lists. The motion compresses time-to-touch from days to minutes and is the dominant outbound pattern in 2026.
A topic is the category or keyword cluster an intent platform tracks, such as cybersecurity training or marketing automation. Vendors publish topic taxonomies in their documentation; matching the right topics to your category is a core configuration step.
Bombora Surge is a specific intent product widely used as a third-party signal source. The term is sometimes used loosely as a synonym for any third-party surge, but it refers to a specific vendor's data set.
G2 Buyer Intent is review-site signal data sourced from category, vendor, and comparison page traffic on G2. The signal is high-fidelity for late-stage research because review-site traffic correlates strongly with shortlisting.
An intent stack is the combination of first-party, third-party, predictive, and review-site signal sources a revenue team uses, plus the unification layer that merges them. See best intent data platforms for vendor coverage.
Ready to see merged intent signals routing to sales in real time? Book a demo of Abmatic AI.
Contact data is who works where (name, title, email, company). Intent data is what those companies are researching right now. Both are useful, and they answer different questions. Contact data scopes outreach. Intent data prioritizes outreach.
Third-party intent has a known false-positive rate because it infers account-level research from publisher and review-site activity that may not be specific to one vendor. The right use is as a first-pass filter that surfaces accounts to investigate, not as a closing signal on its own. Pair with first-party signals before routing to sales.
Thresholds are platform-specific and category-specific. A practical starting point is to look at the top decile of accounts ranked by score, sample 20, and have sales review whether those accounts feel in-market. Adjust the threshold up or down based on the qualitative read. See how to use intent data.
Small companies with high contract values benefit from intent data because the cost-per-account of getting it wrong is large. Small companies with low contract values usually do better starting with first-party signal capture (reverse IP lookup, form submissions) before layering in paid third-party feeds.
Most leading platforms refresh signals daily. For signal-based selling, anything older than 48 hours has lost most of its routing value because the buying committee has moved on.
Intent data is not directly attributable to revenue. It is a routing input, not a touch. Attribution credits the touches that intent data enabled (the sales call, the personalized ad), not the intent signal itself.
Intent data vocabulary varies by vendor. When reading intent platform documentation, always trace the signal back to its source (publisher network, review site, first-party tag) and its decay window. The same intent score from two vendors can mean very different things. Use this glossary to compare like with like.