What Actually Gets Written Down: Rules, Actions, Lessons, Tasks

When we talk about a "second brain" for your business, it can sound abstract. A memory system. A knowledge base. Sure - but what's actually in it? What does the AI reach for when someone asks it a question about your business?

The honest answer: it's a folder of plain text files. Nothing fancier than that. But the contents of those files are what turn a generic AI into something that acts like it's worked at your business for years. Let's break down what actually gets written down.

The Four Things a Second Brain Remembers

Every well-built second brain has four kinds of notes living inside it. They're simple on their own. Together, they're what make the whole system work.

1. Rules - How Your Business Does Things

These are your standing policies. The things you'd tell a new employee on their first day. How you price rush work. Which areas you actually cover. What "urgent" means to your office manager. Which vendors get priority. Whether you take cash.

Nothing here is technical. It's just written down in plain language, once, so that every AI-powered task - a drafted reply, a meeting summary, a follow-up email - pulls from the same source of truth. No more "how do we handle this again?" pinging around a group chat.

2. Actions - A Running Log of What's Been Done

Every time the AI does something on your behalf, it writes down what it did. Drafted the reply to the Alvarez inquiry at 2:14 PM. Sent the proposal to the Henderson account. Moved the Ramirez meeting because the account lead was running late.

This isn't just record-keeping. It's what makes the AI aware of what's already happened, so it doesn't call the same customer three times or send duplicate quotes. It's the difference between an assistant with amnesia and one that remembers the conversation you had this morning.

3. Lessons - Things Learned Along the Way

This is the part most people don't expect. As the AI runs, it takes notes on what worked and what didn't. "Customers in this neighborhood usually ask about pricing before scheduling - offer a range up front." "This referral partner sends leads that convert well; prioritize their callbacks." "Requests flagged urgent from this channel usually aren't; ask two clarifying questions before escalating."

Over time, this file grows. Your business gets smarter in a way that doesn't live in one person's head anymore. If your best employee left tomorrow, most of what they knew would leave with them. A second brain fixes that.

4. Tasks - What's Still Outstanding

The open loops. The quote you promised to send by Friday. The customer who asked for a callback next week. The vendor waiting on your PO. Every commitment made - by you, by a teammate, by the AI itself - gets tracked here until it's closed out.

A Day in the Life

Say you run a small business. It's Tuesday morning.

A customer email lands at 8:47 AM asking whether you can deliver by Thursday. An AI assistant drafts the reply. It checks the rules file - yes, you cover that area; yes, Thursday works. A teammate approves the draft, and the assistant writes to the actions log: "Replied to J. Chen, confirmed Thursday delivery, quoted the standard range."

Twenty minutes later, another inquiry. Different customer, but from the same area as three recent ones. The assistant notices - because a note in the lessons file flagged this area as price-sensitive - and puts the pricing range near the top of the draft. The conversation moves faster. That note got written weeks ago after a pattern emerged.

The customer asks for a follow-up on Monday to confirm details. The AI drops that into the tasks file. Monday morning, it surfaces. Nothing falls through.

At the end of the day, you glance at the log. You can see every action, every commitment, every note. Nothing is hidden inside a black box.

Why This Matters

The reason to build a second brain this way - plain text, four kinds of notes, all visible to you - is that it's the only setup that actually holds up over time. AI without this structure guesses. AI with it acts like it knows your business. And when something needs to change - a new pricing rule, a new service area, a lesson learned from a bad call - you change one file and every tool that reads it picks up the change.

This is also why implementation matters more than the AI itself. Even Microsoft, after spending billions on AI rollouts internally, concluded that the hard part isn't the model - it's the setup around it. That's why the structure matters as much as the tool.

Start Small

You don't need special software to start. Open a folder, create four files - rules, actions, lessons, tasks - and write down the three questions your team asks most often this week. If you want more plain-English pieces like this, the blog has a Friday digest that sums up each week's posts.