The Problem With Treating Every Lead Equally
A CEO downloads your pricing guide. A student downloads the same guide for a university assignment. Your CRM treats them both as "new lead." Your sales team calls them both.
One converts into a $50,000 contract. The other doesn't answer the phone and never will.
Without lead scoring, your sales team wastes time on leads that were never going to buy, while genuinely interested prospects wait in the queue and lose interest. Research shows that 79% of marketing leads never convert to sales — largely because they're either unqualified or contacted at the wrong time.
Lead scoring is the system that separates the signal from the noise.
What Lead Scoring Actually Is
Lead scoring assigns numerical values to each lead based on two categories of information:
1. Who they are (demographic/firmographic fit)
- Job title and seniority
- Company size
- Industry
- Location
- Budget indicators
2. What they've done (behavioural engagement)
- Pages visited on your website
- Content downloaded
- Emails opened and clicked
- Form submissions
- Event attendance
- Social media engagement
Each action or attribute adds (or subtracts) points. When a lead reaches a threshold score, they're flagged as ready for sales contact.
Simple example:
| Action/Attribute | Points | |-----------------|--------| | Job title: Director or above | +20 | | Job title: Student or Intern | -15 | | Company size: 50+ employees | +15 | | Company size: Under 5 employees | +5 | | Visited pricing page | +25 | | Downloaded case study | +15 | | Opened 3+ emails in 30 days | +10 | | Attended webinar | +20 | | No activity in 30 days | -20 | | Unsubscribed from emails | -50 |
Threshold: Leads scoring 60+ are flagged as Marketing Qualified Leads (MQLs) and passed to sales.
The Two Scoring Dimensions
Explicit Scoring (Fit)
Based on information the lead provides or that you can look up. This tells you whether the lead matches your ideal customer profile.
High-value signals:
- Decision-maker title (CEO, Marketing Director, Head of Operations)
- Company in your target industry
- Company size within your sweet spot
- Located in your service area
- Stated budget within your range
Low-value / negative signals:
- Competitor email domain
- Student or intern title
- Company too small or too large for your services
- Located outside your service area
- Personal email (for B2B where company email signals seriousness)
Implicit Scoring (Interest)
Based on behaviour — what the lead does on your website, with your emails, and across your content. This tells you how interested they are right now.
High-intent behaviours:
- Visiting your pricing or services page (multiple times = very high intent)
- Requesting a quote or demo
- Downloading bottom-of-funnel content (case studies, comparison guides, ROI calculators)
- Replying to an email
- Returning to your site after a period of inactivity
- Visiting your "About" or "Team" page (evaluating whether to work with you)
Medium-intent behaviours:
- Downloading educational content (guides, ebooks)
- Attending a webinar
- Following on social media
- Opening emails consistently
- Reading multiple blog posts in one session
Low-intent behaviours:
- Single blog post visit
- Opening one email
- Visiting the homepage once
- Social media like (without deeper engagement)
Building Your First Scoring Model
Step 1: Define Your Ideal Customer
Before assigning points, get crystal clear on who your best customers look like.
Look at your last 20 closed deals:
- What job titles made the buying decision?
- What industries were they in?
- What was the company size?
- How long was the sales cycle?
- What content did they engage with before buying?
- What was the first touchpoint?
Patterns will emerge. Those patterns become your scoring criteria.
Step 2: Map the Buying Journey
Identify the behaviours that signal progression through your funnel:
Awareness stage: Blog visits, social engagement, educational downloads
Consideration stage: Case study downloads, pricing page visits, webinar attendance, repeat visits
Decision stage: Quote requests, demo bookings, team page visits, comparison content downloads, direct contact
Assign more points to behaviours closer to the decision stage.
Step 3: Assign Point Values
Start simple. You can refine later.
Demographic scoring (fit):
| Criteria | Perfect Fit | Good Fit | Okay Fit | Poor Fit | |----------|------------|----------|----------|----------| | Job title | +25 | +15 | +5 | -10 | | Industry | +20 | +10 | +5 | -5 | | Company size | +15 | +10 | +5 | -10 | | Location | +10 | +5 | 0 | -5 |
Behavioural scoring (interest):
| Action | Points | |--------|--------| | Requested quote/demo | +40 | | Visited pricing page | +25 | | Downloaded case study | +20 | | Attended webinar | +20 | | Downloaded ebook/guide | +10 | | Visited services page | +10 | | Opened email | +3 | | Clicked email link | +7 | | Read blog post | +5 | | Social media follow | +3 | | No engagement 30 days | -15 | | No engagement 60 days | -30 |
Step 4: Set Thresholds
| Score Range | Status | Action | |-------------|--------|--------| | 0-20 | Cold | Nurture with content | | 21-40 | Warming | Increase email frequency, targeted content | | 41-60 | Marketing Qualified (MQL) | Alert sales for potential outreach | | 61-80 | Sales Qualified (SQL) | Priority sales follow-up | | 81+ | Hot | Immediate sales contact |
Step 5: Build in Decay
Scores should decrease over time if the lead goes inactive. A lead who scored 75 six months ago but hasn't engaged since isn't a hot lead anymore.
