sales
Lead Scoring
Numerical system that prioritizes leads based on fit and intent signals.
Definition
Lead scoring assigns each lead a numeric score (typically 0-100) combining fit criteria (industry, company size, role) and intent signals (page views, content downloads, demo requests, time on pricing page). Above a threshold, the lead becomes an SQL and gets routed to sales. The split: typically 50% fit, 50% intent. Avoid over-engineering - 5-10 scoring criteria is plenty. Re-weight quarterly based on which scored leads actually converted.
Fit scoring criteria for US B2B
Fit signals match the lead against ICP. Common firmographic criteria with sample point values. Company size in employees (50 to 500 worth 30 points, under 50 worth 10, over 500 worth 5). Industry vertical (target industries 25 points, adjacent 10, off-target 0). Geography (US 20 points, EN-speaking adjacent 10, others negative). Annual revenue (matching ICP band 20 points). Tech stack alignment (uses complementary tools 15 points). Role of contact (decision maker 25 points, influencer 15, end user 5). Total fit score 0 to 100. Re-evaluate weights quarterly based on actual conversion data. Roles and industries that close should weigh more; ones that consistently lose should weigh less.
Intent scoring with behavior signals
Intent signals capture behavior over a recency window (typically 30 to 90 days). Sample point values. Visited pricing page (25 points). Downloaded BOFU asset like case study or ROI calculator (20 points). Attended webinar or demo (30 points). Requested demo or trial (50 points). Opened email in last 7 days (5 points). Clicked email link in last 7 days (10 points). Multiple sessions in same week (15 points). Time on site over 5 minutes in single session (10 points). Engaged with sales rep (20 points). Decay factor: scores decrease 10 percent per week of inactivity. Total intent score 0 to 100. The decay prevents stale leads from accumulating scores indefinitely.
Implementing scoring in HubSpot and Salesforce
HubSpot Marketing Hub Pro and above includes native predictive lead scoring plus manual scoring rules. Salesforce uses Pardot or Marketing Cloud Account Engagement for scoring with separate fit and intent fields. Both platforms support workflow automation: when score thresholds are crossed, leads route to sales, marketing nurture, or disqualification. The setup steps. One, define scoring criteria with marketing and sales agreement. Two, configure rules in platform. Three, test with historical data (would past closed deals have scored as MQL? would past disqualifications have scored low?). Four, deploy with weekly accuracy review. Five, refine quarterly.
Avoiding lead scoring failure modes
Five failure patterns to avoid. One, over-engineering: 30 plus scoring criteria that no one can validate against reality. Two, set-it-and-forget-it: scoring rules deployed in 2022 still running unchanged in 2026, with no quarterly refinement. Three, fit-only or intent-only scoring: missing one dimension hides the other. Four, no decay: stale leads accumulate scores from old behavior and clutter the MQL pool. Five, scoring without action: leads score as MQL but routing and SLA enforcement are missing, so the MQL designation has no operational consequence. The simplest functional scoring (5 to 10 criteria, fit and intent separated, monthly review, automated routing) outperforms sophisticated scoring with poor execution every time.
FAQ
Should I use predictive lead scoring or rule-based?
Rule-based for under 500 leads per month or under 50 closed deals annually. Predictive (machine learning) for higher volume where the algorithm has enough data to train on. Rule-based is more transparent (you know why a lead scored high), easier to refine, and works with limited data. Predictive can catch patterns humans miss but requires substantial historical data and is harder to debug when scores feel wrong. For US small business under 5M revenue, rule-based scoring in HubSpot or similar is the practical answer. For 10M plus with consistent high lead volume, predictive scoring starts to deliver real value.
How often should I refine lead scoring rules?
Quarterly minimum, monthly if volume is sufficient. The refinement process. One, pull closed-won deals from the last 90 days and check their scoring at the time of MQL. Two, pull disqualified MQLs from the last 90 days and check whether scoring should have caught them. Three, identify scoring criteria that consistently predict win or loss. Four, adjust weights or add new criteria based on findings. The process takes 2 to 4 hours per quarter and produces meaningful improvement in MQL quality. Without refinement, scoring drifts out of alignment with reality within 6 to 12 months.
What MQL threshold should I set?
Calibrate so 5 to 15 percent of new leads cross the MQL threshold. Lower than 5 percent means thresholds are too strict and you are leaving qualified leads in nurture too long. Higher than 15 percent means thresholds are too loose and sales is wasting time on weak leads. The exact threshold depends on your scoring rule weights; tune by adjusting threshold until MQL volume matches sales capacity to follow up. The goal: MQL volume that sales can work fully within SLA, not so much that SLA misses become common.
Can I score leads without marketing automation software?
Yes, with limitations. Basic firmographic scoring (industry, company size, role) can be implemented in any CRM via custom fields. Behavior scoring (page views, downloads, email engagement) requires marketing automation to capture the data. Without behavior data, scoring is fit-only and misses the intent dimension. The cost of HubSpot Marketing Hub Pro (under 1000 dollars per month for most US small businesses) is justified by the operational lift from full scoring. CRMs without marketing automation (Pipedrive base, Salesforce without Pardot) typically lack the data infrastructure for effective intent scoring.
Should sales reps see lead scores?
Yes, prominently in the CRM lead and contact views. Reps prioritize based on score visibility; without it, they default to recency or alphabetical order. Display fit score, intent score, and total score with clear thresholds. Some teams use red, yellow, green color coding for instant prioritization. Visibility drives behavior change: reps work high-scored leads first, which lifts overall conversion. Hidden scoring (used only for routing automation) misses this productivity benefit.
In your business
- →Score on fit (50%) + intent (50%)
- →Re-weight quarterly based on actual conversion data
- →Don't over-engineer - 5-10 criteria is plenty