Lead Scoring: What It Is, Models, Criteria & Examples (2026)
Lead scoring explained: what it is, the main lead scoring models, criteria and a worked example, plus how to build a B2B model on engagement signals.
TL;DR: Lead scoring is a method of ranking leads by assigning points for who they are (fit) and what they do (behavior), so sales works the prospects most likely to buy first. A good B2B lead scoring model combines fit criteria from your ICP, behavioral and engagement signals like comments, replies and repeat visits, and negative scoring and decay so stale or poor-fit leads drop down the list. Start with a simple points table, set thresholds for MQL and SQL, and recalibrate against closed-won deals every quarter.
What Is Lead Scoring?
Lead scoring is the process of assigning a numerical value to each lead based on how well they match your ideal customer profile and how strongly they are showing buying intent. The score tells sales and marketing who to contact now, who to nurture, and who to ignore.
Every lead scoring model answers two questions:
Fit: should we sell to them? Company size, industry, role, seniority, geography, tech stack.
Intent: are they in-market right now? What they do: engaging with your content, replying, attending events, visiting high-intent pages, multiple people from the same account showing up.
A lead with high fit and low intent is a nurture target. A lead with high intent and poor fit is noise. A lead with both is where sales should spend today.
Why Lead Scoring Exists: The Problem Before It
Without lead scoring, B2B teams fall into the same pattern:
Every lead looks equal. Reps work the list top to bottom, or by whoever filled out a form most recently.
Sales and marketing disagree on what "qualified" means. Marketing hands over volume; sales complains about quality.
High-intent buyers wait. A decision-maker who has engaged five times this week sits behind fifty cold contacts.
Signals outside the website are invisible. Traditional scoring relies on form fills, email opens and page views, and misses the research buyers do on LinkedIn and in communities before they ever visit your site.
Lead scoring fixes the first three by agreeing on a shared, numeric definition of a good lead. Engagement-based scoring fixes the fourth.
Lead Scoring Models: The Main Types
Most B2B lead scoring models are a combination of the following. Use this as a menu, not a checklist. Start with two and add more once you have data.
Model | What it scores | Best for | Watch out for |
|---|---|---|---|
Fit (demographic / firmographic) scoring | Role, seniority, company size, industry, region, tech stack | Filtering out leads you'd never sell to | Says nothing about timing |
Behavioral scoring | Website visits, pricing page views, email clicks, content downloads, demo requests | Teams with strong inbound traffic | Only sees activity on your own properties |
Engagement scoring | Social interactions: post reactions, comments, reposts, profile views, DMs, connection requests, event RSVPs | B2B teams selling to buyers who research on LinkedIn | Needs a way to capture signals at the person level |
Account-level scoring | Combined activity of everyone at one company; buying-committee breadth | ABM and enterprise deals | Can hide which individual to contact |
Predictive scoring | Statistical or machine-learning model trained on past wins and losses | Teams with large volumes of historical closed-won data | Black box; weak with small datasets |
Negative scoring | Disqualifiers: competitors, students, job seekers, unsubscribes, poor-fit industries | Every model, since it keeps the top of the list clean | Easy to forget |
For most B2B teams under a few hundred deals a year, the practical choice is a rules-based model that combines fit + engagement + negative scoring. Predictive scoring needs more closed-won history than most teams have.
Lead Scoring Criteria: What to Score
Fit criteria (who they are)
Define fit from your ideal customer profile template, then turn it into points or a multiplier:
Job title and seniority: does this person own the problem or the budget?
Department: sales, marketing, RevOps, founders, etc.
Company size: employee range or revenue band you sell into
Industry: verticals where you win
Geography: regions you can actually serve
Tech stack: tools that signal readiness (e.g., a CRM you integrate with)
Behavioral and engagement criteria (what they do)
Not every action carries the same intent. Group signals by how much effort the prospect put in:
Signal category | Examples | Intent level |
|---|---|---|
Passive consumption | Post impressions, article reads, single page view | Low |
Light engagement | Likes and reactions, email opens | Low–medium |
Active engagement | Comments, reposts with commentary, poll votes, content downloads | Medium–high |
Direct engagement | DMs, connection requests with a note, event RSVPs, email replies | High |
Buying signals | Pricing page visits, demo requests, several people from one account engaging in the same week | Very high |
Signals compound. One comment means little; a prospect who viewed your profile, commented on two posts and then sent a connection request is showing a pattern. That's the core idea behind engagement signals.
