Lead Scoring That Actually Predicts Sales: Move Past Vanity Metrics to Conversion Signals

Your marketing automation platform scores leads on 47 criteria. Visited pricing page: 10 points. Downloaded whitepaper: 20 points. Spent 2+ minutes on site: 5 points.

A lead hits 100 points. They're ready for sales.

Sales gets them and says: "This lead visited the pricing page and filled out a form. That's it. They're not ready to sell to."

The score is useless. Not because the logic is wrong, but because the inputs are wrong.

You're scoring on engagement metrics (how interested they are). Sales needs qualification metrics (whether they can actually buy).

These aren't the same. And conflating them destroys the entire handoff.

The Lead Scoring Problem

Most marketing teams build lead scores around engagement:

• Page visits: 10 points each
• Email opens: 5 points each
• Form fills: 20 points
• Case study views: 10 points

This measures interest. But interest doesn't predict sales.

A person can be highly engaged (visiting your site 5 times, opening every email) and still not be able to buy:

• They have no budget
• They're not the decision-maker
• Their company isn't a fit for what you sell
• They're just researching for a competitor

Conversely, someone might have low engagement (visited once, never opened an email) but be perfectly ready to buy:

• They already know what they need
• They're just making a buying decision
• Someone referred them to you
• They're a perfect ICP fit

Lead scoring based on engagement will reject ready-to-buy leads and accept unqualified window-shoppers.

The Qualification Scoring Framework

Instead of engagement scoring, build qualification scoring around buyer attributes.

A lead scores high if they're likely to buy. Low if they're unlikely to buy. Regardless of engagement.

Here's the framework:

Company-level signals (40 points max):

• Company size: is it your ICP size? (0–15 points)
• Industry: is it in your target industries? (0–10 points)
• Growth stage: did they just raise funding or show growth signals? (0–10 points)
• Buying signal: did they post a job in a relevant role, or change executives? (0–15 points)

Role-level signals (30 points max):

• Title: is this person a decision-maker in their org? (0–15 points)
• Seniority: are they senior enough to control budget? (0–15 points)

Engagement signals (30 points max):

• Intent: did they show high intent (downloaded comparison guide, requested demo)? (0–15 points)
• Fit relevance: did they engage with content related to your solution? (0–15 points)

Total score: 0–100 points

A lead scoring 70+ is sales-ready. A lead scoring 30–50 needs nurture. A lead scoring <30 shouldn't get sales attention.

Real-World Example

Lead A:

• Company: £3M ARR SaaS (ICP fit: 15 points)
• Industry: fintech (target industry: 10 points)
• Growth: raised Series B in last 12 months (signal: 10 points)
• Role: VP Sales (decision-maker: 15 points)
• Seniority: VP-level (senior enough: 15 points)
• Intent: downloaded RevOps whitepaper and requested a demo (high intent: 15 points)
• Engagement relevance: specifically looked at RevOps content (fit: 10 points)

Total score: 90

This lead is ready for sales. Send immediately.

Lead B:

• Company: £50M ARR company (not ideal ICP: 5 points)
• Industry: healthcare (not target industry: 0 points)
• Growth: no signal (0 points)
• Role: product manager (not a budget owner: 5 points)
• Seniority: mid-level (not senior: 5 points)
• Intent: opened three marketing emails (low intent: 3 points)
• Engagement relevance: visited blog, no specific area focus (fit: 2 points)

Total score: 20

This lead shouldn't go to sales yet. They need nurture. In 6 months, if they show buying signals, they might be ready.

Lead C:

• Company: £5M ARR SaaS (ICP fit: 15 points)
• Industry: SaaS (target: 10 points)
• Growth: no recent signals (0 points)
• Role: Head of Sales (decision-maker: 15 points)
• Seniority: director+ level (senior enough: 15 points)
• Intent: visited demo page but never requested (medium intent: 5 points)
• Engagement relevance: visited multiple RevOps pages (fit: 12 points)

Total score: 72

This lead is close to sales-ready. They fit your ICP and have the right role, but intent is moderate. Nurture with a targeted email sequence (1–2 weeks). Then if they engage, send to sales.

How This Changes Your Sales Handoff

Before (engagement-based):

• Marketing generates 500 "scored" leads per month
• Sales gets all 500
• Sales says "most of these aren't real"
• Sales focuses on inbound + existing relationships

After (qualification-based):

• Marketing generates 500 leads with qualification scores

• 100 are 70+: send immediately to sales

• 200 are 40–70: nurture with targeted sequences

• 200 are <40: save for future outreach

• Sales gets 100 leads they can actually sell to

• Sales focuses on those + continues outbound

Conversion rate improves because sales is focused on qualified prospects.

Building the Qualification Score

Step 1: Define your ICP. Company size, industry, growth stage, buyer roles. Write it down.

Step 2: For your last 50 closed deals, score them on the qualification framework. Get them to near-perfect (they all bought, so they all scored high).

Step 3: For your last 50 lost deals, score them. You'll see variation. Some scored high (qualified but you lost the deal). Some scored low (never should have been pursued).

Step 4: Calculate: which leads actually converted? What was their average score? If your average converted lead scored 72, set your sales threshold at 70.

Step 5: Build the score into your CRM/marketing automation. Use the formula above. Make it data-driven.

Ongoing Refinement

Track: of all leads you send to sales with a 70+ score, what percentage become customers?

Track: of all leads in nurture (40–70 score), what percentage eventually score 70+ and convert?

After 90 days, you'll see where your thresholds should be. Refine accordingly.

Next Steps

Pull your ICP definition. Score your last 10 closed deals on the qualification framework.

What was their average score?

Then identify your 5 most recent lost leads. Score them. Are they below that average?

If yes, you know your threshold. If no, you need to refine your ICP.

Use that baseline to build your qualification scoring system.