Two CRM campaigns: one simulated, one real
A simulated bank campaign in SQL and Mautic, measured against a group that got no message, and my real job letters sent through Mautic and my own mail server.
Dashboard
Job Outreach (REAL DATA, n = 5, read on 2026-11-11): 5 letters sent, 5 accepted by the receiving server, 1 reply from people, 0 clicks by people (3 raw hits before the scanner filter). Counts, never rates: there are too few letters for any conclusion. Opens are not counted, because the letters carry no tracking pixel.
CRM Campaign (SIMULATED DATA, run 2, 2000 synthetic clients): 1700 chosen by the SQL rule, 9 of 9 safety checks passed; savings account within 30 days: holdout 5 of 163, framing A 13 of 754, framing B 10 of 783. Neither the campaign against the holdout nor A against B is shown: both intervals include zero, and the holdout group is too small for the small effect the simulation gives. No winner is picked.
- CRM Campaign (Simulated data): A bank campaign, from choosing the clients to measuring the result. A fictional bank and synthetic clients, so no message reached a real person. No winner: the first run was inconclusive, and both readings of run 2 are “not shown”.
- Job Outreach (Real data): Job letters from my own mail server, measured without a pixel. Real recipients, and no company is named. No tracking pixel: clicks go through my own redirect, which every letter discloses.
Limits
- Simulated, not a real campaign. In CRM Campaign the bank is made up, the clients are synthetic, and their behaviour comes from a simulator. The runs show the method and the machinery; they say nothing about how real clients behave.
- Few letters, no conclusions. Job Outreach reports counts with their date, never rates, and says nothing about which letter, subject line or kind of company works better.
- Mautic is a stand-in. Mautic is an open-source stand-in, neither Bloomreach nor Salesforce Marketing Cloud Engagement, and I have no production experience with any of the three.
AI agents (Claude Code) collected the target companies and wrote the code, the SQL, the letter drafts and the first versions of the texts under my direction. In CRM Campaign I chose the campaign idea, the target rule and the A/B hypothesis from options the agents proposed; in Job Outreach each batch of letters went out only after my yes.
Write email