Lead scoring

Lead scoring quantifies how sales-ready a lead is. Without scoring, every lead looks equal - the enterprise prospect who visited your pricing page five times looks the same as the student who downloaded one ebook. With scoring, you prioritise the right leads.

This chapter configures scoring criteria based on demographic fit and behavioural engagement, explains how to balance behavioural versus demographic signals, creates score-based workflow triggers that automate actions when leads reach thresholds, and establishes a system to maintain and adjust scores as you learn what actually predicts conversions.

Effective lead scoring improves sales efficiency by helping reps focus on leads most likely to convert. Let's build a scoring model that actually works.

Lead scoring assigns points based on who someone is (demographic) and what they've done (behavioural). The total score indicates sales-readiness.

Planning your scoring model

Before configuring scores in HubSpot, define what makes a lead valuable to your business.

Ask these questions:

• What company characteristics predict good customers? (company size, industry, revenue, tech stack)
• What behaviours indicate buying intent? (pricing page visits, demo requests, competitor comparison searches)
• What disqualifies leads? (students, competitors, wrong geography)

Document your answers. These become your scoring criteria.

Example for B2B SaaS company:

Positive scoring criteria:

• Company size 50-500 employees: +10 points
• Company size 500+: +15 points
• Industry is Technology or Professional Services: +10 points
• Job title contains "Director," "VP," or "Head of": +15 points
• Job title contains "Manager" or "Lead": +10 points
• Visited pricing page: +10 points
• Downloaded whitepaper: +5 points
• Attended webinar: +15 points
• Requested demo: +25 points
• Opened 3+ marketing emails: +5 points
• Visited website 5+ times: +10 points

Negative scoring criteria:

• Email domain is gmail.com, yahoo.com, or similar free provider: -10 points
• Job title contains "Student" or "Intern": -50 points
• Company is a known competitor: -100 points
• Country is outside target markets: -20 points

Total possible score: 0-100+ points. You'll set thresholds (e.g., 60+ = MQL, 80+ = hot lead).

Demographic vs behavioural criteria

Balance both types of signals in your scoring model.

Demographic criteria (firmographic and role-based):

• Company size
• Industry
• Revenue
• Job title / seniority
• Geography
• Tech stack (if you track this)

Demographic criteria answer: "Are they a good fit for our product?"

Behavioural criteria (actions and engagement):

• Website visits
• Page views (especially high-intent pages like pricing, customers, comparison pages)
• Content downloads
• Email engagement
• Form submissions
• Webinar attendance
• Tool/product trial activity

Behavioural criteria answer: "Are they actively researching solutions?"

Best leads score highly on both dimensions: they fit your ICP (ideal customer profile) AND they're actively engaged.

Creating scoring properties

HubSpot uses a default "HubSpot score" property, but you can create custom scoring properties for more control.

Navigate to Settings > Properties > Contact properties. Find "HubSpot score" property.

HubSpot score is calculated automatically based on criteria you configure. Let's configure it.

Click "Set up lead scoring" (appears when you view HubSpot score property details).

You'll see categories: Contact information, Email activity, Web activity, Social activity, CRM activity.

Configure contact information scoring:

• Company size (number of employees): If >= 50, +10 points. If >= 500, +15 points.
• Industry: If Technology, +10 points. If Professional Services, +10 points.
• Job title: Use text contains logic. If contains "Director," +15 points. If contains "Manager," +10 points.
• Country: If United Kingdom, +5 points. If United States, +5 points.

Configure email activity scoring:

• Email opened: +1 point per open (max +5 points)
• Email clicked: +3 points per click (max +15 points)
• Unsubscribed: -20 points

Configure web activity scoring:

• Page views: +1 point per view (max +10 points)
• Specific page views: If visited /pricing, +10 points. If visited /customers, +5 points.
• Form submissions: If submitted "Contact sales" form, +25 points. If submitted "Newsletter signup," +3 points.

Configure CRM activity scoring:

• Meeting scheduled: +20 points
• Deal created: +50 points (they're already an opportunity)

Click "Save scoring criteria."

HubSpot now calculates scores for all contacts based on these criteria. Scores update automatically as contacts take actions.

Alternative: Custom scoring property

If you need multiple scoring models (e.g., separate scores for Product A vs Product B interest), create custom score properties.

Navigate to Settings > Properties > Contact properties > Create property.

Name: "Product A interest score." Type: Number.

Create workflows to increment this score based on Product A-specific behaviours:

• Visited Product A page: +10 points
• Downloaded Product A case study: +15 points
• Requested Product A demo: +30 points

Use "Increment value" action in workflows to add points. This gives you fine-grained control over scoring logic that goes beyond HubSpot's built-in scoring.

Balance demographic fit with behavioural engagement to avoid false positives (engaged leads who aren't a fit) and false negatives (perfect-fit leads who haven't engaged yet).

