Case Study / Lead Qualification / 02
01 / Project Overview
This workflow reads newly enriched Lead Intake records, evaluates each prospect against predefined ICP criteria, assigns an AI qualification score and priority, and determines whether the lead should enter the active CRM sales process. Qualified leads are synchronized with GoHighLevel; non-qualified prospects are retained for visibility without entering automated outreach.
Input
Existing enriched Lead Intake records not yet processed by AI qualification.
Decision
Documented AI assessment with deterministic score and priority normalization.
Output
Qualified GoHighLevel handoff or retained not-qualified result records.
The Challenge
Without a qualification layer, sales teams must manually review job titles, decision-maker relevance, company size, industry fit, business context, contact quality, CRM suitability, lead priority, qualification notes, and location or phone data. This slows outreach, creates inconsistent decisions, and can introduce low-quality leads into the sales pipeline.
The Solution
The n8n workflow processes unqualified Lead Intake records, prepares structured data for the AI model, evaluates ICP fit, normalizes the response, and applies fixed scoring rules before routing the lead. Qualified prospects are synchronized with GoHighLevel; non-qualified prospects remain logged without entering active outreach.
03 / Workflow Architecture
The workflow evaluates enriched prospects, normalizes AI qualification results, applies defined scoring rules, and routes eligible leads into GoHighLevel.
n8n Workflow Architecture
Lead Intake processing, AI qualification, score normalization, qualified routing, and GoHighLevel CRM synchronization.
Architecture Details
Lead Intake: Filters records that are complete and have not yet completed AI qualification.
AI qualification: Evaluates ICP fit, normalizes score and priority, and prepares structured assessment data.
Decision routing: Sends qualified records to GoHighLevel while retaining non-qualified results and updating processing state.
Stack / Tools
Workflow orchestration
AI assessment
Data storage
Data control
Downstream CRM
Workflow Demonstration
A recorded walkthrough of the supplied n8n workflow, presented with native browser controls.
04 / Workflow Breakdown
Common qualification process
Start WF2
Read Lead Intake
Filter Leads to Process
Loop Leads
Prepare Qualification Input
Request Pacing
AI Lead Qualification
Normalize Qualification Result
Qualified?
Yes / Qualified path
Prepare GHL Contact
Upsert GHL Contact
Save Qualified Result
Mark AI Qualification Done
Continue Loop
No / Not qualified path
Save Not Qualified Result
Mark AI Qualification Done
Continue Loop
Loop Done
WF2 Complete
AI Qualification Criteria
The AI evaluates decision-maker seniority, job-title relevance, company size, industry fit, automation potential, business context, contact quality, and overall ICP alignment. Decision-maker relevance is one of the strongest scoring factors, while industry supports the decision without determining qualification by itself.
Qualification Scoring
High priority
Medium priority
Low priority
Qualification threshold
AI Score ≥ 70The workflow recalculates qualification and priority from these thresholds, so routing does not depend only on free-form AI interpretation.Leads Selected for Processing
The workflow checks Person ID, email, verified email status, job title, company name, and AI Qualification Done?. Only records where AI Qualification Done? != true continue, preventing repeated scoring.
Structured AI Output
The model returns structured qualification data that can be normalized and stored reliably before downstream routing decisions are made.
Request Pacing
A deliberate delay before AI qualification prevents large lead batches from being sent to the provider simultaneously, reduces burst traffic, and makes processing more controlled.
GoHighLevel Handoff
Qualified leads are sent through a GoHighLevel contact upsert designed to create contacts when needed, update them when applicable, reduce duplicates, retain AI qualification data, and prepare the record for downstream workflows.
This n8n workflow does not create opportunities or execute the complete sales process. GoHighLevel handles the downstream opportunity, outreach, follow-up, reply, appointment, and pipeline operations.
CRM Data Normalization
Invalid spreadsheet or error values in phone fields are converted to null; only verified phone data is included when available. Location values are parsed into CRM-friendly city, state, and country fields.
Stored on the GoHighLevel contact: Prospeo Person ID, AI Score, Lead Priority, Qualification Summary, Recommended Action, Lead Source Detail, and AI Qualified.
Qualification Results Storage
Results are stored for visibility and reporting. After either route completes, the original Lead Intake record is updated to AI Qualification Done? = true, preventing repeat processing on future runs.
05 / Key Features
Automated AI lead qualification
ICP-based lead evaluation
1–100 AI scoring
HIGH / MEDIUM / LOW priority routing
Deterministic qualification thresholds
Structured AI output
Phone validation
Location normalization
CRM payload preparation
GoHighLevel contact upsert
Qualified / non-qualified routing
Qualification history storage
Duplicate processing prevention
Controlled request pacing
Downstream sales automation handoff
Role in the Full System
The preceding workflow finds and enriches prospects. This workflow decides which leads should enter the sales process. GoHighLevel then manages the downstream communication and appointment lifecycle.
n8n + Prospeo
n8n + OpenRouter
Initial outreach · Follow-ups · Reply handling · Appointment lifecycle
06 / System Outcome
This workflow transforms enriched prospects into structured CRM-ready contacts by evaluating ICP fit, assigning documented scores and priorities, generating qualification summaries and recommended actions, filtering low-fit prospects, normalizing contact data, synchronizing qualified contacts with GoHighLevel, preserving qualification history, and preventing repeat runs.