Featured1 min read
Why we killed our own lead scoring model
We built a lead scoring model, sold it as a core feature, and shut it off eight months later. Here is what replaced it.
Callum AshfordPipeline Metrics
Sales0 articles7 subtopics
Topic
Starting, running and growing companies, from people who are doing it.
We built a lead scoring model, sold it as a core feature, and shut it off eight months later. Here is what replaced it.
We called back every subject from case studies we published in one calendar year to check if the numbers held up. Most did not, in an interesting way.
Not a machine learning model. A basic weighted-stage calculation that outperformed manual forecasts for two straight quarters.
A specific, common pattern in customer quotes that reads as fake even when it is completely true.
A conversation with our own VP of Sales about a compensation change that felt risky and paid off in an unexpected way.