Lead scoring: ranking leads by how likely they are to buy

Scoring exists because there are always more leads than sales hours. When marketing and sales share one score, they stop arguing about which leads are worth calling.

Most models combine two kinds of points:

• Fit points. Do the company and the person match the ideal customer profile? Industry, size, role, region.
• Behaviour points. What have they done? Opened emails, visited pages, booked a demo, used the product.

Many teams also subtract points for decay (no activity for weeks) and for disqualifiers (competitors, job seekers, students).

Behaviour scores have a weak spot: they guess intent from clicks. A pricing-page visit could be a buyer, a competitor or an intern doing research. A person who writes "we need this before Q1" has said what they want.

HearBuyers applies the same idea to public posts. Each post it reads gets a buying-intent score from 0 to 100, along with the sentence that shows the need and a one-line reason. Matches at or above a team's threshold (70 by default) get a drafted reply. Sample sentences:

• Sample, score 96. "I need a bookkeeper urgently, books are three months behind and taxes are due." Explicit need, urgency, a deadline.
• Sample, score 72. "Thinking about outsourcing our bookkeeping next year. Worth it?" Real interest and a vague timeline, so it lands just over the threshold.
• Sample, score 0. "Looking for monthly bookkeeping clients, I can take on three more." This is a seller.

The third is the classic false match: it uses a buyer's keywords with the opposite intent. Our bookkeeping scan, checked on 2026-10-16, scored a post like it 0 with the reason "seeking bookkeeping work rather than hiring a bookkeeper".

• Scoring activity, not intent. Ten email opens can mean nothing.
• Never testing the model. Compare scores against closed deals every quarter and adjust the weights.
• Ignoring negative signals. Job seekers and vendors inflate scores if you don't subtract for them.
• Setting the threshold once. Set it too low and sales drowns; too high and sales runs dry. Tune it to your team's capacity.

• Lead qualification: the human judgment scoring supports.
• MQL vs SQL: the stages scores usually feed.
• Hot lead: what a top score should mean.
• Intent data: the signals behaviour scores try to capture.

To see stated-intent scoring on your own examples, paste a sentence into the buying intent scorer.