B2B Lead Scoring in 2026: The 4-Dimensional Model That Doubles SQL Conversion

Dimension 1: Firmographic fit (25% of score)

Weight: 25%. This dimension is the table stakes. It does not predict intent.

Dimension 2: Behavioral signals (35% of score)

Weight: 35%. This is the strongest intent signal. The more product-related the behaviour, the higher the score.

Dimension 3: Trigger events (25% of score)

Weight: 25%. This is the timing dimension. A firmographic-fit company with a recent trigger event is 3-5x more likely to convert.

Dimension 4: Engagement recency (15% of score)

Weight: 15%. Recency is the multiplier. A lead that engaged 6 months ago is cold. A lead that engaged yesterday is hot.

• Company size matches ICP (target band: 50-200 employees)
• Industry matches ICP (specific verticals, not generic)
• Geography matches ICP (US, UK, EU)
• Revenue matches ICP ($5M-$50M for mid-market)
• Funding stage matches ICP
• Pricing page views
• Product page views
• Demo request page views
• Case study downloads
• Webinar attendance
• Email engagement (opens, clicks, replies)
• Content downloads (gated assets)
• Recent funding round
• Leadership change in target role
• Hiring surge in target team
• Tech adoption (new tool, new integration)
• Expansion event
• Negative signal (layoffs, missed earnings)
• Touched in the last 7 days: +15
• Touched in the last 30 days: +10
• Touched in the last 90 days: +5
• No touch in 90+ days: 0

``` Total Score = (Firmographic * 0.25) + (Behavioral * 0.35) + (Trigger * 0.25) + (Recency * 0.15) ```

The score is 0-100. Bands:

• 80-100: Hot, route to AE within 1 hour
• 60-79: Warm, add to active nurture sequence
• 40-59: Cool, add to long-term nurture
• 0-39: Cold, drop from active outreach

Most teams score on firmographic alone. A VP of Sales at a 100-person SaaS company gets the same score whether they visited the pricing page yesterday or never heard of the company.

The behavioural dimension is the differentiator. Without it, you are scoring companies, not leads.

Lead scoring is the highest-leverage place to deploy an AI agent in 2026:

The result: hot leads are touched in under 5 minutes (vs 24-48 hours). Cold leads do not waste rep time. Conversion to SQL doubles.

• Real-time data ingestion (website, CRM, email, intent data)
• Multi-dimensional scoring in seconds
• Routing decisions based on score and rep availability
• Continuous learning from closed-won / closed-lost data

Step 1: Define the dimensions and weights

Work with sales and marketing leadership to agree on:

Step 2: Connect the data sources

The scoring engine needs:

Step 3: Set up the routing

Hot leads (80+) route to an AE within 1 hour.

Warm leads (60-79) enter a nurture sequence.

Cool leads (40-59) enter a long-term drip.

Cold leads (0-39) are dropped from active outreach.

Step 4: Review weekly

Track:

Step 5: Update the model quarterly

As closed-won data accumulates, the weights should adjust. The 25/35/25/15 split is a starting point. After 90 days of data, recalibrate.

• The 4 dimensions
• The weight for each (this is a judgment call based on what has worked historically)
• The score bands (hot, warm, cool, cold)
• The action per band
• CRM (HubSpot, Salesforce) for firmographic
• Website analytics (Hotjar, Heap) for behavioural
• Intent data (Bombora, 6sense) for trigger
• Email platform (HubSpot, Outreach) for engagement
• Score-to-conversion correlation
• False positive rate (high score, no conversion)
• False negative rate (low score, conversion anyway)
• Weight adjustments needed

If speed to lead is over 1 hour for hot leads, the routing automation is broken. If SQL conversion is under 20% for hot leads, the scoring weights are wrong.

• SQL conversion rate: percentage of scored leads that become SQLs (target: 30%+ for hot leads)
• Speed to lead: time from high-score trigger to AE contact (target: under 1 hour for hot)
• Pipeline contribution: pipeline generated from scored leads (target: 60%+ of total pipeline)