Intent data is any behavioral signal that suggests a person or a company is looking into what you sell, before they ever contact you. That is the whole definition. Everything else is a question of where the signal comes from, how fresh it is, and whether you can act on it today.
I run a LinkedIn prospecting agency. My team contacts thousands of profiles every month for our clients, so I read this subject from the side of the person who has to write the first message, not the side of the vendor who sells the feed. Most B2B intent data content skips that part, and that is exactly where the money is made or lost.
Here is what this page covers: the three types, what each one can honestly tell you, and the short playbook I use to turn a signal into a first line that does not sound like a template.
What intent data means in B2B
A prospect does something observable. A team downloads a comparison guide, a company posts a job ad for a role that only exists when a problem exists, a director visits your pricing page twice in a week. Each of those is a behavior you can see from the outside. Collected and attached to a company or a person, those behaviors become intent data.
Two things follow from that definition, and both matter more than any feature list.
First, the signal is probabilistic. It says "more likely than the average company on your list", not "ready to buy". A team reading about CRM migration might be doing research for a blog post. Treat every signal as a reason to look, not as a qualified lead.
Second, every signal has a shelf life. A pricing page visit is worth something for about a day. A funding round is worth something for a quarter. If your process takes two weeks to move from signal to message, most of your data is expired by the time you use it.
The three types of intent data
Vendors group the category in different ways, but the split that decides what you can do with a signal is the source. Demandbase uses the same first party and third party frame in its intent data FAQ, and Bombora built the reference implementation of the third party model.
| Type | Where it comes from | Resolves to | Best use |
|---|---|---|---|
| First party | Your own site, product, docs, emails | Company, sometimes a person | Same day follow up |
| Second party | A partner's property: review site, publisher, community | Company | Category shortlisting |
| Third party | A network of publisher sites across the open web | Company, by topic | Account prioritization |
| People and event signals | Job posts, funding, hiring, LinkedIn activity, webinars | A named person | A first message you can write |
First party intent data is behavior on property you own. Website visits, repeat visits, docs pages, trial activity, email clicks. You already have the raw material. The work is turning an anonymous visit into a company name and getting it in front of someone before the day ends.
Second party data is another company's first party data, shared with you commercially. The classic case is a review site telling you who compared you against a competitor last week. Narrow, but the intent is unusually clear.
Third party data is aggregated content consumption across a publisher network. A vendor tells you that a company showed a surge in reading about your topic. It is anonymous by construction, which is why it lands on the company and not on a person.
People and event signals sit outside most vendor definitions, and for a small team they are the most useful group of all. A company posting three SDR roles has a prospecting problem right now. That is a signal with a name and a job title attached, which means you can write to someone specific. I broke down the ones I use week to week in how to identify buying signals.
Account level or person level: the only distinction that changes your day
Ask one question about any source: does it give me a company, or does it give me a person?
Account level data tells you a company is in market. That is genuinely useful when you have a marketing team, an ads budget, and a CRM someone maintains. You raise bids on those accounts, you send them a campaign, you tell sales to prioritize them.
But if you are a founder or a two person sales team, an account level surge leaves you with a second job: figuring out who inside that company to contact, and what to say to them. Ninety percent of the effort in outbound sits in that gap. Account intent narrows a list of ten thousand companies to four hundred. It does not write your first line.
Person level signals skip the gap. Someone changed jobs, someone posted about the problem you solve, someone viewed your profile, someone attended a webinar on your topic. You know who they are and you know why you are writing.
If you have to choose one and you are small, choose person level. You will send fewer messages and get more conversations.
Where B2B intent data actually comes from
Sources vary in quality far more than vendor pages admit. This is roughly how I rank them for a small outbound team.
| Source | Signal quality | Freshness | Person level |
|---|---|---|---|
| Your own pricing and docs pages | High | Minutes | Sometimes |
| Job postings | High | Days | Yes, the hiring manager |
| Funding and headcount changes | Medium | Weeks | Yes |
| LinkedIn activity: posts, comments, profile views | Medium to high | Hours | Yes |
| Review site comparisons | High | Days | No |
| Publisher network topic surges | Medium | Weekly | No |
| Technographic installs | Low as intent, useful as a filter | Months | No |
One source is missing from that table on purpose: the words people type into a search engine before they ever land on you. You cannot see them prospect by prospect, but the vocabulary tells you which problems are live in your market, and the same phrases show up in job ads and LinkedIn posts. I collected them in buyer keywords.