Negative criteria (who to push down)
Competitors, vendors, students and job seekers
Companies outside your ICP size or region
Unsubscribes, bounced emails, "not interested" replies
No new activity in 30+ days (handled with decay, below)
Lead Scoring Example: A Starter Points Table
Here is a lead scoring example you can copy and adapt. These are starting weights. Recalibrate after 60–90 days against what actually converts.
Signal | Points | Rationale |
|---|---|---|
Post impression (viewed your content) | +1 | Awareness only |
Post like / reaction | +2 | Low effort but intentional |
Profile view | +3 | Active interest in who you are |
Repost without commentary | +3 | Endorsement signal |
Comment (generic) | +5 | Took time to engage publicly |
Connection request | +5 | Wants ongoing access to your content |
Repost with commentary | +8 | Strong endorsement and amplification |
Comment that asks a question | +10 | Signals active evaluation |
Connection request with a note | +10 | Intentional relationship-building |
Event or webinar RSVP | +10 | Committed time |
DM or email reply initiated by prospect | +15 | Direct buying signal |
Engaged with 3+ posts in 7 days | +15 (bonus) | Sustained interest pattern |
2+ people from the same company engaging | +20 (bonus) | Buying committee is active |
Competitor, student or job seeker | −50 | Negative score: will not buy |
Fit multiplier: multiply the engagement score by 1.5× for ideal ICP fit, 1× for acceptable fit and 0.5× for poor fit. A comment from a VP at a target account should outrank the same comment from someone outside your market.
Decay rule: subtract 20% of accumulated points every 30 days without new activity, so stale leads don't sit at the top forever.
Worked example
A Head of Sales at a 120-person SaaS company (ideal fit, 1.5×) likes two of your posts (+4), comments with a question on a third (+10), views your profile (+3) and triggers the 3-posts-in-7-days bonus (+15). Raw score: 32. With the fit multiplier: 48, an MQL under the thresholds below. If a colleague from the same company comments the next day, the account bonus (+20 × 1.5) pushes them to 78: hot, same-day outreach.
Lead Scoring Thresholds: What Each Score Means
Score range | Status | Action |
|---|---|---|
0–10 | Cold | Keep nurturing with content. No outreach. |
11–30 | Warming | Add to a targeted content sequence. Watch for acceleration. |
31–50 | Marketing qualified (MQL) | Notify sales. Rep reviews profile and recent activity before reaching out. |
51–75 | Sales qualified (SQL) | Priority outreach within 48 hours, referencing the specific engagement. |
76+ | Hot | Same-day outreach. Sustained, multi-signal intent. |
Thresholds depend on sales cycle, deal size and lead volume. A team selling $500/month software will set lower thresholds than one selling six-figure contracts. Review them every quarter.
Traditional Lead Scoring vs. Engagement-Based Lead Scoring
Aspect | Traditional lead scoring | Engagement-based lead scoring |
|---|---|---|
Main inputs | Form fills, email opens and clicks, website visits | Social engagement (comments, reactions, DMs, profile views) plus fit |
When intent shows up | Late, once the buyer reaches your site | Earlier, while the buyer is still researching |
Who it sees | Known contacts in your database | New people engaging with your team's and your market's content |
Account view | Usually per contact | Naturally shows multiple people from one account |
Context for outreach | "You downloaded our ebook" | "You asked about X in the comments on Y's post" |
Main risk | Misses buyers who never fill out a form | Noise from casual engagers without fit filtering |
The best B2B models use both: traditional behavioral data for buyers who come to you, and engagement data for the ones still researching elsewhere.
How to Build a Lead Scoring Model in 5 Steps
Define fit first. Write down your ICP (roles, company size, industries, regions) and decide your fit multiplier or fit points. Disqualify obvious non-buyers with negative scores.
