Buyer intent data is behavioral evidence that an account is researching a topic or category. It comes in four types: first-party (your own site, product and emails), third-party (content consumption across publisher networks, sold as "topic surges"), review and comparison activity, and declared intent (forms, demo requests, replies). It shows that an account is looking, not why, who or when.
Used as a confirming input alongside buying signals, it sharpens prioritization. Used as the primary trigger, it produces a sequencer firing at curious accounts. Either way it raises the odds of a conversation; it does not promise a purchase or a date.
How is each type collected?
| Type | How it is collected | Granularity | Freshness | Shared with competitors? |
|---|---|---|---|---|
| First-party behavioral | Scripts on your site and product; email engagement; de-anonymization of IP to company | Account, sometimes person | Real time | No |
| Third-party topic surge | Publisher networks and data co-ops observe content consumption; a vendor aggregates by company and topic and flags spikes against a baseline | Account, topic | Often weekly | Often; networks can overlap |
| Review and comparison | Activity on review sites and category comparison pages, often sold by the review platform | Account, category | Days to weeks | Often |
| Declared | Your forms, replies, event registrations | Person | Real time | No |
Two consequences follow. First-party and declared intent are the most specific and the smallest; they cover the accounts that already found you. Third-party intent is the broadest and the least exclusive; it covers the market but is visible to everyone in it.
What can intent data tell you, and what can it not?
It can: confirm that a business event has turned into research; narrow which function is researching (by topic); flag accounts that are evaluating late in the journey (review activity); and raise borderline accounts into the working set.
It cannot: explain why the research is happening; distinguish a buyer from an analyst, a competitor or a job seeker; judge timing; or name a decision-maker. Those come from buying signals, revenue timing and decision-maker intelligence.
It also helps to separate intent data from the other data you buy. Firmographic data describes what a company is (industry, size, location). Contact data says who works there and how to reach them. Both are fairly static descriptions of an account or a person, while intent data describes behavior and changes week to week. Firmographic and contact data usually sit inside a sales intelligence platform, and intent data is a layer on top. It cannot replace either, since you still need to know who fits your ICP and who to write to.
How is intent data sold and priced?
Intent data is usually sold as a subscription, and the pricing model matters more than any list price. The common models are: contract tiers based on the number of topics monitored, accounts revealed or users with access; per-seat licenses where intent is a feature of a wider ABM or sales intelligence platform; and credit-based plans where you spend credits to reveal accounts or contacts. Annual contracts are common, and many vendors quote rather than publish prices. We do not quote vendor prices here because they change and depend on scope. Before committing:
- 01Validate on closed-won. Ask the vendor to show intent history for your recent closed-won accounts for the 90 days before they entered your pipeline. If few of them surged, the feed is unlikely to find your buyers.
- 02Validate on closed-lost and churn. If those accounts surged just as often, the feed does not discriminate.
- 03Check the topic taxonomy. If your category maps to a broad topic ("sales software"), the surge will include everyone evaluating anything adjacent.
- 04Check the pricing unit. Ask what you are charged for (topics, accounts revealed, seats or credits) and what happens when you exceed it. A cheap entry tier with expensive overage is not cheap.
- 05Check latency. A weekly refresh plus your own weekly review can add up to two weeks of delay on a signal that competitors may see the same week.
A process for using intent data without burning SDR time
- 01Never trigger on intent alone
A surge without a business event goes to a watch list with a dated check, not to a sequence.
- 02Attach every surge to an account file
What changed at this account in the last 90 days? If the answer is nothing, the surge is probably noise or a competitor's campaign.
- 03Score intent as a capped input
Third-party intent can lift a borderline account into the working set, but it should never put an event-free account at the top. The scoring guide sets a cap.
- 04Map the function the topic points at
A surge on 'sales intelligence' maps to sales and RevOps. Seed the committee from the function, not from the CEO.
- 05Write to the event, mention nothing about the surge
Telling a prospect you watched their research says something about your data and nothing about their business.
- 06Measure reply rate for intent-plus-event against event-only
If intent adds nothing to the reply rate in your market, stop paying for it. If it adds a lot, raise the cap.
How does PravRaha use intent data?
PravRaha is built around business events: hiring, funding, leadership change, tech-stack change and expansion. It does not treat third-party intent as a trigger. If you bring intent data to your own process, use it only to confirm an account that already has a business event. The Messaging agent drafts against that event rather than the observed research. The platform page has the details.
More on buyer intent
- →The parent guide: buyer intent
- →A scoring rubric: buyer intent signals and scoring
- →The causal inputs: buying signals
See how the loop runs inside PravRaha.
Signals, intelligence, prediction, timing, decision-makers and action, as one system rather than eight tools.