AI-Era Information Workflow
Memo to Email to GPT
Keep short memos in emails to yourself, then pick only what you need and organize it with AI. Sending a memo does not share it with any AI: you remove confidential details, paste the memos in yourself, and check the output against your original memos. This guide covers all of that as one routine.
A 3-step routine: review your memos in emails to yourself and give an AI only the parts you choose. It includes checking what you share before you paste it, and checking the AI's output against the original memos.
Step 1: Capture with Memos
The first step in information management is capture. Meeting comments, commute insights, reading notes -- recording these without loss is the foundation.
3 Capture principles:
- Speed first -- use a tool that starts in 0.4 seconds (warm launch, iPhone 16e)
- Quantity over quality -- organize later, write now
- Add one line of context -- write "why I noted this"
Step 2: Accumulate in Email
Reasons to use email as your information hub:
- Universal: Accessible on all platforms and devices
- Searchable: Gmail and Outlook search can find them
- Storage: How long mail is kept and how much fits depend on your mail service, so back up the memos you need
- Structured: Listed by received date with subject lines; labels or folders let you sort them further
- Exportable: Text can be transferred anywhere
Memos sent via Simple Memo accumulate in your inbox. They aren't locked into a dedicated app's database, so you can reread them on any device that gets your email.
Step 3: Leverage with GPT
Pick the memos you need from your email and paste them into an AI yourself, and it can draft the following. Email and the AI are not connected automatically.
- Summarization: Draft a weekly report from a week's notes
- Classification: Auto-sort scattered memos by category
- Analysis: Identify behavioral trends from memo patterns
- Action items: Auto-generate todo lists from notes
- Content creation: Generate blog posts or reports from memo fragments
FAQ
What is the Memo-Email-GPT workflow?
A 3-stage framework: capture via memo, accumulate in email as database, feed to GPT for analysis and structuring.
Best method for information overload?
Split into 3 stages: Capture broadly, Filter strictly, Utilize deeply. Don't try to organize everything at once.
Why use email as information hub?
Available on all platforms, powerful search, chronological auto-organization, and long-term storage (retention and quota depend on your mail provider).
Effective Prompts for Feeding Memos to GPT
Use this prompt when you pick a few of the week's memos for your weekly review. Search your emails to yourself, pull out only the parts you're comfortable sharing, and paste them into the AI with the following prompt.
Prompt example:
"Below are my notes from this week. Please organize them from 3 perspectives: (1) key achievements and progress, (2) recurring themes or patterns, (3) tasks and ideas to carry over to next week. Please summarize as concise bullet points."
The key is specifying "3 perspectives." Letting GPT freely summarize tends to produce verbose output; limiting perspectives yields actionable results.
Use this when expanding fragmented memos into fuller ideas, especially for brainstorming records and insight notes.
Prompt example:
"Below are fragmented ideas noted at different times. Are there common themes between these notes, or combinations that could form new ideas? Please explain each possibility concretely."
Humans struggle to find connections between notes, but GPT excels at pattern recognition. It can discover unexpected links across a month of scattered notes.
Convert rough meeting notes captured via Simple Memo into structured minutes.
Prompt example:
"Below are rough meeting notes. Please create meeting minutes in this format: [Meeting Name] [Date/Time] [Attendees (infer from notes)] [Agenda & Decisions] [Action Items (assignee, deadline)] [Unresolved Items]. Mark uncertain information as [To Confirm]."
Including "[To Confirm]" in the instruction makes GPT less likely to fill gaps with guesses. At work, it matters that uncertain points are flagged rather than made up.
Effective prompts share three common principles:
- Specify output format: "As bullet points," "in table format," "from 3 perspectives" -- explicit formatting prevents verbose prose
- Handle uncertainty explicitly: "Don't guess," "mark unknowns as [To Confirm]" -- an important technique for reducing made-up details at work
- Provide context: Adding background such as "These are sales team weekly meeting notes" helps GPT produce output that fits your purpose
Concrete Memo-to-Email-to-AI Workflow
Understanding theory is one thing; daily practice is what matters. Here is the concrete workflow for integrating Memo-to-Email-to-AI into your daily routine.
Morning (Capture Preparation):
- Place Simple Memo in the most accessible position on your home screen
- During commute: instantly memo insights, ad ideas, podcast takeaways
- Follow "1 memo = 1 topic" strictly. Keep it short. Add one line of context
During the Day (Capture Execution):
- Memo notable meeting remarks ("Tanaka: said Q3 budget needs revision")
- Capture useful Slack/chat information (even just URLs)
- Note lunch conversation ideas
- The memos you send with Simple Memo pile up in your email inbox
Evening (AI Processing):
- From your emails to yourself, pick only the memos you need for today's review
- Remove anything you can't share, check the AI service's data settings, then paste them in with an "organize today's notes" prompt
- Check the AI's output against the original memos, and turn only the confirmed items into tomorrow's tasks
- Star important memos in Gmail for retention
Weekend (Weekly Review):
- Draft a weekly report from the memos you picked, and delete anything that isn't in the originals
- Look for recurring themes, then go back to the actual memos to confirm them
- Use GPT output as reference material for next week's planning
The Memo-to-Email-to-GPT workflow can also be used beyond basic reviews.
- Reading note synthesis: Memo interesting passages while reading via Simple Memo. After finishing, GPT generates "book summary + my key highlights"
- Interview review: Only with memos you're allowed to share outside and that don't identify people or customers, draft a list of issues that come up repeatedly
- Learning reinforcement: Feed study notes to GPT asking "what areas show weak understanding" to identify review priorities
- Diary review: If you'd rather not hand private records to an outside service, skip the AI and reread them yourself at month-end