How to Build an Account Scoring Model in Your CRM

Account scoring separates the high-potential accounts from the noise, letting your team focus where deals are most likely to close. But most scoring models fail - not because the concept is wrong, but because the implementation is sloppy. Scores get stale, thresholds are arbitrary, and nobody trusts the output. This guide walks through how to build a model that actually drives routing and prioritization decisions.

The Two Dimensions of Account Scoring¶

Effective account scores combine two independent signals:

1. Firmographic fit - How well does this company match your Ideal Customer Profile?
2. Behavioral engagement - How actively are people at this account interacting with your brand?

Neither dimension alone is sufficient. A perfect-fit company with zero engagement isn’t ready to buy. A highly engaged company outside your ICP will churn fast. You need both.

Firmographic Scoring Criteria¶

Important: Base these ranges on your closed-won customer data, not assumptions. Pull the firmographic profile of your top 50 customers by ARR and use that to define your tiers.

Behavioral Scoring Criteria¶

Assign points based on engagement signals aggregated at the account level:

• Website visits - 1 point per unique contact visiting, up to 10 points/month
• Content downloads - 3 points per gated asset
• Email engagement - 1 point per open, 2 points per click (across all contacts)
• Event attendance - 5 points per webinar or conference session
• Product page visits - 5 points (pricing page = 10 points)
• Demo request - 20 points
• Multiple contacts engaged - 5-point bonus when 3+ contacts from the same account are active in a 30-day window

Apply time decay: Reduce behavioral scores by 50% every 30 days. Engagement from six months ago shouldn’t drive today’s prioritization.

Implementation: Salesforce¶

In Salesforce, implement account scoring with these components:

1. Custom fields on Account:
2. Firmographic_Score__c (Number)
3. Behavioral_Score__c (Number)
4. Combined_Account_Score__c (Formula: Firmographic + Behavioral)

5. Account_Tier__c (Formula or Flow-assigned: A / B / C / D)

6. Firmographic scoring: Use a Flow or Apex trigger that recalculates Firmographic_Score__c whenever enrichment data updates (employee count, industry, revenue fields).

7. Behavioral scoring: Aggregate engagement data from your MAP (Marketo, Pardot, HubSpot) via integration. Roll up contact-level activity to the Account using a scheduled Flow that runs nightly.

8. Tier assignment thresholds:

Implementation: HubSpot¶

In HubSpot, use the native Company Scoring property or build a custom score:

1. Create a custom Company property for firmographic and behavioral scores
2. Use Workflows to update firmographic scores based on Company properties
3. Roll up contact engagement using the “Associated Contact” activity triggers in workflows
4. Set lifecycle actions based on score thresholds using workflow branching

Routing Rules Based on Scores¶

Connect your scoring model to your lead and account routing:

• Tier A accounts bypass SDR and route directly to the territory AE
• Tier B accounts get fast-tracked in the SDR queue with priority flags
• Tier C accounts enter automated nurture sequences with human checkpoints every 30 days
• Score increases trigger real-time alerts - if an account jumps from C to A, the owning rep gets a Slack notification

Key Takeaways¶

• Score accounts on two independent dimensions: firmographic fit and behavioral engagement
• Derive scoring criteria from closed-won customer data, not assumptions
• Apply time decay to behavioral scores so stale engagement doesn’t inflate priority
• Set clear tier thresholds and connect them to routing rules and SLA commitments
• Recalibrate the model quarterly by comparing predicted tiers against actual outcomes