Almost every HubSpot instance has a lead score. Almost none of them have a sales team that looks at it. The reason is structural, not cultural.
Open the scoring settings and look at where the points come from. Email opens, page views, form submissions, maybe a job title contains rule. Now ask which of those actually separated your won deals from your lost ones last year. Almost nobody can answer, because the model was built forward from assumption rather than backward from outcomes.
An email open is close to meaningless now that mail clients prefetch images. A pricing page visit at 11pm from a competitor scores the same as one from a buying committee. The model cannot tell the difference because nobody asked it to.
Lead scoring ranks individual contacts. Revenue arrives from accounts, and accounts buy through several people who behave differently. One person fills in a form. Another one forwards the pricing page to procurement. A third never touches your site and signs the contract.
Scoring the first person highly and the account not at all is how a 90 point lead sits in a queue while the real deal moves without you.
Your product changes, your pricing changes, your ICP shifts, and the score does not. Most instances we open have a model that has not been touched in two or more years, still awarding points for a webinar series that ended in 2023, still ignoring the product usage signal that arrived last spring.
A number with no explanation is not actionable. If a rep sees 84 and cannot see why it is 84, the only rational response is to ignore it and open the account themselves. Any score that cannot say "procurement joined the thread on Tuesday" is asking for trust it has not earned.
Start from closed deals rather than from opinion. Export last year of closed-won and closed-lost, then check which signals appeared in the wins and not in the losses. Two or three of the things your team treats as qualification will predict nothing. Something nobody tracks will predict a lot.
Score the account, not just the person, and combine what the company is doing with what named contacts are doing. Then show the reasons alongside the number, in the place the rep already works, rather than in a property nobody opens. And re-test the weights every quarter, because the business the model was fitted to is not the business you have now.
The feature is fine. The usual problem is that the model behind it was never tested against outcomes and has not been updated since implementation, so it produces numbers nobody has reason to trust.
Signals that actually separated your won deals from your lost deals, weighted by how much they separated them, and re-tested regularly. Anything else is a guess with decimal places.
Because a score with no visible reasoning cannot be acted on, and because a few weeks of working the list teaches them the ranking does not predict who closes.
Lead scoring ranks individual people, usually on demographics and engagement. Deal scoring ranks open opportunities on what the buying group is doing, weighted against deals you already closed.
Leave an email and I will reply myself, usually the same day. No sequence, no newsletter, and no calendar link unless you ask for one.
Would rather just book time? Here is the calendar.
We use cookies to run the site and to understand how it is used. You can accept all or necessary only. See our privacy notice.Privacy notice