How the 4 lead scoring dimensions interact (and why one number can't replace them)
A working B2B lead scoring framework has four dimensions: engagement (what they’re doing now), profile fit (the person), account fit (the company), and deal context (the opportunity). The previous posts in this series walked through the lead scoring criteria for each dimension in detail. This post covers the part most teams get wrong: how to compose the four scores into something a sales team can act on.
The wrong answer is to average the four into one number. Single-number scoring throws away the asymmetry that makes the framework useful. A lead scoring 22 in engagement and 8 in profile fit lands at the same composite as a lead scoring 8 in engagement and 22 in profile fit, but the two leads are completely different sales plays.
The right answer is a matrix view that preserves the dimensions and pairs them into named plays.
Why averaging fails
Imagine three leads, all scoring 60/100 on a composite:
- Lead A: Engagement 25, Profile Fit 25, Account Fit 5, Deal Context 5
- Lead B: Engagement 5, Profile Fit 25, Account Fit 25, Deal Context 5
- Lead C: Engagement 5, Profile Fit 5, Account Fit 25, Deal Context 25
All three look the same on a single-number score. They are completely different sales situations:
- A is a researcher or competitor. High engagement and right title, but the company isn’t a buyer (wrong size, wrong industry, wrong region) and there’s no open deal context. SDR shouldn’t waste time.
- B is an ABM target. Right person at right company, but no recent engagement and no open opp. Marketing should target the account; sales shouldn’t pitch yet.
- C is a deal-in-flight at the right account but with the wrong contact. Multi-threading or warm intro is the play; not standard SDR outreach.
Averaging the four scores produces 60 in all three cases and produces the same action: a generic “warm lead” assignment. Three leads, three different plays, one wrong action.
The matrix view
A working composite is a 4-cell or 9-cell matrix that pairs the dimensions and tells you what action each cell warrants.
The simplest version is a 2×2 of fit and engagement:
| High Fit | Low Fit | |
|---|---|---|
| High Engagement | Hot lead, AE works this week | Researcher / competitor, deprioritize |
| Low Engagement | ABM / marketing target, nurture | Pass / disqualify |
HubSpot’s modern Combined score natively produces a 3×3 version (A1–C3) on Marketing Hub Enterprise:
| High Fit | Mid Fit | Low Fit | |
|---|---|---|---|
| High Engagement | A1 (Hot) | B1 (Warm) | C1 (Researcher) |
| Mid Engagement | A2 (Warm) | B2 (Mid) | C2 (Cold-but-fit) |
| Low Engagement | A3 (ABM target) | B3 (Mid) | C3 (Pass) |
This gives you nine cells. The HubSpot setup guide groups them into four bands for threshold configuration (Hot, Warm, Nurture, Pass); once routing is involved, each cell deserves its own play:
- A1: AE works it within 24 hours. The strongest signal in the matrix.
- A2: SDR works it this week. Real fit + meaningful activity.
- A3: Marketing targets the account; sales doesn’t pitch yet.
- B1: SDR works it; could become A1 with more engagement.
- B2: Standard nurture; revisit monthly.
- B3: Low priority; revisit quarterly.
- C1: Likely researcher or competitor. Don’t burn SDR cycles.
- C2: Marketing’s problem to solve.
- C3: Pass. Re-evaluate quarterly in case of ICP drift.
Account fit and deal context modify each cell. A1 with an active deal at the account becomes “AE who owns the deal works it”; A1 without one becomes “AE in territory works it.” A3 at a strategic-tier account (high account fit) becomes a top-tier ABM target; A3 at a non-strategic account is just nurture.
Building the matrix in practice
If you’re on HubSpot Enterprise, the Combined score produces the 3×3 matrix natively as a threshold property (A1–C3 labels). Configure thresholds based on your team’s actual capacity:
- A1 should be small enough that AEs can work each one within 24 hours
- A2 + B1 together should be the SDR’s daily pull
- A3 should be a marketing-ops queue, not a sales queue
If you’re on HubSpot Pro, the Combined score isn’t available natively. You’ll either build a manual composite in workflows (assign to a custom Tier property based on Engagement and Fit thresholds), or layer on a tool that produces the matrix view.
For deal context as a modifier, build a separate Contact property like “Deal Context Tier” populated by a workflow that checks for open opps, multi-threading, and account engagement state. Use it as a second sort key after the Combined score.
Where the framework breaks down
A few honest caveats:
The matrix is only as good as the dimensions. If your engagement score over-weights low-quality activity, the engagement axis of the matrix is noisy. If your account fit ignores technographics, the fit axis is missing the highest-predictor signal. Fix the dimensions before composing them.
Routing rules need to follow the matrix. A1 leads going to a round-robin SDR queue when there’s an active AE-owned deal at the account is malpractice. The routing logic has to honor the deal context dimension.
The A1–C3 labels can be misleading. A1 doesn’t mean “definitely will close.” It means “highest expected value per minute of sales time.” Treat it as a sort key, not a forecast.
How to validate the matrix
Build a HubSpot Custom Report tracking MQL→SQL conversion rate by matrix cell over a rolling 90 days. The setup guide’s target bands (Hot 25–40% MQL→SQL, Warm 10–20%, Nurture 2–8%, Pass under 2%) apply here too; per cell, the shape to look for:
- A1: the highest-converting cell in the matrix, at or above the top of the Hot band
- A2 / B1: the rest of the Hot band
- A3: converts below A1 because there’s no current engagement, but carries outsized pipeline value when it does close
- C1: near the bottom; confirms most are researchers
- C3: near zero; confirms the floor is working
If your A1 cell isn’t converting meaningfully better than B2 or C2, the framework isn’t producing useful sort. Diagnose which dimension is broken: is engagement scoring noisy, fit rules outdated, or thresholds wrong?
This is the calibration loop that makes the matrix useful over time. Run it monthly. The day A1 conversion drops below B2 conversion, you have a structural problem to fix this week.
Common mistakes
A few traps:
Hiding the matrix from sales. Reps don’t need to see a number; they need to see “this is an A1 lead in your territory, here’s why it scored that way.” If your alerts and CRM views don’t expose the matrix tier, your reps will sort by gut.
Ignoring deal context as a modifier. The 3×3 matrix tells you fit + engagement. Without deal context layered on top, AEs can’t tell whether to work an A1 as new business or save-the-deal motion.
Static thresholds. Your A1/A2/A3 boundaries shouldn’t be hardcoded forever. As your business grows and your pipeline volume changes, the thresholds need to track sales capacity. A1 should always be the volume your AEs can actually work in 24 hours.
Where this fits
This is the wrap-up to a series on the four dimensions of B2B lead scoring. The dimensions:
- Engagement: what they’re doing now
- Profile fit: the person
- Account fit: the company
- Deal context: open opportunities at the account
The framework is a starting point. Your weights, thresholds, and routing rules are yours to tune against your data. The point is the structure: four dimensions, composed into a matrix, calibrated against actual outcomes.
A note on tooling
kenbun scores all four dimensions natively and keeps them separate instead of blending them into one number. Every score arrives with the per-dimension breakdown visible on the lead, with deal-context signals like deal-level multi-threading depth layered on top. See it on your HubSpot data.