What Is Intent Data? How It Works and What It Costs

What intent data is, where it really comes from, what it costs, and where it falls short. An honest guide for B2B sales and marketing teams.

Hunch Team·July 7, 2026·11 min read

Intent data is behavioral data that shows which companies are researching a product category right now, inferred mostly from the content they consume across the web. Vendors sell it so you can find in-market accounts before those accounts ever contact you. It works, within limits the companies selling it rarely spell out: most of it is aggregated, account-level, and topic-based, and it is priced through annual contracts that almost no vendor publishes.

This guide explains what intent data actually is, where each type comes from, what it looks like in practice, what it costs, and where it falls short. If you are deciding whether to buy it, the last two sections are the ones the vendor explainers skip.

What is intent data?

Intent data is information about a company's online research behavior, used to predict whether that company is in the market for a product like yours. You will also see it called buyer intent data, purchase intent data, or B2B intent data. All four terms mean the same thing.

The core idea: before a B2B team buys anything, its people research. They read comparison articles, download guides, search for category terms, and browse review sites. Individually those actions are invisible to you. Intent data vendors observe them at scale, aggregate them to the company level, and flag when a company's consumption of a topic spikes above its normal baseline.

The word "intent" oversells it slightly. Nobody declared intent. What you actually get is a statistical inference: this company is consuming unusually more content about, say, "sales engagement platforms" than it normally does. That inference is genuinely useful for timing and prioritization. It is not a purchase order, and treating it like one is the most common way teams get disappointed by it.

Where intent data actually comes from

Almost all commercial intent data is collected in one of five ways, and the collection method determines its accuracy, coverage, and compliance profile more than any feature on the vendor's website. This is the part most explainers gloss over, because every vendor's method sounds worse when described plainly.

The biggest co-op is run by Bombora, which reports that 86% of the websites in its data cooperative are exclusive to it. Many platforms you might buy "their" intent data from, including Cognism and several others, are actually reselling Bombora's co-op data under their own interface. That matters when you compare vendors: two products with different logos can be selling you the same underlying signal.

First-party vs third-party intent data

First-party intent data is behavior you observe on properties you own. Third-party intent data is behavior someone else observed elsewhere and sold to you. The distinction matters because it decides both quality and coverage.

First-party signals (pricing page visits, trial signups, repeated documentation views) are the most accurate intent data that exists, because the behavior happens on your site, tied to your product. Their limitation is structural: they only cover companies that already know you exist. If your traffic is small, your first-party intent data is small.

Third-party signals cover the whole market, including companies that have never heard of you. That reach is the entire value proposition, and it comes with a proportional accuracy cost: the behavior was aggregated, anonymized, and mapped to a company through IP-and-cookie inference before it reached you.

There is also a granularity split worth knowing. Most third-party intent data is account-level: it tells you a company spiked on a topic, not who inside the company did the reading. Contact-level intent products exist and cost more, and they concentrate the same inference risk on a named person, which raises both the value and the compliance stakes.

What intent data looks like in practice

In practice, intent data usually arrives as a scored topic feed. Concretely, a weekly delivery might contain rows like these:

• Acme Logistics, topic "warehouse management software", surge score 76, baseline 41, trend up two weeks running
• Northwind Health, topic "HIPAA compliant messaging", surge score 88, first appearance this quarter
• Contoso Financial, topic "your competitor's brand name", surge score 91

Each row says: this company's consumption of this topic is unusually high relative to its own history. Your team then decides which scores are worth an SDR's morning. The topics come from the vendor's fixed taxonomy, typically several thousand predefined topics; Cognism, for example, advertises 14,000+ intent topics from the Bombora co-op it resells. You pick the topics closest to your category, and everyone else in your category picks the same ones.

How sales and marketing teams use intent data

Teams that get value from intent data use it for prioritization and timing, not as a magic lead list. The five uses that consistently pay off:

1. Ranking outbound lists. Given 500 target accounts, work the 40 showing topic surges first. This is the single highest-yield use, because it changes effort allocation without changing anything else.
2. Timing outreach earlier. Third-party signals fire while buyers are still researching, before they have a shortlist. Getting a rep into the conversation at that stage measurably beats arriving after the shortlist forms.
3. Weighting lead scores. Intent adds a behavioral input to fit-based scoring, so a mid-fit account actively researching outranks a perfect-fit account showing nothing.
4. Building ad audiences. Marketing teams sync surging accounts into LinkedIn and programmatic audiences, spending only on companies showing category interest.
5. Spotting churn risk. When a current customer starts surging on competitor topics or "alternatives" terms, customer success gets a warning it would not otherwise have.

What intent data costs

As of mid-2026, almost no major intent data vendor publishes prices. 6sense, Bombora, Demandbase, and ZoomInfo all sell through sales-negotiated annual contracts, which is why "pricing" pages in this category ask for your email instead of showing a number. Public buyer-reported figures consistently put standalone intent platforms in five-figure annual territory, with full ABM platforms that bundle intent (like 6sense and Demandbase) running well into six figures for mid-market and enterprise deployments.

The packaging usually has three parts: a platform fee, a limit on how many intent topics you can track, and per-seat or per-record charges for the contact data you need to act on the signals. The topic limit is the one to watch. Entry tiers often include a handful of topics, and the topics you actually need cost extra.

Two structural reasons explain the opacity. First, the data is cheap to serve once collected, so price tracks willingness to pay rather than cost, and publishing a number would collapse that. Second, intent data is usually sold inside a bigger platform bundle, which makes like-for-like comparison hard by design.

