Intent data: what it can tell you, what it cannot, and how to act on it
A prioritisation signal, never a qualification
Intent data are behavioural signals that suggest an account is researching a problem you solve: content consumption, search behaviour, review-site activity, technology changes. Used well, intent data tells you when an account on your list is in market, so you can act while the buying group is still forming its shortlist. Used badly, it is an expensive way to cold-call people who read one article. This guide covers what intent data can and cannot tell you, which sources to trust, and how to turn a signal into a play.
Why intent matters: the invisible buying journey
The modern B2B buying journey is mostly invisible to you. Buying groups research anonymously, define requirements internally and rank their shortlist before any vendor hears from them. By the time a form is filled in, most of the decision has been made. Intent data is the only instrument that gives you a view into that anonymous phase, imperfect as the view may be.
That is also why intent belongs inside an ABM programme rather than as a standalone lead source: it answers the question "which of the accounts we already chose is moving right now?" If you have not chosen your accounts yet, start with the account selection guide.
What intent data actually is
- First-party intent: behaviour on your own properties. Website visits, content downloads, pricing-page views, product trials, email engagement. The most reliable signal you have, because you can see exactly what happened, at account level if your analytics are set up for it.
- Second-party intent: data a platform observes on its own property and sells to you, for example review-site research activity. Reliable within its scope: someone comparing products on a review platform is genuinely researching.
- Third-party intent: aggregated content-consumption data from publisher networks, bidstream data and data co-ops, matched to accounts. The broadest coverage and the noisiest signal.
What a signal can and cannot tell you
An intent spike tells you that someone at the account is consuming content on a topic. It does not tell you who, why, or whether there is budget. It could be a buying group forming. It could equally be an intern writing a paper, a competitor doing research, or your own champion validating a decision already made against you.
Treat intent as a prioritisation signal, never as a qualification. It decides where you look first, not what you say.
From signal to play
A signal without an agreed follow-up play is trivia. What works in practice:
- Define signal thresholds together with sales. What combination of signals promotes an account from watching to working? Write it down. One spike is weather; a pattern across weeks and topics is climate.
- Match the play to the signal stage. Early research signals earn helpful, ungated content and warm-up advertising, not a phone call. Late-stage signals, pricing pages, comparison keywords, review-site activity, justify direct sales outreach.
- Never reveal the surveillance. "I saw your company is researching X" is how you burn trust. The signal decides timing and channel; the message comes from your account research.
- Agree an SLA. A late-stage signal that waits ten days in a queue was not worth buying. Decide response times per signal tier and measure them.
Common failure modes
- Buying third-party intent before instrumenting first-party. If you cannot see which target accounts visit your own pricing page, fix that first. It is cheaper and more accurate than anything you can buy.
- Keyword lists that are too broad. Intent on a generic category term means little. Intent on your niche problem terms and competitor names means a lot.
- No baseline. A "surge" only exists relative to normal behaviour. Give any new intent source a few weeks to establish baselines before you act on it.
- Using intent to spam. If every spike triggers a sequence, you train the market to ignore you and sales to ignore the data.
Frequently asked questions
What is intent data in B2B marketing?
Intent data are behavioural signals, such as content consumption, search activity and review-site research, that suggest an account is actively researching a problem or solution category. It comes in first-party form (behaviour on your own website and content), second-party (observed and sold by platforms such as review sites) and third-party (aggregated from publisher networks).
Is intent data reliable?
As a prioritisation signal, yes; as a qualification, no. A spike tells you someone at the account is researching a topic, not who, why, or whether budget exists. First-party signals are the most reliable, third-party the noisiest. Look for patterns across weeks and topics rather than single spikes.
Do I need to buy an intent data platform for ABM?
Not necessarily to start. First instrument your own first-party signals: account-level website analytics, pricing-page visits, content engagement. Buy third-party intent when you have a working target account list, agreed follow-up plays with sales, and the capacity to act on more signals than your own properties produce.
How should sales use intent data?
As timing and prioritisation, never as an opening line. Referencing an account's research behaviour in outreach destroys trust. The signal tells sales which account to work this week and through which channel; the message itself should come from account research and the agreed play.
Account-based marketing strategist and full-stack marketer with 20+ years of B2B experience, including ABM programmes in healthcare IT. Writes about account-based marketing and sales, and builds his own marketing and AI tooling. Find me on LinkedIn.