Two accounts. Identical firmographics. Same industry, same size, same tools. On paper, twins.
Except one of them has spent three weeks reading comparison pages, downloading pricing guides, and researching your category across the web. The other one has done nothing.
Your CRM can’t tell them apart. B2B intent data exists because that difference is worth money. So let’s unpack what it is, where it comes from, and how to use it without wasting a budget.
đź“Ś TL;DR: B2B intent data shows which companies are actively researching a topic or category right now. It's collected from content consumption across the web and resolved to the account level. Treat it as a smoke detector, not a prophecy: it tells you where to look first, never who will buy.
What is B2B intent data?
B2B intent data is behavioral information showing which companies are actively researching a topic, product category, or solution. It’s collected from content consumption: searches, page visits, downloads, review-site activity, and engagement across publisher networks. The signal suggests an account may be in a buying cycle right now.
Here’s the simplest way to place it. Everything else in your data stack describes WHO an account is. Intent data is the only layer that tries to answer WHEN they’re ready.
That’s why it gets so much attention. And why it disappoints teams who expect a crystal ball. Intent signals are probabilistic, noisy, and account-level. Used with those limits in mind, they’re genuinely valuable. Used as gospel, they burn quarters.
What are the types of intent data?
Intent data splits into three types by where the signal comes from, and the source decides both its strength and its blind spots:
| Type | Source | Strength | Blind spot |
|---|---|---|---|
| First-party | Your own site and content | Highest accuracy, they came to YOU | Tiny volume; only sees existing awareness |
| Second-party | Review sites and partner platforms | High buying relevance | Limited to those platforms’ visitors |
| Third-party | Publisher and content networks | Sees research you’d never see | Noisier; topic-level, not product-level |
The pattern most teams land on: first-party signals as the trust anchor, third-party signals for reach into accounts that haven’t found you yet.
Sound familiar? It’s the same framing used for first, second, and third-party data in general. Intent just applies it to behavior instead of records.
How is intent data collected?
Third-party providers watch content consumption across networks of publisher sites, then resolve that activity back to companies, usually through IP-to-company matching (mapping a visitor’s network address to an employer) and cookie pools. When people at an account read noticeably more about a topic than their normal baseline, the account “surges” on that topic.
Read that twice, because both halves matter:
- Resolved to companies, not people. Intent tells you an ACCOUNT is researching, not which human. Pairing the signal with person-level data is your job, not the provider’s.
- Against a baseline. Good providers measure unusual spikes, not raw volume. A tech blog’s readers always read about CRMs. That’s not intent, that’s Tuesday.
Now the honest limitations. IP-to-company matching gets fuzzier every year as remote work spreads. Topic taxonomies (the category labels providers track) are broad. Cookie restrictions keep tightening the identity layer underneath. And two providers watching different networks will surge different accounts. Bombora’s own intent data explainer and Demandbase’s FAQ are refreshingly open about the mechanics if you want the provider-side view.
Intent data is genuinely useful AND genuinely noisy. Both things are true. Plan for both.
What does that planning look like in practice? Treat every surge as a hypothesis. Check it against fit, check it against your own first-party signals, and let agreement between sources raise your confidence. One source alone never should.
How do you actually use intent data?
Use intent data to change the ORDER and TIMING of work you were already doing. The four plays that reliably pay:
1. Prioritization. The simplest and best. Your reps have 500 target accounts; intent tells them which 30 to work this week. No new motion, just better sequencing.
2. Account scoring. Add surge signals as a scoring input next to fit. Fit says “worth winning.” Intent says “worth calling now.” The accounts scoring high on both are your pipeline for next quarter.
3. Ad audiences. Point paid spend at surging accounts instead of the whole addressable market. Same budget, warmer eyeballs.
4. Churn early-warning. An existing customer surging on your competitor’s category is a renewal conversation you want to have EARLY. Intent pointed inward is criminally underused, and almost nobody writes about it. Yet the same subscription covers it, and a saved renewal often pays for the whole contract.
One thing intent should NOT do: trigger a creepy “we saw you researching…” email. Nobody enjoys being watched. Use the signal to time and prioritize. The prospect never needs to know why you showed up this week.
What does intent data need before it works?
Intent data needs three foundations in place first: a defined ICP, complete account records, and contacts to act with. It multiplies the quality of your existing data. It can’t fix it.
- A defined ICP. A surge from an account you’d never sell to is noise with a price tag. Fit filters first.
- Complete account records. Intent arrives at the account level, so your accounts need clean domains and firmographics to match the signal against. That’s account enrichment territory. No domain, no match: resolving names to domains (Company URL Finder’s specialty) is the quiet prep work that decides whether intent signals land anywhere at all.
