Short answer
Intent data is information about research behavior that suggests a company or person may be in the market for a product, such as reading articles on a topic, comparing vendors on a review site or visiting your pricing page. It comes from your own website (first-party), review sites and publisher networks (third-party). Sales and marketing teams use it to decide which accounts that fit their ideal customer profile to contact first.
- Most intent data is account-level: it tells you a company is researching, not who.
- Third-party surge scores are relative to each company's own baseline, not absolute counts.
- Combine intent with fit. A surging account outside your ICP is still a poor prospect.
- Test it with a holdout group before you trust a vendor's lift figures.
- Since 2023, California's privacy law covers B2B contact data, so ask vendors how they collect it.
On this page
This page is for marketing and sales leaders deciding whether to buy intent data, and for SDRs and RevOps teams who have it and want to use it well. It covers the types, how the scores work, a worked test, the privacy rules and the questions to ask a vendor.
What intent data means
Intent data is behavioral information that suggests a company or person is researching a purchase. The behavior is the evidence: reading about a problem, comparing vendors, visiting a pricing page. The intent is an inference from that behavior, and it can be wrong.
Two things to know from the start. First, most third-party intent data is reported by company, not by person, so it tells you that someone at an account is researching, not who. Second, it's a timing signal. It helps you choose which accounts to contact this week. It doesn't tell you which accounts are worth contacting at all; that's your ideal customer profile.
Types of intent data
| Type | Where it comes from | What it shows | Level | Main limit |
|---|---|---|---|---|
| First-party | Your website, emails, product and CRM | Visits to pricing or comparison pages, content downloads, replies | Person when known, otherwise company by IP | Only shows people who already found you |
| Review-site | Software review and comparison sites | Companies viewing your category, your profile, competitors or pricing | Company | Mostly software categories |
| Third-party co-op | Networks of B2B publishers and sites sharing research activity | Companies reading more than usual about a topic | Company | Topic matches can be loose; relative scores |
| Bidstream | Data passed in online ad auctions | Pages viewed by devices linked to companies | Company | Weakest provenance and consent trail |
| Public signals | Hiring, funding, new sites, leadership changes, filings | Events that often come before a purchase | Company | An event isn't research; needs interpretation |
Public events are often called buying signals or trigger events rather than intent data, but many tools mix them into one score. Ask which you're getting.
How a third-party surge score works
Third-party providers watch research activity across a network of sites, map each page to topics, and flag an account when its reading on a topic rises above its own normal level. Bombora, one of the largest, describes comparing "the most recent three weeks of activity against a 12-week historical baseline" for each account and topic, across more than 25,300 topics.
The baseline matters. A 10,000-person company reads more about everything than a 50-person one, so raw counts would always favor big companies. A relative score fixes that, but it also means a small company can surge on a handful of page views. Read a surge as "more than usual for them," not "a lot."
Review sites work differently. G2 describes its buyer intent as company-level signals from people researching on G2 itself, such as "category research, competitor comparisons, pricing page visits." That's closer to an active evaluation, but it only covers buyers who use review sites.
Worked example: testing intent before you trust it
Illustrative example. The company and figures are invented to show the method and arithmetic.
Trackwell (trackwell.example) sells fleet telematics. It has 800 accounts that fit its ICP: delivery, field service and distribution companies running 50 to 500 vehicles. Its intent provider flags 160 of them as surging on topics such as fleet tracking, ELD compliance and fuel management.
To test whether the flag helps, Trackwell splits the work:
| Group | Accounts | Treatment | Meetings | Meeting rate |
|---|---|---|---|---|
| Surging, contacted | 100 | Same sequence | 9 | 9% |
| Not surging, contacted | 100 | Same sequence | 4 | 4% |
The surging group booked meetings at more than twice the rate. But with 100 accounts per group, a gap of 9% against 4% could still be chance: a standard two-proportion test gives a p-value of about 0.15. To detect a difference that size reliably (80% power at 5% significance), Trackwell needs about 380 accounts per group. With only 160 surging accounts this month, it keeps the test running as new accounts surge and pools the results before changing how it routes accounts.
Three rules make the test fair: pick both groups from accounts that fit the ICP, give them the same sequence and reps, and decide the success measure (meetings, not opens) before you start.