Two notes on the table itself. Technographic data ("they use HubSpot") is not intent, it is a filter. Using it as intent is the most common way teams convince themselves a cold list is warm. And publisher network surges are only as good as the topic taxonomy behind them, which is why two vendors can disagree completely about the same account.
How to use intent data without a marketing team
Here is the loop I run. Four steps, no revenue operations hire required.
1. Pick one trigger you can see reliably. Not seven. One. Job ads for a specific role, or funding rounds in your segment, or people who viewed your profile. A single trigger you check every week beats a data platform you log into twice.
2. Filter against your customer profile before you look at the signal. A signal on a company you cannot serve is noise with better packaging. Company size, market, geography, language. Cut first, then read the signal.
3. Write the first line from the signal itself. This is the part tools cannot do for you, and it is the part that decides your reply rate. The signal has to be visible in the message, and the message has to be about them.
Weak, because it could be sent to anyone:
Hi Sarah, I saw your company is growing. We help B2B teams book more meetings. Open to a chat?
Better, because it could only be sent to one person:
Hi Sarah, saw you are hiring two SDRs in Berlin. Curious how you are handling the list building side while the seats are still empty, since that is usually what slows the ramp. Happy to share what we see working if useful.
4. Act inside the freshness window of the signal. Same day for a site visit, same week for a job ad, same month for a funding round. Late is the same as never, except you also burned the contact.
What to check before you pay for a signal feed
Prices in this category are mostly private, and the ones that are public tend to belong to products built for small teams. Before a demo call, get answers to these:
- Whose data is it? Plenty of tools resell the same underlying network. If two vendors quote you the same signal, you are comparing packaging.
- Account or person? Covered above. Ask for a real example, not a screenshot of a dashboard.
- How does the signal leave the tool? Webhook, CRM object, Slack alert, or straight into a sequence. This is where most of this data dies: it arrives somewhere nobody looks.
- What is the coverage? Several visitor identification products resolve individuals in the United States only. If you sell in Europe, ask before you sign.
- What happens on day one? If the setup needs a data engineer, you are buying a project, not a tool.
I compared the main vendors on these points, with the published prices where they exist, in intent data providers.
The privacy question, honestly
Third party data works by observing behavior across sites people do not know are connected. In Europe that runs into the GDPR, and the usual vendor answer is that the data stays at the company level and is therefore not personal data. That answer is fine until a signal resolves to a named individual, which is exactly what makes person level data useful.
The practical position I take: prefer signals a person published on purpose. A job ad, a public post, a profile view, a webinar registration. Those are visible by design, they are easier to defend, and they make a better first message anyway, because you can reference them out loud without sounding creepy.
Where Sendable fits
Sendable, the product I am building with my cofounder, sits on the narrow end of this category. It watches a small set of signals (job posts, webinars, profile visits, first degree network activity), filters the profiles against the ideal customer you describe, and pushes the ones that match into a LinkedIn sequence you built in a visual workflow, with a shared inbox and multiple sender accounts. Published pricing: $49 per sender per month from 1 to 4 senders, $39 from 5 to 25, $1,000 a month for unlimited senders.
What it is not: a topic intent platform across the open web, and not an email tool. It is LinkedIn only. If your buyers are not on LinkedIn, or you need account level surges to feed an ads budget, this is the wrong shape and one of the platforms in the comparison above will serve you better. More on getsendable.io.
FAQ
What is B2B intent data?
B2B intent data is behavioral data showing that a company or a person is researching a product category. It comes from your own site, from a partner's property such as a review site, or from a network of publisher sites across the open web. It indicates likelihood, not a decision, so it is best used to prioritize who you contact and when.
What is the difference between first party and third party intent data?
First party intent data is behavior on property you own: your site, your docs, your product. You already have it and it is the freshest signal you will ever get. Third party data is content consumption observed across external sites, sold by a vendor, and delivered at the company level. First party tells you who is looking at you; third party tells you who is looking at your category.
Is intent data accurate?
It is directional. Third party topic surges tell you a company read more than usual about a subject, which can mean a buying committee, a curious intern, or a competitor. Accuracy improves when you combine a signal with a fit filter and act quickly. Judge any provider on a sample of your own accounts, not on a case study.
How much does intent data cost?
Account level platforms sell annual contracts and rarely publish prices, so budget for a five figure yearly commitment and a procurement cycle. Tools built around person level signals more often publish a monthly price. A published price is a decent proxy for whether a product was designed for a team of your size.
How do I use these signals with a small team?
Pick one trigger you can check weekly, filter it against your customer profile, write the first message from the signal itself, and contact people inside the freshness window of that signal. One trigger used consistently beats a platform full of feeds nobody opens.