Lead Scoring model¶
A proposed lead-qualification model that clearly separates a freshly captured lead from an MQL (Marketing Qualified Lead) ready for accelerated nurturing and an SQL (Sales Qualified Lead) ready for handoff to sales.
Context¶
Without a formal scoring model, every lead captured by the campaign enters the CRM with the same weight, which makes it harder to measure the true efficiency of paid investment and creates noise for the sales team. A CFO at a manufacturing company who requests a demo should not be treated the same as a visitor who only downloads a brochure.
The objective of the model is to assign a 0–100 score that combines firmographic signals (industry, size, role, geography) with behavioral signals (lead actions across site, email, calls, and sales sequences).
Model structure¶
Two qualification axes that add up to 100 points:
| Axis | Maximum points | What it measures |
|---|---|---|
| Firmographic | 50 | How well the lead fits the ICP (industry, size, role, country) |
| Behavioral | 50 | Observable intent (form, browsing, email, calls) |
Qualification thresholds:
- Lead: 0 – 49 points. Enters automated nurturing.
- MQL: 50 – 74 points. Accelerated nurturing; sales notified for review.
- SQL: 75 points or more, plus a sales action (call scheduled, proposal sent, etc.).
Axis A — Firmographic (50 points)¶
| Criterion | Weight | Logic |
|---|---|---|
| Target industry | 15 | +15 if the lead belongs to one of the 7 priority industries (Manufacturing, Retail, Automotive, Food & Beverage, Oil & Gas, Construction, Logistics). 0 otherwise. |
| Company size | 15 | +15 if the company exceeds MXN 50M in annual revenue or has more than 250 employees. +8 if between MXN 20–50M. 0 if smaller. |
| Role | 15 | +15 for C-level or Director (CFO, CRO, Credit Director). +8 for Manager (Credit Manager, Compliance Manager). +3 for Analyst. 0 for others. |
| Country | 5 | +5 for Mexico. 0 outside. |
Enrichment is done by combining the corporate email domain with enrichment services (Clearbit, ZoomInfo, Salesforce LinkedIn enrichment) to complete industry, size, and role when the form does not ask for them directly.
Axis B — Behavioral (50 points)¶
| Action | Weight | Data source |
|---|---|---|
| Requests demo (full form) | 20 | Webflow → Salesforce |
| Visits 3 or more pages in a session | 5 | GA4 |
| Returns to the site (second session or more) | 5 | GA4 |
| Downloads whitepaper or case study | 10 | Webflow form |
| Opens 3 or more nurture emails | 5 | Customer.io → Salesforce |
| Clicks a CTA in a nurture email | 10 | Customer.io → Salesforce |
| Opens an email from a sales cadence | 5 | Gong Engage → Salesforce |
| Click or reply within a sales cadence | 15 | Gong Engage → Salesforce |
| Call scheduled or completed | 20 | Gong Foundation → Salesforce |
| Positive or negative sentiment on the call | +/−10 | Gong Foundation |
| Attends a webinar | 15 | (future) |
| 30 days of inactivity with no new signals | −10 | Automatic decay |
Control rules:
- Signals are capped at 50 points on this axis. Accumulation cannot exceed that limit.
- Email opens and clicks count only once per person within a 30-day rolling window, to avoid inflation from bots or email-warming.
- Automatic decay prevents inactive leads from occupying pipeline.
Example¶
A CFO at a manufacturing company with 1,500 employees, located in Mexico, who:
- Requests a demo (20 points)
- Target industry: manufacturing (15 points)
- Size: large company (15 points)
- Role: C-level (15 points)
- Country: Mexico (5 points)
- Visits 4 pages in the session (5 points)
Total: 75 points → SQL. Immediate notification to sales for handoff.
Technical implementation¶
The model is implemented in Salesforce Enterprise Edition using Flow Builder. Each lead action triggers a Flow that adds or subtracts points on a custom field (Lead_Score__c). Firmographic criteria are computed at lead-creation time.
Advantages of this implementation:
- No additional cost. All required capability is included in the Enterprise license.
- Transparent. The sales team can see why a lead has a given score, rule by rule.
- Tunable. Weights and thresholds are tuned monthly based on real outcomes.
- Auditable. Every change is logged in Flow.
Future paths:
- Einstein Lead Scoring (Salesforce add-on): from month 6 onward, if CIAL chooses to license it, it can run in parallel with the rule-based model to compare predictions. The recommendation is to complement, not replace.
Required integrations¶
For the behavioral axis to operate in full, the following integrations must be active and forwarding events to Salesforce:
Customer.io → Salesforce. Captures open, click, and unsubscribe events from marketing-automation emails. Native integration available; validated during technical configuration.
Gong Engage → Salesforce. Captures opens, clicks, and replies from the outbound sales cadences. Native integration available. Confirm which events sync and to which object (Lead, Contact, Activity).
Gong Foundation → Salesforce. Associates each call with metadata (duration, sentiment, topics, talk ratio) to the corresponding lead or opportunity. Enables automatic SQL promotion when voice activity is present.
GA4 → Salesforce. Not a native connection. Resolved by capturing UTMs and the GA4 client_id as hidden fields on the Webflow form, which travel with the lead into the CRM.
Webflow → Salesforce. Critical flow to preserve UTMs and attribution. Mapped during technical configuration.
Next steps¶
- Joint validation of weights and thresholds with CIAL's marketing and sales teams.
- Technical mapping of the integrations detailed above.
- Build of the Flows in Salesforce with the
Lead_Score__ccustom field and routing rules. - Sandbox testing with sample leads.
- Production activation at the start of the pilot.
- Monthly review of weights during the optimization cycle.