CVM is having its AI moment
The old playbook — segment, target, offer, measure — didn't die. It just got a new engine. Here's why I'm writing about it.
Customer Value Management has always been a simple promise with brutal execution: get the right offer, to the right customer, at the right moment — and prove it moved the needle.
For twenty years the bottleneck wasn't the idea. It was the labor. Building segments, wiring campaigns, waiting on analysts, arguing about attribution. By the time a play shipped, the moment had passed.
AI doesn't replace that playbook. It removes the friction between having an idea and testing it.
What actually changes
- Segmentation stops being a quarterly project and becomes a sentence you type.
- Targeting moves from broad rules to per-customer propensity, cheaply.
- Listening scales — every ticket, review and survey becomes a signal, not a sample.
- Iteration collapses from weeks to hours, so you run more, smaller bets.
Idea → live play (median days)
None of this is magic. It's the same discipline, with the boring parts automated.
Why this blog
I've spent my career close to lifecycle and retention work, and I keep building small tools to scratch my own itches. This is where I'll think out loud about the new CVM — part notebook, part playbook, and occasionally a place to show what I'm tinkering with in the lab.
Short posts. Real opinions. Things you can use on Monday.