Setting up scoring criteria

Demographic scoring tends to be binary and stable: a contact either fits your ICP or doesn't, and this rarely changes.

Behavioural scoring is continuous and dynamic: engagement increases as contacts interact with your content, and can decrease if they disengage.

Weighting the balance:

High demographic fit + Low engagement = Qualified but cold lead (nurture until engagement increases)

Low demographic fit + High engagement = Engaged but unqualified lead (probably not a customer, but could be partner/influencer/researcher)

High demographic fit + High engagement = Hot lead (prioritise immediately)

Low demographic fit + Low engagement = Ignore (lowest priority)

Configure scoring so both dimensions matter. A lead needs 30+ demographic points AND 30+ behavioural points to reach your MQL threshold of 60 points. This prevents unqualified-but-engaged leads from becoming MQLs.

Score-based workflow triggers

Example weighted model:

Total score = Demographic score + Behavioural score

Demographic score max: 50 points (company size, industry, job title, geography)Behavioural score max: 50 points (web activity, email engagement, content consumption)

Set thresholds:

• 30-59 points: Nurture lead (not yet MQL)
• 60-79 points: MQL (route to sales)
• 80+ points: Hot lead (immediate Slack alert to sales manager)

Create workflows for each threshold:

Workflow: Nurture threshold (30-59 points)

Enrollment trigger: "HubSpot score is greater than or equal to 30" AND "HubSpot score is less than 60" AND "Lifecycle stage is Lead."

Add to nurture sequence: Enroll in email nurture workflow designed to increase engagement.

Workflow: MQL threshold (60-79 points)

Enrollment trigger: "HubSpot score is greater than or equal to 60" AND "Lifecycle stage is Lead."

Actions:

1. Set property value: Lifecycle stage → MQL
2. Rotate record to owner (sales team)
3. Send Slack notification: "New MQL: {{contact.firstname}} {{contact.lastname}} (Score: {{contact.hubspot_score}})"

Workflow: Hot lead threshold (80+ points)

Enrollment trigger: "HubSpot score is greater than or equal to 80" AND "Lifecycle stage is any of (Lead, MQL)."

Actions:

1. Set property value: Lifecycle stage → MQL (if not already)
2. Set property value: Lead status → Hot lead
3. Send Slack notification to sales manager: "🔥 Hot lead: {{contact.firstname}} {{contact.lastname}} (Score: {{contact.hubspot_score}}). Requires immediate attention."
4. Create task for assigned owner: "Call this hot lead within 2 hours."

This creates automatic prioritisation based on score ranges.

Maintain and adjust scores

Lead scoring models require ongoing optimisation. What you think predicts conversion might not actually predict conversion.

Quarterly scoring review process:

1. Export closed-won customers from last quarter. Include their lead score at the time they became MQL.
2. Calculate average lead score of customers who converted. Example: average score was 72 when they became MQL.
3. Export closed-lost opportunities. Calculate their average lead score. Example: average score was 58 when they became MQL.
4. Compare: If closed-won leads scored significantly higher than closed-lost leads, your model is working. If scores are similar, your model isn't predictive.
5. Analyse which criteria matter most:
   • Do customers who attended webinars convert at higher rates than those who didn't? (If yes, increase webinar attendance points.)
   • Do company size scores correlate with win rates? (If yes, maintain or increase those points.)
   • Do email opens correlate with win rates? (Often they don't - people open emails but don't buy. Consider reducing email open points.)
6. Adjust scoring criteria based on findings. Increase points for behaviours that correlate with wins. Decrease points for behaviours that don't.
7. Document changes and re-evaluate next quarter.

This data-driven approach improves your scoring model over time, making it more predictive of actual conversions.

‍

Use lead scores to trigger automated actions at key thresholds.

Set up scoring criteria

We've already configured scoring criteria in Section 1. Now use those scores to drive automation.

Score increase workflows:

When scores increase past thresholds, trigger actions:

Workflow: Score reached 40 (mid-level engagement)

Enrollment trigger: "HubSpot score changed to greater than or equal to 40" AND "Lifecycle stage is Lead."

Action: Send targeted email: "Based on your interest in [topic they engaged with], here's a relevant case study."

Workflow: Score reached 70 (MQL threshold)

Enrollment trigger: "HubSpot score changed to greater than or equal to 70" AND "Lifecycle stage is Lead."

Actions:

1. Set lifecycle stage → MQL
2. Assign owner (rotate or territory-based)
3. Send internal notification
4. Create task for owner: "Reach out to new MQL within 24 hours"

Score decrease workflows:

If scores decrease (due to inactivity or negative signals), trigger different actions:

Workflow: Score dropped below 40 (disengagement)

Enrollment trigger: "HubSpot score changed to less than 40" AND "Lifecycle stage is MQL" AND "Last engagement date is more than 60 days ago."