- Contacts to act with. Intent says “this account is warm.” You still need the right humans, which is where contact data enrichment picks up the handoff.
Think of intent as the top of the B2B data enrichment stack. The timing layer over the identity layers. Build the base first.
đź’ˇ Free first-party intent: pull the list of accounts that visited your pricing page, downloaded anything, or showed up twice in thirty days. That's real intent data, it's sitting in your analytics right now, and it costs nothing. Work that list before you pay for anyone else's signals.
How do you choose an intent data provider?
Choose intent providers by testing their signal against deals you already closed. I call it the retroactive test: take your last 20 closed-won deals, and ask the provider to show what their data said about those accounts in the months before each deal. If their surges line up with your real buying cycles, the signal is real for YOUR market.
The big names here (Bombora, 6sense, ZoomInfo) each watch different networks and score differently, which is exactly why the retroactive test matters more than any comparison chart. Then ask four hard questions:
- Which publisher network do you actually watch?
- How do you resolve identity as cookies fade?
- What does your topic taxonomy look like for MY category?
- What does a surge score mean, mathematically?
Vague answers to the last one are disqualifying. A score nobody can explain is a vibe, not a signal.
The intent data mistakes I keep seeing
Five patterns, over and over:
- Treating surges as leads. A surge is a reason to look, not a person to call.
- Buying intent before the basics. No ICP, messy account records, missing domains. The signal arrives and lands nowhere.
- Creepy outreach. “Noticed your team researching…” reads as surveillance. Because it is.
- Ignoring free first-party signals. Paying for third-party surges while pricing-page visitors go unworked is backwards.
- Assuming one provider sees everything. Each network is a window, not the whole house.
My retroactive test, honestly
We piloted a third-party intent provider at my Hamburg job a few years back, and I ran the retroactive test myself on our last 20 closed-won deals.
The result was a genuine split. About half the accounts showed clear surges in the two quarters before closing. The rest showed nothing at all, mostly smaller companies where IP matching had little to grab onto.
So we bought it, but we bought it for what it proved it could do: reordering the week’s outreach across our larger target accounts. Not for forecasting revenue. Not for small accounts. That narrower expectation is why the team still liked it a year later, while two peers at other companies quietly let their contracts lapse. Expectations, not data quality, decided all three outcomes.
🧠Privacy note: intent signals arrive at the account level, but the moment you act on individual people, person-level rules apply. The GDPR framework, ICO guidance, and the Data Privacy Framework for cross-border handling all still apply to your outreach. Ask providers how their collection is consented. The good ones answer fast.
How I know this (and what to check yourself)
This guide comes from piloting intent tools hands-on, running the retroactive test, and watching teams succeed or churn based on their expectations. The honest limit: every market surfaces differently in every network. My split won’t be your split. Run your own 20-deal test before signing anything, and re-check yearly, because the identity layer under intent data is shifting fast.
Frequently Asked Questions
What is B2B intent data?
B2B intent data is behavioral information showing which companies are actively researching a topic or product category. It’s collected from content consumption across websites, review platforms, and publisher networks, and resolved to the account level.
What is an example of intent data?
A 400-person software company suddenly reading far more content about “data enrichment tools” than its baseline, across multiple publisher sites, is an intent surge. A vendor in that category would move the account to the top of this week’s outreach list.
How does ZoomInfo get intent data?
Like most third-party providers, ZoomInfo tracks content consumption signals across a network of sites and data partnerships, then resolves that activity to companies. The exact network differs by provider, which is why the same account can surge in one platform and stay silent in another.
Who offers the best intent data?
Honestly, it depends on your market. Bombora, 6sense, and ZoomInfo are the most established names, but “best” is whichever network sees YOUR buyers. Run the retroactive test on your closed-won deals and let the results pick the winner.
Is intent data worth it for small teams?
Usually only after the basics are solid. If your account list is small, your first-party signals (site visits, downloads, demo requests) carry most of the intent value for free. Paid third-party intent earns its cost when you have more target accounts than your team can work.
What’s the difference between intent data and technographic data?
Technographic data shows what a company currently runs; intent data shows what it’s currently researching. One describes the present stack, the other hints at the next purchase. My technographic data guide covers the first half of that pair.
It’s time to work the warm ones first
You don’t need a platform to start thinking in intent. This week, pull the accounts that visited your pricing page, downloaded anything, or showed up twice in thirty days. Free, and already yours.
Work those accounts first. When that habit runs out of accounts, THEN look at third-party providers, retroactive test in hand.
You’ve got this. Tell me in the comments: do your reps work accounts in any particular order today, or is it alphabetical-by-vibes? Be honest.
