Third-Party Intent vs Account Scoring: Which Signals Matter?

Third-party intent data (Bombora, G2, web monitoring) identifies which accounts are actively researching solutions right now; account scoring (6sense, Demandbase, AI models) ranks which accounts are most likely to buy based on firmographic/technographic fit; they answer different questions and work best together: intent identifies buying activity (timing), scoring identifies best-fit accounts (who to pursue). Combined, they deliver 70%+ accuracy in predicting which accounts will close deals within 6-12 months.

Key Takeaways

• Intent data answers the timing question: Which accounts are actively researching and likely to buy soon (30-90 days); sources include Bombora (web monitoring), G2 (review/comparison activity), ZoomInfo (company changes), Clearbit (hiring/growth signals), and LinkedIn (job postings, funding); intent is fresh (real-time to weekly) but biased toward web-active companies
• Account scoring answers the priority question: Which accounts match your ICP and are most likely to become customers; based on firmographic (company size, revenue, industry), technographic (technology stack, maturity), and behavioral signals; scoring models are built by 6sense, Demandbase, or custom AI and remain stable (change weekly/monthly)
• Intent data strengths: High accuracy (directly observes activity), real-time freshness, activates urgent follow-up (accounts showing intent now), efficient budget (focus on warm accounts not cold lists); weaknesses: misses companies using non-web research, biases toward larger tech-forward companies, expensive ($10K-$30K/year per source)
• Account scoring strengths: Holistic fit evaluation (combines multiple signals), stable prioritization (not just recent activity spikes), scalable (works across your entire TAM), identifies best accounts even if not currently active; weaknesses: doesn't capture current buying stage (many high-fit accounts aren't ready to buy yet), requires constant retraining, misses accounts below threshold
• Combined approach (best practice): Layer intent on top of scoring: use scoring to identify 100-300 best-fit accounts (your TAL), use intent to identify which scored accounts are actively researching (activate immediate outreach), creates 70%+ accuracy in deal prediction and focuses resources on warm accounts with high fit
• Budget allocation: Intent data ($10K-$30K/year) + scoring model ($20K-$50K/year) + combined platform ($30K-$100K/year) for comprehensive signal stack; most mid-market teams start with scoring alone, add intent once proving ROI

Understanding Third-Party Intent Data

Third-party intent data is observed buying behavior from external sources:

Bombora: Monitors B2B websites. When IP addresses from company devices visit articles tagged "CRM alternatives" or "sales engagement software," Bombora logs that activity by company.

G2: Tracks who reads vendor review and comparison pages. Reading "Salesforce vs HubSpot" or "best CRM software" indicates research activity.

ZoomInfo: Monitors company changes (funding, acquisitions, executive moves) that trigger buying need.

Clearbit: Tracks company growth signals and hiring patterns.

Intent data characteristics: - What it measures: Active research behavior (company is looking at solutions) - Accuracy: High (directly observing activity) - Freshness: Real-time to weekly - Bias: Skews toward web-active companies; misses companies using non-web research - Use case: Identifying accounts actively shopping for your solution

Strengths of Intent Data

• Identifies accounts in active buying stage
• Proactive signal (find them before they visit you)
• Measurable action (they visited X page, downloaded Y content)
• Timing signal (if activity is recent, they're buying now)

Limitations of Intent Data

• Doesn't predict likelihood (they're researching but may not buy)
• False positives (someone visited a page out of curiosity)
• Biased toward tech-savvy companies
• Doesn't work for sectors with limited web research (manufacturing, construction)
• Missing stakeholders (only measures person with access to intent data)

Understanding Account Scoring

Account scoring is a model that predicts which accounts are most likely to buy based on multiple factors:

Firmographic signals: - Company size (employees, revenue) - Industry and vertical - Growth rate - Funding status

Behavioral signals: - Website visits - Email engagement - Content consumption - Product trial activity - Sales team interaction

Intent signals (from multiple sources): - Third-party intent data (Bombora, G2) - Job posting analysis - News and company changes - Executive movement

Model output: Account score (0-100) predicting likelihood to buy

Scoring Model Types

Rule-based scoring: "If company size 100-1000 AND revenue $10M-$100M AND in tech industry AND visited website in last 30 days, score 80."

Probabilistic scoring (AI): ML model learns from your historical data which accounts bought. Then predicts score for new accounts based on patterns.