Time-based decay options:
- Subtract 5 points per week of inactivity
- Halve the behavioural score after 90 days of no engagement
- Reset to baseline after 6 months of zero activity
Decay prevents stale leads from clogging your sales pipeline.
Implementing Lead Scoring
In Your CRM/Marketing Automation Platform
Most modern platforms have built-in lead scoring:
- HubSpot: Built-in scoring with predictive lead scoring on higher tiers
- ActiveCampaign: Contact scoring with automation triggers
- Salesforce: Einstein Lead Scoring (AI-driven)
- Pipedrive: Smart scoring features
- Mailchimp: Tags and segments can approximate scoring
- Monday CRM: Customisable lead scoring rules
Automation Triggers
Once scoring is set up, automate the handoffs:
- Lead reaches MQL threshold → Notify sales via Slack/email, create task in CRM
- Lead visits pricing page 3+ times → Trigger personalised email from sales rep
- Lead score drops below threshold → Move back to nurture sequence
- Lead requests demo → Instant notification to sales + calendar booking link
The Sales-Marketing Handoff
This is where most lead scoring implementations fail — not in the scoring itself, but in what happens after.
Define clearly:
- At what score does sales receive the lead?
- How quickly must sales follow up? (Research shows 5 minutes is optimal)
- What information does sales receive? (Score, key behaviours, company info)
- What happens if sales doesn't convert the lead? (Feedback loop back to marketing)
- What feedback does sales give marketing about lead quality? (Critical for refining the model)
AI-Powered Lead Scoring
In 2026, AI lead scoring is increasingly accessible:
How it works: Machine learning analyses your historical conversion data and identifies patterns humans miss — combinations of behaviours, timing, and attributes that predict conversion.
Advantages over manual scoring:
- Discovers non-obvious patterns
- Adapts automatically as buyer behaviour changes
- Removes human bias from scoring
- Processes more signals simultaneously
When to use AI scoring:
- You have 500+ leads in your database
- You have at least 50 closed deals to train the model
- Your manual scoring model has plateaued in accuracy
- You want to scale beyond what manual rules can handle
When manual scoring is better:
- Small lead volume (under 200)
- New business without historical data
- Simple, clear buying patterns
- You need full control and explainability
Measuring and Refining Your Model
Key Metrics
- MQL-to-SQL conversion rate: What percentage of marketing qualified leads become sales qualified? Target: 30-50%
- SQL-to-customer rate: What percentage of sales qualified leads close? Target: 15-30%
- Average score of closed deals: What score did your actual customers reach before buying?
- Average score of dead leads: At what score range do leads that never buy tend to plateau?
- Sales acceptance rate: What percentage of MQLs does sales agree are worth pursuing?
Quarterly Reviews
Every quarter, sit down with sales and ask:
- Are the leads coming through at the right time? (Too early? Too late?)
- Are we passing leads that shouldn't have qualified?
- Are good leads being missed or scored too low?
- Have buying behaviours changed? (New content types, new pages that signal intent?)
Adjust point values based on real data. The first version of your model will be wrong — that's expected. The value is in iterating.
Common Red Flags
- Too many MQLs, too few conversions: Threshold is too low or scoring is too generous
- Too few MQLs: Threshold is too high or scoring is too stingy
- Sales ignoring MQLs: Either lead quality is poor (scoring problem) or sales doesn't trust the system (communication problem)
- High scores but low engagement: Demographic scoring is weighted too heavily vs. behavioural
Common Mistakes
- Over-complicating v1 — Start with 5-10 scoring criteria, not 50. Complexity kills adoption.
- Scoring without decay — A lead from 18 months ago with a high score is misleading. Build in time-based decay.
- Not involving sales — If sales doesn't help build the model, they won't trust the output.
- Scoring everything equally — A pricing page visit is worth far more than a blog post read. Weight accordingly.
- Set and forget — Buyer behaviour changes. Your model must change with it.
- No negative scoring — Competitor visits, unsubscribes, and inactivity should subtract points.
- Ignoring the feedback loop — Sales must report back on lead quality. Without this, the model can't improve.
- Treating the score as the only signal — A lead with a score of 45 who just called your office is more important than a lead with a score of 80 who hasn't done anything in weeks. Scoring is a tool, not a replacement for judgement.
Start Here
- Analyse your last 20 closed deals — who were they, what did they do before buying?
- List 5 demographic criteria and 5 behavioural criteria
- Assign simple point values (start with 5, 10, 15, 20, 25 increments)
- Set your MQL threshold (start at 50 and adjust)
- Build the scoring model in your CRM or marketing platform
- Add decay rules for inactivity
- Set up automated notifications when leads hit the threshold
- Review with sales after 30 days and adjust
Lead scoring doesn't have to be perfect from day one. It just has to be better than treating every lead the same — which is a very low bar to clear. Even a basic model will help your sales team spend their time on the people most likely to become customers.