The practical advice: never buy intent data without a defined workflow for who acts on a signal within 48 hours of it firing. Signal feeds that land in an unwatched dashboard are the most expensive kind of shelfware, and at these contract sizes the difference between used and unused is a rep's quota. If your budget does not stretch to enterprise contracts, signal monitoring priced per account exists as an alternative; Hunch runs at $0.75 per monitored account per month with pricing on the page.

Where intent data falls short

Intent data has real, structural limitations, and knowing them in advance is the difference between a tool that focuses your team and a line item that quietly disappoints. Five come up over and over in practitioner reviews:

1. Topics are category-level, not company-level. You cannot track "companies evaluating a switch from our specific competitor's mid-tier plan." You can track "CRM software." Every one of your competitors can, and does, buy the same topic.
2. Account identification is inference. Mapping anonymous web behavior to a company runs through corporate IP ranges and cookies. Remote work broke a lot of that mapping: an employee researching on home Wi-Fi looks like a consumer, not like their employer.
3. Research is not buying. Surges are triggered by competitor employees doing market research, new hires learning the space, students, journalists, and content marketers. A topic spike means attention, and attention has many causes besides purchase intent.
4. The same big accounts light up repeatedly. Large enterprises generate research traffic constantly across thousands of employees, so they surge on everything, always. Practitioners have noticed their intent feeds re-serving the same large accounts month after month. Baseline-relative scoring is supposed to correct for this; it only partly does.
5. You cannot audit a score. When a vendor says a company surged to 82 on your topic, there is no underlying evidence you can read, no article list, no source trail. You are trusting the methodology wholesale, which makes quality differences between vendors nearly impossible to verify before you buy.

None of this makes intent data useless. It makes intent data a probability instrument: it moves attention toward accounts more likely to be in-market, and that is genuinely worth money at the right price. The failure mode is expecting named, evidenced, ready-to-buy accounts from a product that is architecturally incapable of delivering them.

Intent data vs buying signals: which do you need?

Intent data is one category of buying signal, and for many teams it is not the best one to start with. A buying signal is any observable event that correlates with purchase likelihood: an executive hire, a funding round, a hiring surge, a technology change, a pricing page visit. Topic-level research surges are one member of that family. Our guide to buying signals, with 15 concrete examples covers the full taxonomy.

The practical difference is specificity and evidence. Topic-based intent data answers "is this company researching my category?" from a fixed menu of topics, with no evidence trail attached. Signal-based approaches answer questions you define yourself, from public, checkable events: "did this company hire a VP of Sales in the last 90 days?", "did they post three or more RevOps roles this month?", "did they just expand into Europe?". Those events are readable by anyone, which means a rep can open an email with them, and a skeptical buyer can verify them.

This is the gap Hunch was built for. Instead of picking from a topic taxonomy, you describe the signal in plain English, and Hunch checks it daily against the live web, across your account list and across companies you have never named. Every detection comes with dated evidence and sources a rep can actually read, and the people to contact, with verified emails, are included on every plan. Writing a precise signal takes a few minutes; the signals documentation shows how to make one testable.

If your motion depends on catching anonymous category research at enterprise scale, third-party intent data earns its contract. If your motion depends on reaching specific kinds of companies at the moment something changed, custom signals give you evidence, specificity, and a price you can read before you buy. Mature teams often run both.

Frequently asked questions

What does intent data mean in B2B sales?

Intent data means behavioral data indicating which companies are actively researching a product category. It is collected from web content consumption, review-site activity, search behavior, and website visits, then aggregated to the company level. Sales teams use it to decide which accounts to contact first and when, on the theory that active research correlates with near-term buying.

How do you get intent data?

First-party intent data comes free from tools you already run: website analytics, form fills, email engagement, and visitor identification. Third-party intent data is bought from providers such as Bombora, 6sense, ZoomInfo, Demandbase, and G2, almost always on annual contracts. An alternative route is monitoring public buying signals (hiring, funding, leadership changes, technology shifts), which platforms like Hunch automate with per-account pricing.

Is intent data accurate?

Intent data is directionally accurate and precisely unverifiable. Aggregate studies show intent-flagged accounts convert at higher rates than cold lists, so the prioritization value is real. But individual scores cannot be audited, account identification through IP mapping has degraded with remote work, and topic surges are regularly triggered by research that has nothing to do with buying. Treat it as a probability boost, not a verdict.

Is intent data GDPR compliant?

It depends entirely on the collection method. Co-op data gathered under consent-based frameworks, like Bombora's, is generally considered compliant, and vendors reselling it lean on that. Bidstream-derived data is far more contested, because the people whose behavior was captured in ad auctions never meaningfully consented to its resale. If compliance matters to you, ask any vendor to state in writing which collection methods feed their product.

Who are the best intent data providers?

The major third-party providers are Bombora (the largest co-op, resold by many other platforms), 6sense and Demandbase (intent bundled into full ABM platforms), ZoomInfo (intent attached to its contact database), and G2 and TrustRadius (review-site intent). The right one depends on whether you want raw signal, a full ABM suite, or contact data with intent attached. For custom, evidence-backed signals rather than topic scores, signal monitoring platforms like Hunch are the emerging alternative.

What is the difference between intent data and buying signals?

Intent data is a subset of buying signals: specifically, aggregated data about a company's topic-level research behavior. Buying signals is the broader category covering any purchase-correlated event, including executive hires, funding rounds, hiring surges, technology adoption, and actions inside your own funnel. Intent data comes from a vendor's fixed topic list, while buying signals can be defined as specifically as your market requires.