Scoring Rules

Scoring rules are the fundamental building blocks of kenbun. Each rule connects an event type or attribute to a point value, creating a direct relationship between a lead's behavior or profile and their score. Rules add up to scores; scores roll up into Levels; levels feed segments and triggers.

How Scoring Rules Work

When something happens — a lead opens an email, a CRM sync updates a company's industry, a deal moves stage — kenbun walks the active rules to see which ones match. Every match adds (or subtracts) points. The total is the lead's, account's, or deal's score for that ruleset.

There are four scoring surfaces, each configured under Configure > Scoring:

Scoring Type	What It Scores	Source Of Points	Page
Engagement	Per-lead behavior	Events (page views, opens, form fills)	Engagement Scoring
Profile	Per-lead fit	Lead attributes (title, role, industry)	Profile Scoring
Account	Per-account fit (ABM)	Account attributes (size, revenue, industry)	Account Scoring
Deal	Per-deal	Deal-level signals (feature-flagged)	Deal Scoring

Each surface has its own rules, its own ruleset structure, and its own level system. They run independently — engagement scoring never affects profile scoring, and vice versa.

Engagement vs. Fit Scoring

The biggest conceptual divide is between engagement (behavior over time) and fit (who the person or company is):

• Engagement scoring is dynamic. Scores rise on activity and decay during inactivity. Use it to find who's interested right now.
• Profile and account scoring is slow-moving. Scores change only when attributes change. Use them to filter for ICP fit regardless of activity.
• Deal scoring combines signals at the deal level — useful for forecasting and pipeline triage.

The most actionable signal in kenbun is the cross-section: a lead who's both highly engaged and a strong profile fit, on an account that's a strong ABM target. That's why you typically configure two or three of these scoring surfaces in parallel, not just one.

Common Concepts Across All Scoring Types

Every scoring surface shares the same building blocks:

Rulesets

Rules live inside rulesets. A ruleset is a collection that activates as a unit, can run side-by-side with others for A/B testing, and produces its own score per lead/account/deal. Each surface has a "primary" ruleset that supplies the score shown by default in the UI.

Rules

A rule has:

• Trigger — the event type (engagement) or attribute (profile/account/deal)
• Weight — points awarded when the rule matches; positive or negative whole numbers
• Active status — toggle on/off without deleting
• Optional date range — for seasonal or campaign-only rules
• Optional attribute filters — narrow when the rule fires (e.g., "Page View where path contains /pricing")

Templates

Each scoring surface offers pre-built templates for common scoring strategies (SaaS, PLG, Content Marketing, Enterprise ABM). When the rules list is empty, kenbun shows template cards in the empty state — click Load Template to seed a starting configuration you can then tune.

Test Before Activation

Before enabling a new rule in production, click Test Rule in the rule editor. Pick a recent lead from the dropdown to see whether the rule would match, why it does or doesn't, and how the lead's projected score would change. This catches over-broad and under-specific rules before they affect every lead in your workspace.

Preview Rescore Impact

When you change rules, you usually want to rescore historical leads/accounts so the changes apply retroactively. Before triggering that, click Preview Impact on the Rescore All Leads button to see how many leads will be affected and the average score change. This is a dry run — no scores change, no triggers fire — but it lets you catch unintended consequences before applying.

Best Practices

Start Simple

3–5 rules per surface is enough to get value. You can always add more once you see how the scoring distribution lands.

Document The "Why"

Use rule names that describe intent, not just the trigger. "Email Open" tells you nothing; "Email Open — Top of Funnel Nurture +5" tells the next person what the rule is doing in your model.

Calibrate Against Closed-Won

The most defensible scoring model is grounded in conversion data. If your "Hot" tier doesn't include most of your closed-won leads in the 30 days before they signed, the model isn't doing what you think.

Audit Quarterly

Disable rules that haven't matched anything in 90 days. Re-weight rules that don't reflect your current sales motion. Templates make sense when you start; your scoring model should be uniquely yours within a quarter.

Troubleshooting

Symptom	Likely Cause / Fix
Rule not triggering	Verify event type or attribute name (case-sensitive) and that the rule is active
Incorrect point values	Check for conflicting rules; if using attribute filters, verify priority order
Weight not saving	Weights must be whole numbers — 10.5 and non-numeric text are rejected
Regex filter errors	Check that parentheses and brackets are balanced in the pattern
Unexpected total scores	Use Score Explain on the lead detail page to see which rules matched

Related

• Rulesets — group rules into collections
• Levels — turn scores into named tiers
• Decay Rules — automatic score reduction on inactivity
• Half-Life Decay — per-signal exponential decay with a configurable floor
• Filters — keep artificial activity out of scoring