Definition
Lead scoring is a method of ranking leads by how likely they are to buy, by adding points for who they are and what they do. Fit points come from facts like company size and job title. Behaviour points come from actions like visiting pricing or replying. The total decides who sales sees first.
Why it matters when you sell B2B
Sales time is limited, and scoring decides who gets it. A good model sends reps to the leads most likely to buy. A bad one sends them to the leads who download the most.
Fit should usually carry more weight than behaviour. The fit side comes from your ideal customer profile, the behaviour side from buyer intent. Crossing a threshold makes a lead a marketing qualified lead.
A worked example
Here is one invented lead, scored.
One lead, scoredExample
Score = fit points + behaviour points
- Company size and industry in the ICP+30
- Job title owns the problem+20
- Visited the pricing page twice+20
- Downloaded a guide+5
Score = 30 + 20 + 20 + 5 = 75. The threshold is 60, so the lead goes to sales.
Notice the weights: fit is worth 50 of the 75. A lead with no fit and lots of downloads would score 25 and stay with marketing, which is the right call.
Common mistakes
- 1Points from guesses. Set weights from past wins and losses, not from what feels important.
- 2Too much weight on behaviour. Busy readers who will never buy rise to the top.
- 3No decay. A pricing visit from six months ago should not count like one from yesterday.
- 4Scoring people, ignoring accounts. Three medium leads from one company can matter more than one high lead alone.
- 5Never checking against wins. If winners do not score higher than losers, the model is not working.
Test your scoring model
- Take your last 50 won deals and 50 lost ones.
- Score each lead as it looked before the first call.
- Check that winners score clearly higher.
- Adjust the weights that mislead the most.
- Add decay, so old actions fade.