The method is in your head. Write it down once, and AI can run it.
Nina Brenes··6 min read·Updated Aug 2026
Key takeaways
- Playbooking turns the method in your head into a written playbook AI can run reliably.
- Two layers: the mechanical steps go to AI, and the judgment, what to keep and what only you would notice, stays with you.
- Five steps: name the purpose, write the steps, apply MASTER to each step, test and iterate, keep sharpening it as you use it.
- MASTER lives inside step three: each step gets its own context, examples, success criteria, and instructions.
- Playbooking is a framework from the AMP program (runonamp.com), where Nina certified as an AI Operator.
Watch how most people hand work to AI and it looks a lot like flipping a coin. You paste in a few examples, cross your fingers, and hope it comes back right. Sometimes it does. Often it does not, and you quietly redo it yourself. Real delegation, the kind where AI runs a whole process reliably instead of gambling on each try, has a name: playbooking. And here is the part that matters most if you are the expert and the business at once: the thing you are trying to hand over is not a task, it is the method that lives in your head, the one you have never fully written down.
Playbooking is a framework from AMP (formerly The AI Exchange), the program where I trained as an AI Operator. The idea is simple, and it changes how you work: instead of hoping for good output, you take what you know and how you actually do the work and write it into a playbook AI can run reliably. Imagine a doctor who has a method for turning a messy patient intake into a clean, structured note, a method she has never written down. Playbooking is the act of getting that method out of her head and onto the page, so the mechanical part stops living only in her attention. Then the hours you used to spend redoing things get spent on the work only you can do.
What a playbook is (and is not)
A playbook is not a long prompt, and it is not a folder of examples. It is your process written down as steps, with your standard for good baked into each one, so AI can run it without you looking over its shoulder. Think of the difference between doing a thing once, by feel, from memory, and writing the manual so it comes out your way every time even on a tired afternoon. That gap is exactly where the payoff lives: Bain's 2025 Technology Report found that using AI as a basic assistant yields only 10 to 15% productivity gains, while redesigning the whole process around it yields 25 to 30% (Bain & Company, 2025). The distance between those two numbers is the distance between prompting and playbooking, and it holds whether you are a whole company or a single practitioner.
The five steps to build a playbook
- Name the playbook's purpose: which part of your work it runs, and what a good result looks like.
- Write out the steps in the order you actually do them, and mark which are mechanical (AI can run) and which need your judgment (you keep).
- For each mechanical step, give it context, templates, examples, success criteria, and instructions (the MASTER Method).
- Test it against real work, and iterate on what actually breaks.
- Keep using it and keep sharpening it, so the playbook gets better every time your real work touches it.
MASTER lives inside each step
Step three is where the craft lives: each step of the playbook is written with the MASTER Method, also from AMP. Look closely and a playbook is just a chain of steps, each one carrying its own context, examples, and definition of “done well.” That's why the two frameworks are inseparable: MASTER is how you talk to AI inside a single step, and playbooking is how you chain those steps into a process that runs on its own.
Keep the judgment layer for yourself
The point of playbooking is not to hand the whole thing over. It is to give AI the mechanical layer, gathering, structuring, the first pass, and keep the judgment layer, what it means and what only you would notice, on rested attention. That layer runs on focus you have not spent, which is a reason to build rest and structure into your week on purpose, not by accident.
FAQ
What is playbooking in AI?
Playbooking is turning a process you already carry in your head into a structured, step-by-step playbook that AI can run reliably, with your standard built into each step. It is how you delegate a whole process, not just a single task.
How do I teach AI to follow the way I do my work?
Write your process as ordered steps, and for each mechanical step give AI context, examples, and success criteria using the MASTER Method. Keep the judgment steps for yourself. Then test and iterate against real work. That structured playbook is what AI follows reliably.
What is the difference between a prompt and a playbook?
A prompt handles one step or task. A playbook chains many steps into a whole process, each step written with its own prompt. A prompt is a sentence, a playbook is the manual.
What parts of my work can I turn into an AI playbook?
Recurring work with a definable standard: client onboarding, first drafts of proposals or reports, weekly summaries, content production, turning notes into structured documents. Anything you could write a manual for, you can turn into a playbook, while you keep the judgment calls.
Nina Brenes
AI partner for thoughtful experts · Founder of The Harmony Labs
Nina helps thoughtful experts turn what's in their head into AI-powered work that still sounds like them. She reads the research, tests the tools, and sits in the conferences so the people she works with get the short version: what is real, what to try this week, and what to ignore. Her focus is practical judgment, not hype.
Connect on LinkedIn
How to actually tell AI what you want, so you are not fighting it all day.
What are AI operations for one expert? The playbooks that let your work run without living only in your head
