Lead scoring: complete guide
How to prioritise leads so your team focuses on those most likely to buy
Not all leads are equal. Some are ready to buy; others are just browsing. Lead scoring is a rating system that automatically ranks your leads by conversion probability, so your sales team focuses on those with the highest potential.
Without lead scoring, reps waste time on cold leads while hot opportunities cool down unattended. With a well-configured system, marketing only sends sales the leads that meet minimum qualification criteria.
What is lead scoring?
Lead scoring assigns a numerical score to each lead based on two dimensions: who they are (demographic and firmographic profile) and what they do (digital behaviour). The combination of both dimensions determines the likelihood of that lead becoming a customer.
When a lead reaches a predefined threshold score, it’s marked as "sales qualified" (SQL) and automatically assigned to a rep. This removes subjectivity from the qualification process and aligns marketing and sales with objective criteria.
Demographic and firmographic criteria
These criteria evaluate whether the lead matches your ideal customer profile (ICP). It doesn’t matter how interested a lead is if they don’t fit the profile for your product or service.
- Job title or role: a marketing director scores higher than an intern for a B2B product
- Company size: if your product targets SMEs of 10–50 employees, a 5,000-person company scores low
- Industry: if your solution fits retail best, retail leads score higher
- Geographic location: relevant if you operate in specific markets
- Declared budget: if captured via a form, it’s a direct indicator of intent
Behavioural criteria
Digital behaviour reveals intent. A lead who visits the pricing page three times in a week is more likely to buy than one who only read a blog post. Behavioural scoring captures these signals.
- Key page visits: pricing, demos, case studies, contact (+10–20 points each)
- Email opens and clicks: engagement with marketing campaigns (+5 points per click)
- Content downloads: whitepapers, guides, templates (+10–15 points)
- Demo or contact request: the strongest intent signal (+30–50 points)
- Extended inactivity: subtract points if no activity for 30+ days (−10 points)
Scoring models: manual vs predictive
Manual scoring is the starting point: you define rules and scores based on your commercial experience and analysis of won customers. It’s simple to implement but requires periodic review to adjust weights.
Predictive scoring uses machine learning to analyse patterns in your historical data and predict which leads are most likely to close. Tools like HubSpot, Salesforce Einstein or Madkudu offer predictive scoring that adjusts automatically as it accumulates data.
- Manual scoring: you define rules and weights. Best for teams starting out or with limited historical data
- Predictive scoring: the algorithm learns from your data. Best when you have 500+ leads with conversion history
- Hybrid scoring: combines manual rules with predictive adjustments. The most robust mid-term approach
Automating scoring in your CRM
Most modern CRMs include lead scoring capabilities. HubSpot offers it from its Professional plan, Salesforce with Einstein Lead Scoring, and Zoho with Zia. Typical configuration involves defining criteria, assigning scores and setting the qualification threshold.
- Define your SQL threshold: the minimum score for a lead to be passed to sales (typically 50–80 points)
- Create an automation: when the lead hits the threshold, assign to a rep and create a contact task
- Set up alerts: notification to the rep when a hot lead takes a high-value action
- Review and adjust: each quarter, analyse whether high-scoring leads actually convert at higher rates
Aligning marketing and sales with scoring
Lead scoring only works if marketing and sales agree on what a qualified lead means. This alignment requires a definition meeting where both teams agree on criteria, scores and thresholds.
A common mistake is for marketing to define scoring without sales input. The result: leads that marketing considers qualified but sales rejects. The solution is a continuous feedback loop: sales reports on the quality of leads received and marketing adjusts the model.
Common lead scoring mistakes
Implementing lead scoring poorly can be worse than not having it at all. If the criteria don’t reflect reality, reps lose trust in the system and revert to qualifying manually.
- Too many criteria: start with 5–8 clear criteria, not 30 vague variables
- Not subtracting points: inactivity and email bounces should reduce the score
- Not reviewing the model: scoring that isn’t adjusted quarterly loses accuracy fast
- Ignoring sales feedback: if reps reject "qualified" leads, the model needs revision
Key Takeaways
- Lead scoring combines profile (who they are) and behaviour (what they do) to prioritise leads
- Start with a manual model of 5–8 clear criteria and evolve toward predictive scoring
- Define a qualification threshold (SQL) agreed upon by both marketing and sales
- Automate assignment: when a lead hits the threshold, it should reach a rep within minutes
- Review the model quarterly with real conversion data to maintain its accuracy
Want to implement lead scoring in your CRM?
We design and configure a lead scoring system tailored to your sales process and buyer persona. No strings attached.