Buying-intent signals: what they are, where they come from, and how to weight them

The three sources of intent data

First-party intent is behaviour on properties you own: pricing page visits, documentation reads, demo requests, repeat visits from the same account. It is the most reliable because you can see the whole context.

Second-party intent comes from a partner you have a direct relationship with — a review site or publisher telling you an account was researching your category on their property.

Third-party intent is aggregated from wider publisher networks and sold as topic surges at account level. It is the broadest and the noisiest: it can tell you a company is researching a topic, but rarely who, or why.

Depth beats frequency

One visit to a pricing page or an implementation guide says more than ten visits to a blog post. Weight pages by how close they sit to a purchase decision, not by traffic volume.

Similarly, weight recency hard. Intent decays quickly; a surge from six weeks ago is a research memory, not a buying window.

Human signals still outrank digital ones

A prospect who names a deadline, loops in a colleague, asks implementation-specific questions, states a budget range, or follows up before you do has given you information no data vendor can supply. Any scoring model that ranks an anonymous topic surge above a stated deadline is mis-weighted.

This is why we treat digital intent as a prioritisation tool for outreach and human confirmation as the qualification decision. The first tells you who to call; only the second decides whether a lead is passed on.

A weighting approach that survives contact with reality

Score three inputs from 0 to 3: fit against your ICP, pressure to act now, and access to a decision maker. That gives a score out of nine, with a defined action per band — call today, sequence and re-score, or nurture only.

Keep the model small enough that a rep can explain it out loud. Complex weighted models fail not because the maths is wrong but because nobody maintains them after the first busy month.

Check the model against outcomes, not opinions

Review monthly: of the leads scored highest, how many reached a first meeting, and how many reached proposal? If high scores are not converting, the fit definition is usually the culprit — not the reps working the list.