Strengths of Account Scoring

• Predicts likelihood to buy (not just activity)
• Combines multiple signals (intent + firmographic + behavioral)
• Personalized to your data (learns from your customers)
• Ranking system (tells you which accounts to prioritize)
• Explainable (you can see why an account got high score)

Limitations of Account Scoring

• Requires historical data to train (new products have no training data)
• Assumes future = past (market changes break models)
• Garbage in, garbage out (bad input data = bad scores)
• False negatives (may miss buyers who look different from your current customers)

Intent vs Scoring: What's the Difference?

Real-World Example: Identifying and Prioritizing Accounts

Scenario: You sell sales engagement software to mid-market companies.

Step 1: Identify (Intent Data)

"Who's researching sales engagement solutions?"

Using Bombora, you find 500 accounts showing research intent: - Visited articles about "sales engagement alternatives" - Downloaded whitepapers on "sales enablement ROI" - Visited your competitor's pricing page - Searched "sales engagement software comparison"

Output: 500 accounts actively looking

Step 2: Qualify (Firmographic + ICP)

"Do they match our ICP?"

Filter 500 accounts down to 300 that match: - 100-2,000 employees - $10M-$500M revenue - Technology, professional services, or other relevant verticals - US-based (your target region)

Output: 300 qualified accounts actively shopping

Step 3: Score (Account Scoring)

"Which of the 300 are most likely to buy?"

Apply 6sense or Demandbase scoring model:

Score 90-100 (High priority - 30 accounts): - Match ICP perfectly - Visiting your site regularly (first-party intent) - Showing intent for your use case - Recently hired related role (e.g., VP Sales)

Score 70-89 (Medium priority - 100 accounts): - Match ICP well - Showing intent but not visiting your site yet - Not yet showing hiring signals

Score 50-69 (Lower priority - 170 accounts): - Match ICP loosely - Showing intent but many other indicators suggest not ready

Output: Prioritized list (30 hot, 100 warm, 170 cooler)

Step 4: Engage (Sales Execution)

Apply different strategies by score:

Top 30 (Score 90+): - Personalized outreach to buying committee - Account-based campaigns across email, ads, calls - Executive engagement - Expected win rate: 25-35%

Mid 100 (Score 70-89): - Standard sales engagement sequences - Email + calling - Group campaigns - Expected win rate: 10-15%

Lower 170 (Score 50-69): - Light touch nurture - Email campaigns, ads - Expected win rate: 3-5%

When to Use Intent Data vs Scoring

Use Intent Data When:

• You want to find accounts actively in market
• Your sales cycle is short (timing matters)
• You have many potential buyers (need to prioritize by buying signal)
• You're trying to find new accounts to target (cold outreach)
• You want to know "are they buying NOW?"

Use Account Scoring When:

• You want to prioritize within your target account list
• Your sales cycle is long (need to identify early-stage buyers)
• You want to know "are they likely to buy eventually?"
• You have limited sales capacity (need to focus on highest-probability)
• You want explainability (why is this account a priority?)

Use Both When:

• Running account-based marketing at scale
• You need both identification (intent) and prioritization (scoring)
• You want to balance timing (intent) with fit (scoring)
• Managing complex, multi-stakeholder sales cycles

Comparing Intent Data Providers

Comparing Account Scoring Approaches

The Bottom Line

Intent data tells you WHO is buying. Account scoring tells you WHICH accounts to prioritize.

Best practices:

1. Identify with intent data (who's actively researching?)
2. Qualify with firmographic data (do they fit your ICP?)
3. Score with account scoring (who's most likely to close?)
4. Engage based on score (adjust cadence and outreach by priority)

Most successful ABM programs use both. Intent data identifies active buyers (50-60% of your addressable market). Account scoring determines which of those active buyers to pursue with full campaigns (top 10-20%).

The combination: intent + scoring achieves 70-80% accuracy in identifying accounts that will close, vs 20-30% with intent or scoring alone.

Next Steps

• Evaluate third-party intent providers for your industry (Bombora for tech, G2 for tech buyers)
• Identify your account scoring approach (rules vs probabilistic)
• Look for ABM platforms that include both intent and scoring (6sense, RollWorks)
• Model the ROI: cost of intent data and scoring vs revenue from better prioritization
• Book a demo at abmatic.ai/demo to see how intent data and account scoring work together

---

For more context: Account-Based Marketing vs Demand Generation