Not all accounts are equal.
You could have 500 good-fit companies on your target account list. But which ones should your team focus on this quarter?
That's where account scoring comes in. It's the discipline of quantifying account potential and prioritizing sales and marketing effort where it'll have the biggest payoff.
This guide breaks down how to build an account scoring model that actually works.
The Three Dimensions of Account Scoring
A solid account score combines three factors:
1. Fit Score: How well does this account match your ideal customer profile? - Company size (employee count) - Industry vertical - Revenue range - Technology stack - Geographic location
2. Intent Score: How actively is this account showing buying signals? - Recent job hires in buying roles (VP Sales, VP Marketing) - Engagement with your content - Competitor tool searches - Mentions in news, earnings calls, funding announcements - Engagement with account-based ads
3. Engagement Score: How much is this account interacting with you? - Email opens and clicks from company domain - Website visits and asset downloads - Social interactions with your brand - Sales call acceptance rate - Meeting attendance rate
A strong account has high scores across all three. An account that fits your ICP but shows no intent or engagement? Not ready. An account showing intent signals but weak fit? Probably not a right customer.
Step 1: Build Your Fit Scoring Model
Start with your Ideal Customer Profile (ICP). Convert it to numerical attributes.
Example for a B2B SaaS sales platform:
| Attribute | Weight | Scoring |
|---|---|---|
| Company size (employees) | 30% | 50-500 employees = 100, 501-2000 = 80, 2001+ = 60, <50 = 0 |
| Revenue range | 20% | $10M-100M ARR = 100, $100M-500M = 80, > $500M = 40 |
| Industry | 30% | Tech/SaaS = 100, Financial Services = 80, Other B2B = 40 |
| Geographic footprint | 10% | North America = 100, Europe = 80, APAC = 50 |
| Growth rate | 10% | Positive growth in last 12 months = 100, flat/declining = 0 |
For each account, gather firmographic data (from ZoomInfo, Apollo, LinkedIn, or your own research), plug it in, and calculate a composite fit score.
Example: - Company size: 150 employees = 100 points - Revenue: $45M ARR = 100 points - Industry: Tech = 100 points - Geography: USA = 100 points - Growth: +20% YoY = 100 points - Fit Score: 92/100
Accounts above 80 are strong fit. Below 60, they're probably not worth your time.
Step 2: Add Intent Scoring
Fit alone is lazy. Layer in intent.
Intent signals come from three buckets:
Buyer research behavior: - Visiting your website (especially pricing, demo request pages) - Downloading assets (whitepapers, case studies) - Attending webinars or events - Reading your blog or content
Company signals: - Recent job postings for buyer roles (VP Sales, VP Marketing, Sales Ops) - Funding announcements or equity raises - New product launches - Earnings calls mentioning go-to-market challenges - Org restructuring announcements
Competitive signals: - Searches for competitor tools - Contract renewals approaching (if you have that data) - Switching indicators (if someone who worked there now uses your platform)
Create a simple scoring matrix:
| Signal | Points | Decay |
|---|---|---|
| Website visit this week | 5 | -50% every 2 weeks |
| Asset download | 15 | -50% every 4 weeks |
| Job posting for buyer role | 20 | -10% per month |
| Webinar attendance | 10 | -50% every 2 weeks |
| News mention (go-to-market) | 25 | -20% per month |
Intent scores decay over time. A job posting for a VP of Sales 6 months ago is less relevant than one posted yesterday.
Calculate rolling intent scores weekly. An account with: - Website visit this week: 5 points - Job posting from last month: 18 points - Webinar attendance 3 weeks ago: 3 points - Intent Score: 26/100 (moderate, increasing)
Step 3: Add Engagement Scoring
Finally, measure how much the account is engaging with your brand.
Track:
Email engagement (from your email platform): - First-time opener from this company: 5 points - Multiple opens: 10 points - Click: 15 points
Web engagement (from analytics or intent tools): - Page views this month: 1 point per 5 views (capped at 20) - Asset download: 10 points each - Demo request: 50 points - Trial signup: 50 points
Sales engagement (from your CRM): - Inbound inquiry: 30 points - Meeting booked: 20 points - Meeting attended: 30 points
Reset engagement scores monthly. It reflects recent interaction, not historical.
Example: - 12 email opens and 3 clicks this month: 25 points - 22 web page views: 4 points - 1 asset download: 10 points - 1 demo request: 50 points - Engagement Score: 89/100 (highly engaged this month)
Skip the manual work
Abmatic AI runs targets, sequences, ads, meetings, and attribution autonomously. One platform replaces 9 tools.
See the demo →Step 4: Combine into a Composite Account Score
Weight the three components:
- Fit: 50% (foundational; wrong ICP = wrong fit, no matter how engaged)
- Intent: 30% (shows they're in market)
- Engagement: 20% (shows active interest now)
Composite Score = (Fit × 0.5) + (Intent × 0.3) + (Engagement × 0.2)
Example: - Fit: 92 × 0.5 = 46 - Intent: 26 × 0.3 = 7.8 - Engagement: 89 × 0.2 = 17.8 - Composite Score: 71.6/100
Tier these: - 80-100: Tier 1 (Hotlist) - your most ready accounts - 60-79: Tier 2 (Growth) - good fit, building intent - 40-59: Tier 3 (Exploratory) - okay fit, early signals - <40: Out of scope for now
Step 5: Update Continuously and Act
Account scores are not static. Recalculate weekly or biweekly.
Create a dashboard your sales and marketing teams watch:
- Accounts entering Tier 1 this week (new signals, high engagement)
- Accounts dropping from Tier 1 (engagement declining, signals cooling)
- Fastest-growing intent signals (job postings, funding news)
When an account jumps from Tier 2 to Tier 1, trigger immediate actions:
- Sales gets notified (prioritize outreach)
- Marketing scales resources (personalized content, account ads)
- Operations updates the CRM and flags for account-based plays
Avoiding Common Mistakes
Mistake 1: Too many signals, too much complexity Start simple: 5-7 signals. Add more as you mature. Easier to maintain and explain.
Mistake 2: Static weights Your Fit/Intent/Engagement weighting might shift over time. For mature logos, maybe Intent scores less. For early-stage accounts, maybe Fit scores more. Test and iterate.
Mistake 3: Scoring without action A beautiful scoring model is useless if nobody acts on it. Connect scores to CRM workflows, Slack notifications, and sales plays.
Mistake 4: Ignoring negative signals An account showing strong intent but active use of your competitor might not be a real opportunity. Add negative signal decay. A strong competitor adoption score should drop account scores.
Tools to Automate This
Manual account scoring is labor-intensive. Automate:
- Intent data: 6sense, Demandbase, Clearbit (auto-score based on fit, signals)
- Email engagement: HubSpot, Marketo (track opens, clicks by company domain)
- Web engagement: Segment, Google Analytics with company-level roll-up
- Job posting monitoring: Hunter.io, ZoomInfo, LinkedIn (auto-capture job postings)
- Orchestration: HubSpot/Salesforce workflows (update scores weekly, trigger plays)
The Output: Actionable Prioritization
A good account scoring model does one job: tell your sales and marketing teams where to focus.
"Spend this quarter on Tier 1 accounts. Here are the 25 accounts that fit, have strong intent, and are engaging now. Sales, here's who to call. Marketing, here's who to personalize content to."
That focus is worth its weight in pipeline.
Ready to run ABM the right way? Book a demo with Abmatic.ai
