Module 4: Pre-Launch Risk Assessment Agent

Catch campaign failures before they happen with systematic pre-launch checks

What This Agent Does

Score launch readiness before spend
Flag critical setup gaps (tracking, creative, targeting)
Assess risk level (low/medium/high)
Suggest fixes for each identified risk
Generate launch approval checklist

This prevents costly mistakes before budget is wasted.

Prerequisites

Completed Modules 1-3
Campaign setup documentation available
Access to campaign preview/setup details

Step 1: Define Pre-Launch Standard

Create: `workflows/prelaunch_assessment_standard.md`

Objective: Prevent avoidable failures by scoring launch readiness.

Critical Checks (Must Pass)

Conversion tracking verified
Landing page live and functional
Budget aligns with expected volume
Creative approved and spec-compliant
Targeting matches ICP

Red Flags (Auto-Reject)

No conversion tracking
Broken landing page
Budget under $50/day for cold traffic
Single creative asset (no backup)

Risk Scoring

Low risk: All critical checks pass, 1-2 minor issues
Medium risk: 1 critical check fails OR 3+ minor issues
High risk: 2+ critical checks fail OR red flag present

Step 2: Prepare Launch Inputs

Before launch, gather these files in `/inputs`:

bash

campaign_setup_details.md (targeting, budget, creative assets)

bash

tracking_verification.md (conversion events, pixels)

bash

landing_page_url.txt

Be thorough. Missing details will result in incomplete risk assessment.

Step 3: Create Risk Assessment Workflow

Assessment Steps

Run through critical checks
Identify red flags
Calculate risk score
List gaps with priority
Recommend: approve, fix, or reject

Rules

Never approve high-risk launches
Require fix timeline for medium-risk
Document assumptions made

Output Format

Risk score
Gap list (prioritized)
Approval decision
Fix recommendations

Step 4: Run Pre-Launch Agent

bash

claude "Execute workflows/prelaunch_risk_workflow.md for [campaign name]."

Step 5: Save Output

bash

outputs/prelaunch_assessment_[campaign_name]_YYYY_MM_DD.md

Step 6: Human Override

Verify agent's assumptions about setup
Add context agent can't see (brand approvals, legal)
Make final launch decision

Step 7: Iterate After Launches

Review: Which red flags were accurate?
Update critical checks based on new failures
Refine risk scoring thresholds

Common Failure Modes

Too many false positives

Cause: Critical checks too strict

Fix: Separate "critical" from "best practice"

Checklist fatigue

Cause: Too many minor checks

Fix: Focus on top 5 failure modes only

Missed edge cases

Cause: Standard not updated after failures

Fix: Add post-mortem review to iteration step

Why This Works

Prevention is cheaper than fixing live campaigns
Checklists prevent human oversight
Risk is scored, not assumed
Launch quality becomes consistent

This is how bad launches stop happening.

← Previous: Module 3: Budget Reallocation Agent Next: Module 5: Team Adoption & Guardrails →