HUBSPOT GUIDE
Lead Scoring vs Predictive Lead Scoring in HubSpot: What's the Difference
HubSpot ships two distinct lead scoring systems, and the names are close enough that they get conflated at onboarding all the time. Manual lead scoring is the model you build with your own rules. Predictive lead scoring is the model HubSpot's machine learning builds for you. They answer different questions - and the smartest teams run both. Here's exactly how they differ, when to use each, and how to set them up properly.
Two Systems, Two Different Questions.
Lead scoring assigns a numerical value to each contact so your sales team knows who to call first. HubSpot does this in two fundamentally different ways. The distinction matters because they're built differently, priced differently, and trusted differently - and confusing them leads to teams either over-relying on a black box or never switching on a tool that's already in their licence.
You define the rules and assign the points. Page visit adds 5, a senior job title adds 10, an unsubscribe subtracts 20. Every weight is yours to set and edit.
- Fully transparent - every point is traceable
- Splits into Fit (who they are) and Engagement (what they do)
- Total control over the logic
- Available on Professional and Enterprise
- Works from day one, no data threshold
HubSpot's machine learning analyses your closed-won and closed-lost history, then scores every open contact on its probability of becoming a customer within 90 days.
- No rules to write - the model self-trains
- Produces "Likelihood to close" and "Contact priority"
- Spots patterns a human rubric would miss
- Enterprise only
- Needs 100+ customers and 1,000+ non-customers
Four Things That Actually Set Them Apart.
Who Builds the Model
With manual scoring, you are the model. You decide that a "Director" title is worth 10 points and a pricing-page visit is worth 5, and you can change those weights any time. With predictive scoring, HubSpot's machine learning builds the model by analysing the contacts who became customers versus those who didn't - then scores new contacts on how closely they resemble your historical buyers. You don't set any rules; the AI infers them.
Transparency vs Pattern Recognition
This is the central trade-off. Manual scoring is fully transparent - every rule is visible, every weight is editable, and any contact's score traces straight back to the rules that touched it. Predictive scoring is a black box: it can surface patterns your manual rubric would never catch, but it's not possible to know exactly how each input contributes to a given contact's score. You gain reach and lose explainability.
What the Output Looks Like
Manual scoring produces a points total you define the meaning of - commonly a 0-100 score split into Fit and Engagement. Predictive scoring produces two specific, read-only properties: "Likelihood to close" (a percentage probability of closing within 90 days - a contact at 22 has a 22% chance) and "Contact priority" (a tiered ranking: Very High, High, Medium, Low, or Closed Won). These predictive properties are set automatically by HubSpot and cannot be edited.
Plan and Data Requirements
Manual scoring is available on Marketing Hub Professional and Sales Hub Professional and above, and works from day one regardless of how much data you hold. Predictive scoring is Enterprise only, and HubSpot won't even enable it until you have at least 100 customers and 1,000 non-customers - because below that, there simply isn't enough history to train a reliable model.
Each priority tier holds roughly 25% of your contacts, sorted by their likelihood-to-close score. Because the tiers are relative groupings, the boundaries shift over time as your database changes - a contact that's "High" today could slip to "Medium" next month if stronger leads enter the pool.
Manual vs Predictive at a Glance.
| Factor | Manual | Predictive |
|---|---|---|
| You define the rules | ✓ | ✗ |
| Fully transparent and editable | ✓ | ✗ |
| Finds patterns humans miss | ✗ | ✓ |
| Works with little data | ✓ | ✗ |
| Splits into Fit and Engagement | ✓ | N/A |
| Outputs a 90-day close probability | ✗ | ✓ |
| Score can be manually adjusted | ✓ | ✗ |
| Available on Professional | ✓ | ✗ |
| Available on Enterprise | ✓ | ✓ |
| Usable in lists, workflows and reports | ✓ | ✓ |
A Simple Way to Decide.
You don't have to pick just one - and most mature teams don't. But here's the decision logic we walk clients through when they're starting out.
Is Your Lead Scoring Pulling Its Weight?
A score that nobody routes from is shelfware. Here's how to tell whether your lead scoring is actually driving sales behaviour - or just sitting on a contact property.
Signs It's Working
Warning Signs
Frequently Asked Questions.
We've had a great experience with Greg and Hayley at warbble·digital, who have been endlessly helpful and patient with all our team, thank you!
- Parker, H. · Jul 2025 · Customer Success Training
Is your lead scoring built on clean data?
Get a free Thrive Score in 30 minutes. We benchmark your portal across Feature Adoption, Data Quality, Automation Maturity, and Revenue Readiness - the exact foundations that make lead scoring (manual or predictive) actually work.
200+ Implementations