Actions:

1. Set lifecycle stage → Lead (downgrade from MQL)
2. Send internal notification to owner: "{{contact.firstname}} disengaged - downgraded to Lead."
3. Remove owner (set to unknown) so they don't clutter active MQL lists
4. Add to re-engagement nurture campaign

This prevents stale MQLs from clogging your sales pipeline.

Behavioural vs demographic scoring

Balance immediate actions (triggered by behavioural score spikes) with strategic routing (based on demographic fit).

Workflow: Hot behaviour spike

Enrollment trigger: "HubSpot score increased by 20 points or more in the last 7 days" AND "Visited pricing page within last 7 days."

This catches sudden engagement spikes - someone who was dormant suddenly visited multiple high-intent pages in a week.

Actions:

1. Send Slack notification: "{{contact.firstname}} showing high buying intent - score increased by {{score increase}} in last week"
2. Create high-priority task for owner: "Call immediately - recent spike in engagement"

Workflow: Perfect fit but low engagement

Enrollment trigger: "Number of employees is greater than 500" AND "Industry is Technology" AND "HubSpot score is less than 30" AND "Created date is more than 30 days ago."

This catches high-value prospects who fit your ICP perfectly but haven't engaged much yet.

Actions:

1. Add to targeted ABM campaign
2. Notify account-based marketing team: "High-value target needs outreach: {{contact.company}}"
3. Research and personalise outreach (manual step for sales)

This ensures perfect-fit prospects don't get ignored just because they're not clicking emails.

Lead scoring isn't "set and forget." Continuously optimise based on actual conversion data.

Quarterly review process

Month 1 of quarter: Collect data.

Export contacts who became customers this quarter. Include fields:

• Lead score when they became MQL
• Lead score when they became SQL
• Lead score when they became Opportunity
• Time from MQL to Customer
• Close reason

Export contacts who became closed-lost this quarter. Include same fields plus Loss reason.

Month 2 of quarter: Analyse data.

Calculate correlations:

• What was average lead score of closed-won vs closed-lost?
• Which scoring criteria appeared most frequently in closed-won customers?
• Which high-scoring leads didn't convert? (False positives - score was high but they didn't buy. Why?)
• Which low-scoring leads did convert? (False negatives - score was low but they bought anyway. What signals did we miss?)

Create pivot tables or use HubSpot's attribution reports to identify patterns.

Month 3 of quarter: Adjust scoring.

Based on analysis, modify scoring criteria:

Increase points for criteria that predict wins:Example finding: "Customers who attended webinars converted at 3x higher rate."Action: Increase "Webinar attendance" from +15 points to +25 points.

Decrease points for criteria that don't predict wins:Example finding: "Email open rate showed no correlation with conversion."Action: Reduce "Email opened" from +1 point to +0.5 points.

Add new criteria you discovered matter:Example finding: "Customers who visited the integration page converted at higher rates."Action: Add new rule: "Visited /integrations page: +10 points."

Remove or reduce negative scoring that's too aggressive:Example finding: "Some customers used Gmail addresses initially but still converted."Action: Reduce "Free email domain" penalty from -10 points to -5 points.

Re-calibrate thresholds:

If most customers score 75+ at MQL stage but your threshold is 60, you're wasting sales time on leads who won't convert.

Raise threshold: MQL = 70+ points instead of 60+ points.

If many customers scored only 50-60 at MQL stage and still converted, your threshold might be too high and you're missing good leads.

Lower threshold: MQL = 50+ points instead of 60+ points.

Track changes over time:

Document all scoring changes in a shared document:

• Date changed
• What changed (criteria added/removed, point values adjusted, thresholds changed)
• Why changed (data insight that drove the change)
• Expected impact

This creates a history of your scoring model evolution and helps you understand whether changes improved predictiveness.

A/B test scoring changes:

If you're uncertain whether a change will improve things, A/B test it:

Create two scoring properties: "HubSpot score" (existing) and "Test score" (new model).

Run both models simultaneously for a month. Compare which model better predicts conversions.

After validation, implement the winning model for all leads.

‍

Your lead scoring model now quantifies sales-readiness based on demographic fit and behavioural engagement. Scores trigger automated workflows at key thresholds, routing hot leads immediately while nurturing lower-scoring prospects. You've established a quarterly review process to continuously refine scoring based on actual conversion data.

Lead scoring transforms subjective lead quality judgements into objective, measurable criteria. Sales reps know which leads to prioritise. Marketing knows which campaigns generate the highest-scoring leads. Everyone speaks the same language about lead quality.

Next, we'll configure marketing dashboards and reports to measure campaign effectiveness and prove marketing's impact on revenue.