Lead Scoring 101: Rank Your Pipeline by Likelihood to Buy

Why most pipelines leak revenue at the top

Walk into almost any small sales team and you'll find reps working their list top to bottom, oldest lead first, treating a tire-kicker the same as a buyer with a budget and a deadline. That feels fair. It's also why deals slip. When every lead gets equal attention, your best opportunities wait in line behind your worst ones, and the prospect who was ready to sign on Tuesday has gone cold by Friday.

Scoring fixes the ordering problem. Instead of a flat list, you rank prospects by how likely they are to buy, then spend your limited hours where the math says you'll win. The payoff is concrete: teams that prioritize by fit and intent routinely convert qualified leads at 2-3x the rate of an unsorted list, because reps stop spreading themselves thin. The point of generating more leads was never volume for its own sake. It's giving a scoring model enough raw material to surface the gold.

The two ingredients: fit and intent

Every useful score is built from two questions. First, fit: is this the kind of customer we're good for? Second, intent: are they showing signs of buying right now? A great-fit company that isn't looking is a nurture project. A red-hot lead that can't afford you is a waste of a demo. You want both lights green.

Fit is about who they are and rarely changes week to week. Intent is about what they're doing and can swing overnight. Keep them as separate scores so you can read the situation at a glance instead of blending everything into one mushy number.

• Fit signals: company size, industry, role/seniority of the contact, region, tech stack, budget band
• Intent signals: pricing-page visits, demo requests, email opens and replies, content downloads, return visits, free-trial activity

Build a points model in an afternoon

You don't need a data science team. Start with a 100-point scale split evenly: 50 points of fit, 50 points of intent. Assign weights to the handful of signals that actually predicted your last 20 closed deals, and ignore the rest. Resist the urge to score everything.

A starter rubric might look like this. The exact numbers matter less than being consistent and revisiting them once you have results.

• Decision-maker title (VP/owner): +20 fit
• Company in your core industry: +15 fit
• Right company size band: +15 fit
• Requested a demo or quote: +25 intent
• Visited pricing page twice in a week: +15 intent
• Opened 3+ emails, no reply: +10 intent
• Generic free email (gmail/yahoo) for a B2B product: -10 fit
• No activity in 30 days: -15 intent

Turn scores into actions, not decoration

A score that just sits in a column is useless. Tie ranges to behavior so the number triggers a clear next step. The classic split is three tiers: hot, warm, and cold.

Define thresholds and the playbook for each, then hold the team to it. Hot leads (say 70+) get a call within an hour and a named owner. Warm leads (40-69) go into a tight email sequence with a check-in in five business days. Cold leads (under 40) drop into automated nurture and re-enter the queue only if their intent score climbs. This is also where lead generation pays off twice: a steady top-of-funnel means your hot tier is never empty, so reps always have something worth calling.

• 70-100 Hot: personal outreach within the hour, book a meeting
• 40-69 Warm: structured sequence, value content, follow up in 5 days
• 0-39 Cold: automated nurture, revisit when intent rises

Tune it with real outcomes

Your first model will be wrong, and that's fine. After 30-60 days, pull your closed-won and closed-lost deals and check the scores they had when they entered the pipeline. If winners clustered at 55 instead of 70, lower your hot threshold. If a signal you weighted heavily (say, email opens) shows up just as often in losses, cut its points.

Watch two numbers to know it's working: win rate by tier (hot should clearly beat warm) and average days-to-close (prioritized leads should move faster). When those gaps are obvious, your model is earning its keep. Re-tune quarterly, or whenever you change your offer, pricing, or target market — a score calibrated for last year's customer will quietly misrank this year's.