Account Scoring for ABM
Account Scoring is the engine behind your ABM tiering — the rules that produce the scores Account Levels sit on top of. This page covers the ABM strategy behind account scoring. For the configuration walkthrough — how to add rules, create rulesets, and run Score Explain — see the canonical Account Scoring page.
Why Account Scoring Matters For ABM
Lead-level engagement scoring tells you who's interested. ABM-grade account scoring answers a different question: is this company even worth pursuing?
A perfect score here means:
- Targeting budget is spent on companies with real LTV potential.
- Sales territory and AE assignment is grounded in fit, not pipeline guesswork.
- Marketing campaigns segment cleanly against accounts likely to convert.
- Customer-success teams know which logos to invest in for expansion.
Anchoring Your Model
The single most important step in building an account scoring model is grounding it in conversion data, not intuition. Before you write a single rule:
- Pull a list of your closed-won accounts from the last 12–24 months.
- Pull a list of closed-lost accounts (or accounts that disqualified out).
- Compare them on:
- Company size (employees and revenue)
- Industry
- Geography
- Technographic signals (if available)
- Customer stage (new prospect, expansion candidate, etc.)
The attributes that most differentiate won from lost are your highest-weight rules.
Common ABM Scoring Patterns
Enterprise B2B SaaS
employees gte 1000 → +50
revenue gte 50000000 → +45
country in [US, UK, CA, AU, DE] → +20
industry in [Technology, Finance, Healthcare] → +30
A perfect-fit account hits 145 points; a marginal account hits 50.
Mid-Market Land Strategy
employees in 100-1000 → +50
revenue in 10M-100M → +40
industry equals target_vertical → +30
not_already_a_customer → +20
PLG Expansion
existing_customer = true → +40
plan in [Pro, Business] → +30
employees gte 100 → +25
expansion_signal_in_last_30_days = true → +25
Translating Scores Into Plays
The score itself isn't actionable — what you do at each tier is. As you design your model, decide upfront:
| Tier | Targeting | Sales Motion | Success Metric |
|---|---|---|---|
| Top tier | High-touch ABM, named ad audiences, direct mail | AE-led, multi-stakeholder | Pipeline created |
| Mid tier | Segmented campaigns, intent-based outreach | SDR-led, single-threaded | MQL conversion |
| Lower tier | Marketing nurture only, retargeting | Inbound only | Demo requests |
If two tiers have the same plays, your model has too many tiers. If a tier has no clear plays, it doesn't earn its place.
Avoid Common Pitfalls
- Too many rules. Five to ten well-chosen rules outperform 30 mediocre ones. Pick the attributes that actually predict conversion.
- Static models. ICP shifts as your product matures. Plan to revisit weights every quarter.
- Negative weights. Tempting, but they make Score Explain confusing. Prefer not awarding points and using Account Levels to separate the bottom tier.
- Missing mappings. Account scoring rules silently match nothing if the property isn't populated by Account Mapping. Verify mapping coverage before debugging weights.
Where To Go Next
- Account Scoring — configuration walkthrough — how to add rules, create rulesets, run Score Explain
- Account Levels — turn account scores into ABM tiers
- Account Mapping — populate the fields you'll score against
- ABM Overview — the broader ABM strategy framework in kenbun
- Account Triggers — automations that fire on account changes