Intent data

Intent data is a record of research activity that suggests a company may be getting ready to buy, like four people from one account reading your pricing page or comparing you with a competitor on G2. Sales and marketing use it to decide which accounts to reach first.
What is intent data?
Intent data is information about what companies are researching online, collected so you can spot the ones that may be getting ready to buy. It comes from behavior: the pages people read, the comparisons they run, the topics they look up and how often they come back.
Let's say you sell test automation software. Over two weeks, four people from one fintech company read your pricing page. A QA lead there compares you with two other tools on G2. Across that company, reading about "Selenium alternatives" jumps well above normal. None of them filled in a form, but intent data tells you this account deserves a closer look.
There are two main kinds, and each one answers a different question:
| First-party intent data | Third-party intent data | |
|---|---|---|
| Where it comes from | Your website, product, emails and events | Publisher networks, review sites and data co-ops |
| What it shows | Accounts already interacting with you | Accounts researching your topic anywhere |
| Examples | Pricing page visits, demo page views, trial sign-ups | Bombora Company Surge, G2 Buyer Intent |
| Main limit | Only sees people who've found you | Broader signals that need checking |
Why intent data matters for B2B SaaS
B2B buyers do most of their research before they talk to a sales rep. They read reviews, compare tools and ask peers, and several people from the same company often do it at once. By the time one of them books a demo, the shortlist is mostly set.
Intent data gives you a view into that research phase while it's still happening. Here's where I've seen it help the most:
- Picking which accounts to work: Your sales team can't reach every account on your target list each week. Intent shows which ones are active right now.
- Timing outreach: A message about switching test tools lands very differently when the account is comparing test tools this month.
- Running ABM: In an ABM program, intent decides which accounts get ads, which get personal outreach from the account owner and which wait for next quarter.
- Planning content: The topics your target accounts read about most tell you what's on their mind. I use them next to SEMRush and Reddit research when I plan topics.
How to use intent data well
Intent data is easy to buy and hard to use. Most of the value comes from what you do with a signal after it arrives. Here's my process:
- Filter by fit firstRun every signal through your ICP. A surge from a company that could never buy from you is noise, however strong it looks.
- Use your own signals firstPricing page visits, return visits and trial sign-ups from known accounts are the clearest intent you'll get, and you already own them.
- Add one outside sourcePick a third-party provider that covers your category well and test it on a small set of accounts before you roll it out.
- Feed it into a scoreAdd intent to your lead scoring so one blog visit doesn't count the same as four people on the pricing page.
- Agree on the next actionDecide with sales what each level triggers, like an ad, a personal email or a call from the account owner.
It helps to know how a third-party score is built before you trust it. According to Bombora, its Company Surge score compares an account's last three weeks of research on a topic with that same account's 12-week baseline. So a surge means a company is reading about a topic more than it usually does, which tells you more than a raw page count.
An intent data example
G2's Buyer Intent is a good real example of signals that line up with buying stages. According to G2's documentation, it reports when a company views your G2 profile, your pricing, a comparison page with you and a competitor, alternatives pages or a category page that lists you. A category view is early research. A comparison view usually means you're on a shortlist.
Here's how a hypothetical test automation company could sort the signals it sees in a typical month:
Which intent signals deserve a sales call
Act within days
- Three people from one ICP account on your pricing page in the same week
- A G2 comparison of your product and a competitor
- A return visit to a case study from the account's industry
- A trial sign-up with a work email from a target account
Use for ads and content
- One visit to a top-of-funnel blog post
- A topic surge at a company outside your ICP
- A job seeker reading your careers page
- A single email open
The right column still has value. Those accounts can go into ad audiences or a nurture track, so they see you again when their research gets serious.
How to measure intent data
Intent data costs money and sales time, so it has to show up in pipeline. Here's what I track:
- Pipeline from intent accounts: Compare opportunities created from accounts that showed intent with accounts that didn't. If the intent group doesn't convert better, your signals or your follow-up need work.
- Time to first touch: Measure how long sales takes to reach out after an account crosses your threshold. A surge from three weeks ago has often moved on.
- Win rate by signal: Look at which signals came before won deals. After a few quarters, you'll know which ones to weight higher, and your win rate by signal tells you where to focus.
- Coverage: Check what share of your target accounts show any signal at all. Very low coverage from a third-party tool usually means it doesn't see your niche well.
Common intent data mistakes
- Treating a surge like a hand raise: A topic surge says a company is reading about something. It doesn't say who, or that they want to talk. Opening with "I saw you're researching test automation" tends to feel invasive.
- Buying data before fixing the basics: If your site doesn't track pricing page visits by account, start there. Your own signals cost less and are more accurate.
- Tracking too many topics: Fifty loosely related topics flood sales with alerts. Pick a short list that only real buyers would research.
- No feedback from sales: If reps never report which alerts led to real conversations, you can't tune the model.
- Watching one person: One visit means little. Several people from the same buying committee researching together is the pattern worth acting on.
Here's a quick test you can run this week. Pull the accounts that visited your pricing page last month. How many of them did your sales team reach out to?
Read more
FAQ
What is the difference between first-party and third-party intent data?
First-party intent data comes from your own channels, like pricing page visits, product sign-ups and webinar attendance. Third-party intent data comes from outside sources, like Bombora's publisher co-op or G2's review site, and shows research that happens before a buyer reaches your site.
Does intent data identify individual people?
Most B2B intent tools report at the company level. G2's Buyer Intent, for example, gives the company name and domain, plus only the region and country of the person who triggered the signal. Check each provider's privacy and consent terms with your legal team before you buy.
How accurate is intent data?
It tells you a company researched a topic. It can't tell you who, or whether a real project exists. Treat a signal as a reason to look closer and combine it with fit and your own first-party data before sales reaches out.
Do small B2B SaaS teams need paid intent data?
Start with first-party intent. Tracking pricing page visits, return visits and trial sign-ups by account costs little and gives you the clearest signals. Add a paid third-party source once sales can act on what you already have.



