Lead scoring
Quick answer
Scoring leads by fit with the ideal customer profile and by behaviour, so the team works the highest-probability opportunities first.
Key takeaways
- Start from customers won in the last 12 months and identify shared attributes
- Score fit and behaviour separately; a good-fit but inactive lead needs a different action than an active but poor-fit one
- Recalibrate quarterly against real outcomes
What it is
Lead scoring assigns each lead a score on two axes: how well it fits the ideal customer profile (industry, size, region) and how active it is (requests, opens, visits, replies).
Building a model that works
- Start from customers won in the last 12 months and identify shared attributes
- Score fit and behaviour separately; a good-fit but inactive lead needs a different action than an active but poor-fit one
- Recalibrate quarterly against real outcomes
The common mistake
Copying models from other industries. A score that works in B2B software does not transfer to distribution, where order frequency and geography weigh far more than digital engagement.
Real-world example
A 12-rep team cut time spent on poor-fit leads by 40% after the score was recalibrated on actual won customers rather than marketing assumptions.
Last updated:
Frequently asked
How much data does a useful model need?+
Roughly 50-100 closed opportunities, won and lost. Below that, the score reflects noise rather than pattern.
Does lead scoring replace human qualification?+
No. It orders the work queue; the qualification decision stays with the rep, based on the conversation.
Where CRMconnect fits
Related terms
Sales pipeline
A structured view of open opportunities grouped by stage, used to estimate future revenue and prioritise commercial effort.
SalesWin rate
The share of closed opportunities that were won - the main indicator of sales process and qualification quality.
SalesCRM adoption
How far the team actually uses the system in daily work - the factor that decides whether an implementation produces results.