Lead Scoring Examples That Actually Change How Reps Prioritize
Most lead scoring models give you a number nobody trusts. Reps glance at the score, ignore it, and call whoever emailed them last. That's a model problem, not a rep problem. The scoring criteria were set once during a HubSpot onboarding call two years ago and nobody has touched them since.
Good lead scoring examples have two things in common. The signals are specific enough to mean something. And the scores decay when a lead goes cold. Without decay, a lead who downloaded a whitepaper in March is still sitting at 85 points in October, looking warm when they're stone dead.
Lead Scoring Examples by Signal Type
Behavioral signals predict intent. Demographic fit doesn't. Firmographic data alone doesn't either. Behavior does. Here's how to score it with some specificity.
Pricing page visit: +20 points. This is the single highest-intent page on most B2B sites and it's consistently underweighted. Someone hits pricing twice in a week? That's a conversation waiting to happen. A rep who doesn't call within 24 hours of that second visit is leaving money on the table.
Demo request form submission: +50 points and immediate MQL status, full stop. The lead told you they want to see the product. No nurture sequence needed. Route it.
Opened 3+ emails in a sequence without clicking: +5 points, maybe. They're reading but not ready. Don't over-score passive behavior. A lot of scoring models treat opens like intent. Consistent opens with no clicks usually means the person is curious and nowhere near a buying cycle yet.
Attended a live webinar versus just registered: +15 points difference. Registration is almost meaningless at this point. Attendance means something. Staying for Q&A means more.
Job title match to ICP: +10 to +25 points depending on seniority. A VP of Sales at a 200-person SaaS company is worth more than the same VP at a five-person agency, which is where negative scoring on company size earns its keep. Give points, take points. The model should net out.
Lead Scoring Examples That Include Negative Scores
Negative scoring is the part most teams skip. It's also the part that keeps reps from burning half their week on leads who will never buy.
Competitor domain in email: -50 points immediately. Someone from a rival company poking around your site is doing research, not buying. This should disqualify a lead from MQL status automatically.
No activity in 45 days: -30 points. That's your decay. The exact number matters less than the principle. Leads go stale. A score that ignores time since last engagement is a frozen timestamp, not a signal.
Job title is student, intern, or analyst at a sub-50-person company for an enterprise product: -20 points. Sounds harsh. It's accurate. If your ACV is $40k, a marketing intern isn't your buyer.
Only visited the careers page: -10 points. They're looking for a job, not a product. These leads clog pipelines and make dashboards look healthier than they are.
How to Calibrate Your Lead Scoring Model
Build this backwards. Pull the last 50 closed-won deals from your CRM. Look at what those contacts did before they converted. Which pages did they visit? How many emails did they open? Did they request a demo or did a rep reach out cold? That activity pattern is your scoring rubric. You're reading good behavior off actual customers instead of guessing at it.
Then pull 50 leads who reached MQL status but never converted. Compare. If they hit the same behavioral thresholds as your closed-won customers, your scoring is fine and the problem lives in the handoff. If they look completely different, your thresholds are off.
This calibration exercise takes a half-day in HubSpot or Salesforce and most teams never do it. They set up scoring during implementation and treat it like infrastructure, when it's a living model. Quarterly reviews take 30 minutes. The return on that 30 minutes shows up in your conversion rates.
Frequently Asked Questions
What score threshold should trigger MQL status? There's no universal number. For most B2B models on a 0-100 scale, 50 to 60 is a reasonable starting point. Set the real threshold by looking at your historical conversion rates at each score band, not by pulling a figure off a blog post.
Should I use AI-based predictive scoring instead of manual rules? Predictive scoring tools like the ones built into HubSpot's Enterprise tier are genuinely useful once you have enough data, typically 500+ closed deals minimum. Below that, rule-based scoring with good negative signals beats a model trained on too little history. AI automation can sharpen lead scoring later. You need a baseline first.
Why don't my reps trust the lead score? Usually because it's been wrong enough times that they learned to tune it out. Close the loop: when a rep works a high-score lead that doesn't convert, find out why and adjust the model. Reps trust scores that visibly improve from their feedback.
A scoring model your sales team actually uses is less about the perfect formula and more about how RevOps routes and prioritizes leads in practice. A score no one acts on is a vanity metric.
