AI Scoring Engine: Real-Time Sequence Flow

here's how GPTWeb's AI Scoring Engine works in real-time as a visitor engages with your site. This sequence diagram shows the flow from first interaction through DQL qualification.

Initial Score Calculation:
The moment a visitor arrives, GPTWeb creates a profile and assigns a base score based on source (organic search vs paid ad), referrer (coming from high-intent page), and device (desktop users often score higher than mobile browsers). First Question Analysis:
The AI Engine processes the question through RAG (Retrieval Augmented Generation) and updates two scores:

• Engagement Score (+5): Measures conversation depth


• Behavioral Score (+3): Captures basic intent signal Intent Signal Detection:

As conversation continues, the AI analyzes each question for buying intent signals. Questions about pricing, implementation, or integrations trigger higher behavioral scores. A pricing question might add +8 points because it indicates active evaluation. Firmographic Scoring:
When the visitor reveals their company (either explicitly or through email domain), GPTWeb looks up company size, industry, and revenue. If the company matches your ideal customer profile, the firmographic score increases significantly (+15 in this example). Demographic Scoring:
Job title and department information trigger demographic scoring. A VP of Marketing scores higher than an intern. Decision makers get +12 or more points because they have buying authority. DQL Threshold Crossed:
The Scoring Engine continuously calculates the composite score (sum of all four dimensions). When it crosses your configured threshold (typically 60-70), the visitor becomes a Discussion Qualified Lead. Automatic Actions:
Once DQL status is achieved:

Record is created in Data Storage with full conversation history

CRM sync triggers automatically (HubSpot/Salesforce)

Alert agent notifies sales team

Frontend can display relevant campaigns or offers

The sequence diagram shows score updates, but here's what each dimension actually measures:

Traditional lead scoring systems batch-process scores every few hours or overnight. By the time sales sees a hot lead, the visitor is long gone. GPTWeb's sequence is real-time:

• Scores update with every message


• DQL status triggers instantly


• CRM syncs happen in under 2 seconds


• Sales alerts fire while the visitor is still engaged This means your sales team can reach out while the conversation is fresh, while the visitor is still on your site, and while their intent is highest. That's the power of real-time AI scoring - you're not chasing yesterday's leads, you're engaging today's buyers.

this sequence represents what's happening behind the scenes right now as you're using GPTWeb. Every question you ask updates your engagement score. Every topic you explore signals behavioral intent. The system is continuously evaluating and scoring - not to be creepy, but to ensure that when you cross that threshold, your organization's team knows you're serious. GPTWeb is the future of engagement, websites and marketing automation combined - built for the AI era, built for now.