Account Scoring for ABM: Step-by-Step Setup

Introduction

Account-based marketing requires precision. Your sales team has limited capacity and must focus on accounts that generate real revenue, not tire-kickers. That's where account scoring becomes critical.

An account scoring model quantifies which prospects fit your solution (match to ICP), which are actively buying (intent signals), and which combinations deserve your AE's focus. It transforms subjective account selection into a repeatable, data-driven framework that your entire GTM team can trust and execute against.

This guide walks you through building a practical account scoring system: from basic firmographics to behavioral intent signals, validated against your actual closed deal data.

The Core Logic: Fit Meets Intent

Account scoring works when it answers three questions simultaneously:

Fit: Does this account match the profile of companies that typically buy from you and succeed with your product?

Intent: Is someone at this account actively evaluating a solution in your category right now?

Timing: Are they in a buying cycle where you can realistically influence the decision?

The strongest accounts score high on all three. Most scoring systems collapse fit and intent into a single number, which creates problems: high-intent accounts that don't fit your model will waste your sales team's time, and high-fit accounts that show no current intent will frustrate everyone when nothing closes.

The 2026 approach is to keep these signals separate long enough to make routing decisions, then combine them strategically.

Building Your Fit Tier: The Permanent Foundation

Fit score should change rarely. It captures the structural match between an account and your ideal customer profile.

Start by identifying the firmographic attributes that appear consistently in your closed-won deals. Pull your last 30-40 closed deals and extract:

• Company size range (by employee count or revenue)
• Industry classification
• Geographic regions where you operate
• Budget pattern (public, private, bootstrapped)
• Company stage (early-stage, growth, mature)
• Use case alignment (are they in a use case you support?)

Weight each attribute based on frequency in your closed-won data, not gut instinct. If 85% of your closed deals are enterprise companies and only 15% are mid-market, give enterprise a significantly higher weight.

Document this explicitly. Your sales team needs to understand why certain accounts score the way they do, or they'll ignore the scoring system entirely.

A practical fit model might look like: base of 50 points for any account in your target verticals, plus modifiers for company size, geographic restrictions, and use case match. Keep it simple.

Scoring Intent: The Active Signal

Intent scoring captures buying activity today and decays naturally over time. Unlike fit, intent changes week by week.

The strongest intent signals include:

• Website behavior: Accounts whose contacts visit your pricing page, comparison pages, or security/compliance pages. These suggest active evaluation. High time-on-page is stronger than raw pageviews.
• Search and digital signals: Accounts running searches for your category or competitor names. This suggests they're in market.
• Email engagement: Opens, clicks, and replies to your campaigns. Replies are the strongest signal.
• Direct sales engagement: Responses to outreach, meeting requests, and proposal requests. High-intent but late-stage in the buying cycle.
• Content engagement: Demo registrations, trial signups, and webinar attendance. These are explicit commitment signals.

Set intent scores to decay over time. An account that engaged two weeks ago is more likely to close than one that engaged three months ago. Use rolling time windows: activity in the last 7 days gets full points, activity 8-30 days back gets 50% of points, activity older than 30 days decays further or resets.

Combining Signals: The Routing Decision

The simplest approach is multiplicative: multiply fit by intent. This ensures high-intent accounts that don't fit still score low, and high-fit accounts with no intent don't monopolize your sales team.

For a practical implementation:

• High fit (70+) + High intent (70+): Immediate sales focus. These accounts are worth aggressive prospecting and rapid deal progression.
• High fit + Medium intent (40-70): Nurture and engagement. Your marketing can stay active while sales watches for intent to increase.
• High fit + Low intent: Pipeline building. Stay top-of-mind through content and light outreach, but don't expect deals this quarter.
• Low fit + Any intent: Skip or re-evaluate. High-intent unfit accounts can sometimes reveal segments you're missing, but they shouldn't consume sales time.

Technical Implementation: Where Account Scoring Lives

For teams under 500 target accounts, CRM-native scoring (HubSpot formulas, Salesforce flows) is sufficient. Define custom fields for each firmographic attribute and create formulas that add points based on those fields. Layer in email engagement tracking and website behavioral data if your CRM integrates with your MarTech stack. For intent data integration, see how to build your ABM tech stack.

For teams managing thousands of accounts, consider a separate scoring layer. This can be a data pipeline (using Python, SQL, or your data warehouse) that pulls firmographic and behavioral data from multiple sources, calculates a composite score, and syncs results back to your CRM daily or weekly.

Whoever owns the scoring system (marketing ops, revenue ops, or data) needs to have clear ownership and quarterly review cycles. Scoring models degrade if they're never updated.

Validation: Testing Your Model

Before rolling out account scoring to your sales team, validate it against actual outcomes.

Take your last 15 closed-won deals. Score those accounts retroactively using your new model. What was the average score? Set that as your baseline for "worth-pursuing" accounts.

Now take 10 opportunities you lost in the last 90 days. Score those as well. If lost deals scored significantly higher than won deals, either your fit model is capturing the wrong attributes or your weighting is inverted. Adjust and test again.

Ask your strongest salespeople: does this ranking match their gut feel about which accounts are actually closeable? Sales intuition often captures nuances your model misses, so disagreement signals missing variables or poor weighting.

Iterate quarterly. As you close more deals, the model should improve.

Common Implementation Pitfalls

Overweighting intent at the expense of fit. You can't close accounts that don't fit your solution. If a low-fit account shows high engagement, investigate why. You might be missing an addressable segment, or your messaging might be attracting the wrong buyers.

Never decaying engagement scores. Stale engagement is worthless. If you're not automating score decay over time, your old activity pollutes your current prioritization. Review and reset quarterly at minimum.

Siloing scoring from sales strategy. If your scoring model doesn't align with how your sales team actually works, it will be ignored. Build it with them, not for them. Sales input should shape weighting decisions.

Conflating account scoring with contact scoring. These are separate problems. You might score an account highly, but within that account, only the CFO and VP of Engineering matter for your use case. Don't confuse account-level fit with contact-level relevance. Layer contact fit on top of account scoring for routing decisions.

Treating all customer segments identically. If you serve both startups and enterprises, build separate scoring models. Their fit signals, intent patterns, and buying cycles are fundamentally different.

Getting Started

Day one: Document your core firmographic attributes by pulling your last 20 closed-won deals. Identify what they had in common. That's your fit model foundation.

Day two: Layer in one engagement signal from your CRM (email opens, or website visits if you have that integrated). Use a simple decay formula.

Days three and four: Test against recent closed deals and get sales feedback.

Once your sales team trusts the model, expand to additional signals and refine weights based on quarterly reviews.

The goal isn't a perfect score. It's a transparent, repeatable process that gets better every quarter and aligns your sales and marketing teams on account priority.

Ready to implement account scoring in your ABM program? Book a demo with Abmatic AI to see how to build a scoring framework that prioritizes high-value accounts and aligns your sales and marketing teams.

Related resources: - What Is an Ideal Customer Profile (ICP)? - ABM vs Demand Generation: Which Strategy Is Right for You?