HubSpot's native scoring property is straightforward to turn on but easy to configure badly — most broken lead scores come from weighting only positive signals, never revisiting the model, or ignoring how recently a behavior happened. Here's a step-by-step setup that avoids the common failure modes.
Step 1: Decide explicit vs. implicit criteria
Split your scoring inputs into two categories before touching HubSpot's settings:
- Explicit (fit) criteria — job title, company size, industry, and other firmographic data that indicates whether this contact matches your ideal customer profile, regardless of behavior.
- Implicit (engagement) criteria — page views, email opens and clicks, form submissions, and content downloads that indicate active interest.
A contact can score high on fit and low on engagement (a perfect-profile contact who's never engaged) or the reverse (heavy engagement from someone outside your target profile) — both are different problems, and collapsing them into one number hides which one you're dealing with.
Step 2: Build the score in HubSpot's settings
- Go to Settings > Properties, and locate (or create) the HubSpot Score property under Contact properties.
- Add positive attributes: target job titles, company size ranges, industries that match your ICP, and key behaviors like pricing page visits or demo requests.
- Add negative attributes: generic free-email domains if you sell B2B, unsubscribes, hard bounces, and — if relevant — known competitor domains.
- Weight behavioral actions by intent strength: a pricing page visit or demo request should score meaningfully higher than a single blog post view.
Step 3: Set the MQL threshold with sales, not alone
The score number is meaningless until it's tied to an action. Sit down with whoever owns follow-up and agree on the score threshold that defines a Marketing Qualified Lead, and revisit that threshold after the first month of real data — the initial number is a hypothesis, not a fixed rule.
| Mistake | Why It Breaks the Model |
|---|---|
| Only scoring positive signals | A contact who unsubscribed or bounced can still show as "high-scoring" from past activity |
| Never revisiting the model | Buyer behavior and content offerings change; a score built once and left alone drifts out of accuracy |
| Ignoring recency | Activity from eight months ago shouldn't weigh the same as activity from this week |
| Setting the MQL threshold without sales input | Marketing and sales end up disagreeing on what "qualified" means, undermining trust in the score |
Step 4: Review quarterly
Pull a sample of contacts that scored as MQLs each quarter and check with sales whether they were actually sales-ready. If the hit rate is low, the weighting needs adjustment — this is an ongoing calibration process, not a one-time setup task.
FAQ
What's the difference between explicit and implicit lead scoring criteria in HubSpot?
Explicit criteria are firmographic fit signals — job title, company size, industry — that indicate whether a contact matches your ideal customer profile regardless of behavior. Implicit criteria are engagement signals like page views, email opens, and form submissions that indicate active interest. Scoring them separately reveals whether a low-scoring contact is a fit problem or an engagement problem.
- A high-fit, low-engagement contact needs a different follow-up than a high-engagement, low-fit one.
- Collapsing both into a single score hides which type of gap you're actually looking at.
How often should a HubSpot lead scoring model be reviewed?
Review the model quarterly at minimum — pull a sample of contacts that scored as MQLs and check with sales whether they were actually sales-ready. A low hit rate signals the weighting needs adjustment. Buyer behavior and content offerings change over time, so a score configured once and left alone gradually drifts out of accuracy.
- Quarterly review with sales input keeps the MQL threshold aligned with actual sales readiness.
- A scoring model is a living configuration, not a one-time setup task.