TG3 47 SaaS brands scaled and $84M+ in client pipeline generated. See the proof → Free: 20 SaaS calculators, no signup. CAC, LTV, churn, Rule of 40. Open the tools → We rebuild attribution at the warehouse so every channel gets honest credit. See how →
TG3 SaaS/Glossary/Lead scoring
SaaS metrics glossary

Lead scoring.

A simple idea that teams routinely over-build. Here is what lead scoring is and how to do it so sales actually uses the score.

Definition
Lead scoring ranks leads by how likely they are to buy, by combining a fit score (how well they match your ICP) with a behaviour score (what they have done).

Two axes, not one. Fit tells you whether they should buy. Behaviour tells you whether they are ready to. A lead high on both is hot. High on one and low on the other means very different next steps.

How lead scoring works

How lead scoring works.

Fit score  +  behaviour score  =  lead score
Fit scoreindustry, company size, role and region against your ICP
Behaviour scoreactions that predict intent, weighted by how strongly they predict a buy

The trap is over-engineering the weights. Start simple, validate against real conversions and adjust, see MQL for where the score feeds.

Benchmarks

Why most models are over-built.

Teams love to assign points to dozens of actions, most of which do not predict anything. A model with five well-chosen signals usually beats one with forty guessed ones.

The only validation that matters is whether high-scoring leads actually convert better than low-scoring ones. If they do not, the model is decoration, not a tool.

How to improve it

How to build a lead score sales trusts.

01

Score fit and behaviour separately

Keep the two axes distinct so a bad-fit, high-activity lead never scores hot.

02

Start with five signals

Use the few actions that actually predict conversion, not everything you can track.

03

Validate against conversions

Check that high scores really do convert better. If not, fix the weights.

04

Decay old behaviour

A demo request from six months ago is not current intent. Let behaviour scores fade.

Common questions

Questions about lead scoring.

What is lead scoring?+

Ranking leads by how likely they are to buy, usually by combining a fit score with a behaviour score.

What are the two parts of a lead score?+

Fit, how well the lead matches your ICP and behaviour, what actions they have taken that predict intent.

Why are most lead scoring models too complex?+

Because teams assign points to many actions that do not predict conversion. A few well-chosen signals usually work better.

How do you know if a lead score works?+

High-scoring leads should convert at a higher rate than low-scoring ones. If they do not, the model needs fixing.

Marketing qualified leadSales qualified leadProduct qualified leadHow we run lifecycle

A score sales will use?

The 30-minute audit includes whether your lead score predicts conversion or just adds noise. No sales sequence.

Book the 30-minute audit →
Response inside 4 business hours · We turn down 1 in 3