Pick 8–12 signals and weight them. Use the starter table above. Weight by effort: a question in the comments beats a like; a DM beats a comment. Add bonuses for repeat engagement and multiple people from one account.
Set thresholds and owners. Agree with sales on what score makes an MQL and an SQL, who gets notified, and how fast they follow up (48 hours for SQL, same day for hot).
Wire it into your workflow. Manual spreadsheet for under ~50 active prospects; CRM custom score fields and alerts for a few hundred; automated signal capture beyond that. See our guide to LinkedIn CRM integrations for the CRM route.
Close the loop every 90 days. Pull closed-won and closed-lost deals, look at their scores at first sales touch, and adjust: raise weights on signals that preceded wins, cut signals that didn't, and move thresholds if MQLs are too noisy.
Common lead scoring mistakes
Over-weighting a single action. One comment on a viral post doesn't make a buyer. Require patterns.
No negative scoring or decay. Leads that only gain points never leave the top of the list.
Scoring everyone the same regardless of fit. Engagement from outside your ICP should count for less.
Ignoring dark social. DMs and private shares are among the highest-intent signals. Have reps log them, even manually at first.
Set and forget. Buyer behavior and your content change; your model should too.
Frequently Asked Questions
What is lead scoring in simple terms?
Lead scoring is giving each lead a number that reflects how good a fit they are and how interested they seem, so your team contacts the best leads first. Points come from fit criteria like role and company size, and from actions like comments, replies, demo requests and pricing page visits.
What are the most common lead scoring models?
The most common are fit (demographic and firmographic) scoring, behavioral scoring, engagement scoring, account-level scoring, predictive scoring and negative scoring. Most B2B teams combine fit, engagement and negative scoring in a rules-based model, and only move to predictive scoring once they have enough historical deal data.
What tools help prioritize B2B prospects based on recent engagement and fit?
You need two things: a way to capture engagement at the person level, and a way to apply fit rules to it. CRMs like HubSpot and Salesforce can hold custom scores and alerts, while engagement-intelligence tools such as traxy capture LinkedIn engagement automatically, match each person to your ICP and surface the ones showing buying patterns.
How do I prioritize prospects who engage with several relevant LinkedIn profiles?
Treat engagement across several relevant profiles (yours, your team's, industry voices in your space) as a stronger signal than engagement with one post. Give a bonus for repeat engagement within a short window (for example +15 for 3+ engagements in 7 days) and another for multiple people from the same account, then apply your fit multiplier.
How does lead scoring differ from lead qualification?
Lead scoring is the numeric ranking that decides who to look at first; lead qualification is the judgment, often a conversation using criteria like budget, authority, need and timing, that confirms whether a lead is a real opportunity. Scoring feeds qualification: high scores get qualified first. For the qualification step itself, see how to qualify B2B leads from buying signals.
How often should I update my lead scoring model?
Review weights and thresholds every quarter, and do a deeper recalibration against closed-won and closed-lost deals every 90 days in the first year. Also revisit the model whenever your ICP, pricing or main channels change.
Score on Signals, Not Guesswork
Lead scoring works when it reflects how your buyers actually behave, and for most B2B buyers, that research happens in public on LinkedIn long before a form fill. Start with the fit criteria and points table above, add decay and negative scoring, and recalibrate against real deals every quarter. If you want the engagement side captured and scored for you, traxy turns LinkedIn engagement from ICP-matched buyers into prioritized, qualified leads, so your team spends its time on outreach instead of spreadsheets.
Keep reading
All Buyer Intent & Signals guides
Buyer Intent Signals (2026): What Buying Signals Are, Examples by Type and How Sales Teams Act on Them
Buyer intent signals explained: what buying signals are, examples by type, how reps act on each one and a simple scoring framework to prioritize accounts.
B2B Buying Signals: How to Spot the Dark Funnel
B2B buying signals explained: what they are, how the dark funnel hides them, and how to turn anonymous intent into pipeline in 2026.
How to Qualify B2B Leads From Buying Signals
A B2B lead qualification framework for buying signals: score signal strength and ICP fit step by step, and know exactly when to route a lead to sales.