In partnership with
Last Thursday I deleted about 40% of my Notion vault. Not in a rage-slowly, the way you clean out a closet when you finally admit a jacket no longer fits.
I had spent years inside Tiago Forte’s PARA method. Projects, Areas, Resources, Archives. Hundreds of notes, all tagged. I was proud of it the way some people are proud of a tidy garage.
Then I pasted chunks of it into Claude to think through a positioning problem. I watched the model do, in eleven seconds, the connecting work I’d been promising myself I would get to next quarter. It didn’t care which folder things lived in. It cared what I gave it.
PARA solved a real problem. A decade ago, finding that one article on pricing psychology three months later required a system-search was bad, tags were inconsistent. Forte gave us four buckets and a discipline. It worked.
Want to get the most out of ChatGPT?
ChatGPT is a superpower if you know how to use it correctly.
Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.
Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.
The retrieval problem is mostly gone. A model can find anything across thousands of notes in seconds, including things you forgot you wrote.
The bottleneck has moved. The question is no longer “where did I put it?” It’s “what subset of my thinking should this model see right now?”
That’s a different skill. The old work was organized for human filing satisfaction. The new work is curating context-picking the twenty pages out of two thousand that actually shape how the model thinks about your problem.
Mara, a product manager at a fintech I worked with, ran into this directly. She had two years of notes sorted neatly into PARA. When she started using Claude for strategy work, the answers came back generically. The model was smart; her inputs were too broad.
So one Saturday she did something different. For each active project, she wrote a one-page brief: who the customer is, what the team has tried, what didn’t work, what the current bet is, what she’s confused about. She kept them at the top of each project folder.
The Architecture Behind AI-Native Revenue Automation
In our new white paper, The Architecture Behind AI-Native Revenue Automation, Tabs CTO Deepak Bapat breaks down what it actually takes to apply AI to revenue workflows without breaking the books.
You’ll learn why probabilistic reasoning isn’t enough for finance, how Tabs pairs LLMs with deterministic logic, and why a unified Commercial Graph is the foundation for scalable, audit-ready automation. From contract interpretation to cash application, this paper goes deep on where AI belongs—and where it absolutely doesn’t.
If you’re evaluating AI for billing, collections, or revenue operations, this is the architecture perspective most vendors won’t show you.
Monday she dropped one brief into Claude with a question about pricing. The response was sharper than three consultant decks she had paid for. Same model. Same person. Different input.
Meta’s analytics team published a writeup in April 2026 on an AI assistant they built on top of PARA, now used by more than 60,000 people across the company-engineers, PMs, designers, legal, finance, sales.
The design choice is worth copying: each session starts with a lean root context file summarizing who you are and what you’re working on, and drills into specific project folders only when the conversation requires it. They called it progressive disclosure.
Don’t dump your whole vault into the model. Give it the right one page first, then let it ask for more.
Your accounting should keep pace with your business. BELAY's Financial Experts handle books, payroll, and reporting so you can lead with confidence. Download the Free Guide
Pick one active project. Open a blank document and write a single page: who it’s for, what you’ve tried, what failed, what you’re betting on, what’s still unclear. Don’t make it pretty. Make it honest.
That page is now the most valuable file in your second brain. Next time you open a model to think through that project, paste it first, then ask your question.
You’ll know in one exchange whether this is real. I think it is, but I’m still figuring out where it breaks.
Meta’s full writeup is here.
—Prompt N Productive—