Intent data vs related terms
| Term | What it is | How it relates to intent data |
|---|---|---|
| Firmographics | Company facts: industry, size, location, revenue | Decides fit; intent decides timing |
| Technographics | The technology a company uses | A fit signal; a change in tools can be an intent signal |
| Buying signals | Events and behaviors that suggest readiness to buy | The broader category; intent data is one source |
| Lead scoring | A score that ranks leads by fit and engagement | Intent is often one input to the score |
Limits and rules to know
- Noise. Students, journalists, job seekers and your competitors read about your topic too. Account-level data can't tell them apart.
- Lag and decay. A three-week window means a surge can reflect research that's already turned into a shortlist. Act on new surges within days, and let old ones expire.
- Topic fit. Check the exact topics you subscribe to. A broad topic such as "logistics" will surge for companies that will never buy telematics.
- Privacy law. California's exemptions for business-to-business personal information expired on 31 December 2022, according to the California Privacy Protection Agency, so the CCPA now covers B2B contact data about California residents for businesses that meet its thresholds. In the EU and UK, GDPR applies to business contacts too. Ask every provider how its data is collected and what consent sits behind it.
Questions to ask an intent data vendor
- Where exactly does the data come from: your own network, partners, review sites or ad auctions?
- Is it reported by company or by person, and how do you match activity to a company?
- How is a surge calculated, against what baseline, and how often is it updated?
- Which topics would we track, and can we see the pages that map to them?
- What share of our ICP accounts can you match at all?
- How is consent collected, and how do you handle CCPA and GDPR requests?
- Will you support a holdout test before we sign a long contract?
How to use intent data in practice
- Start from fit. Apply intent only to accounts that match your ICP.
- Pick five to ten narrow topics that someone close to buying would research.
- Route surging accounts to a rep within days, with the topics attached so the first message is relevant.
- Find the person yourself. Intent says the company is researching; you still need the right contact.
- Measure meetings against a holdout, and drop topics that don't lift results.
If you want signals joined to your own CRM, Bebop Insights delivers firmographic, technographic and signal data by API or bulk file.
Next step: see the wider set of events worth acting on in buying signals.
Sources
- Bombora Company Surge: B2B intent data, Bombora, accessed 26 October 2026
- Buyer Intent Data, G2, accessed 26 October 2026
- Frequently Asked Questions (FAQs), California Privacy Protection Agency, accessed 26 October 2026
- What Is Intent Data?, Boomi, 29 February 2024
Written by the Bebop team at Blackpearl Group. Last updated .
Questions people ask
What is an example of intent data?
A distribution company whose staff read several articles about fleet tracking and ELD compliance this month, well above their normal level, would show up as surging on those topics in a third-party intent feed. Other examples are a company comparing vendors in your category on a review site, or a visitor from a target account returning to your pricing page three times in a week.
What is the difference between first-party and third-party intent data?
First-party intent data comes from your own channels: website visits, content downloads, email clicks and product use. It's accurate but only covers people who already know you. Third-party intent data comes from outside sources such as publisher networks and review sites, and shows companies researching your topic before they visit you. It reaches further but is usually company-level and less precise.
Is intent data legal?
Generally yes, but it depends on how the data was collected and where the people are. California's exemptions for business-to-business personal information expired at the end of 2022, so the CCPA now applies to B2B contact data about California residents for businesses that meet its thresholds. GDPR covers business contacts in the EU and UK. Ask each vendor how consent is collected and how it handles deletion requests.
How accurate is intent data?
It varies by source and isn't accurate in the sense of predicting a purchase. First-party and review-site signals reflect activity tied to your product or category. Third-party surges are inferred from reading behavior and include noise from students, job seekers and competitors. Treat intent as a way to choose which fitting accounts to contact first, and measure the effect with a holdout test.
What is the difference between intent data and buying signals?
Buying signals is the broader term: any event or behavior that suggests a company may be ready to buy, such as a funding round, new location, leadership change, job posting or research activity. Intent data usually means the research behavior specifically, collected from websites, review sites and publisher networks. Many tools combine both into one score, so check which signals a score is built from.
How do sales teams use intent data?
They use it to decide which accounts to contact first. A typical setup filters intent to accounts that fit the ideal customer profile, sends surging accounts to the owning rep within a few days with the topics attached, and uses those topics to make the first message relevant. Good teams also keep a holdout group so they can see whether intent actually lifts meeting rates.