I implemented a repeatable phishing simulation program using an open source platform. I configured sending profiles, built safe landing pages, launched consented campaigns, and used the results to target awareness training. (How to Run a Phishing Simulation Campaign)
Stand up a simulation platform with strong safeguards
Create templates and landing pages that record events only
Launch a scoped campaign and monitor engagement
Convert findings into coaching and process changes
Organizational approval and documented consent
Access to a test Simple Mail Transfer Protocol account for sending
A server or virtual machine to host the platform
Open source phishing simulation platform
Simple Mail Transfer Protocol configuration tools
HTML and Cascading Style Sheets for email and landing design
1. Platform setup
Deployed the simulation server and secured the admin account.
Configured a sending profile and validated delivery to test inboxes.
2. Assets
Built a landing page that tracks visits and attempted submissions without storing secrets.
Authored a realistic but benign email template with a single call to action.
3. Audience and launch
Created a test group from consented participants.
Scheduled the campaign and verified telemetry end to end with a pilot.
4. Monitor and analyze
Tracked delivery, opens, clicks, and report events in real time.
Identified risk patterns by department and by template variant.
5. Train and iterate
Delivered targeted micro-trainings and sent tip sheets.
Ran a follow-up campaign to measure improvement.
A safe, repeatable program that measures susceptibility without shaming
Higher report rates and lower click rates on subsequent runs
Phishing simulation configuration and operations
Email deliverability testing and telemetry
Data analysis and awareness training design
Note: I only run simulations with explicit approval. Landing pages record events but never store credentials.