Getting started
The first month with a team. What to teach first, what to leave alone until later, and how to tell three weeks afterwards whether any of it actually stuck.
Answering customer emails with AI without sounding like a form letter
Customers can tell. Not because a tool drafted the reply, but because it answers a question they didn't ask, in a voice nobody at your business uses.
4 minute readGetting startedBefore it leaves the building: how to check an AI draft in five minutes
The draft reads well. That is precisely the problem: fluent and correct are different properties, and the tools are much better at the first.
4 minute readGetting startedHow to pick the one job to fix first
Everyone wants to know where to start. The answer is one job, and it's rarely the one that irritates you most.
4 minute readGetting startedThe training that didn't stick: five reasons a good session changes nothing
A room full of enthusiastic people is not a result. Here's what separates a session that changes a job from one that changes a mood.
4 minute readGetting startedThe first ninety minutes: what to teach a team before anything else
There's an order to this, and prompts are not first. Here's what ninety minutes should actually contain.
4 minute readThe short version
Where information goes, then one real job start to finish, then the tool decision, then the setups your team keeps. Prompt technique sits in the middle and is not the opener. Three weeks later, ask which job is being done differently and by whom — one named job is a successful session, and enthusiasm with no named job changed nothing.
The full curriculum is under training.
Not sure which of these applies to you?
Two minutes describing your situation will get you to the right starting point faster than reading four articles.