More in Dashboards & reporting.
Analyze a Substack newsletter export with ChatGPT agent mode
WorkedEvidence: Described by posterPosted Last checked
Clint Cain merged a Substack export with ChatGPT Agent Mode and got blunt advice: stay on 1 topic and use fewer CTAs.
How to use this
- Tap Copy prompt.
- Open the tool the case used (or a similar one) and paste the prompt; results may differ from the original.
- Paste it into the message box.
- Send it, then read the answer before acting on it.
New to this? What's a prompt?
A prompt is the message you send to the AI tool. Copy the one on this page, put your own details where it asks, and read the answer before you act on it.
Prompt
These are the person's own words, as shared in the source.
For each row in posts.csv, use the post_id value to find the matching HTML file(s). Extract the plain‑text content from that HTML file and add it to a new column called post_content in the CSV. If no matching HTML file exists for a row, leave post_content blank. The result should be an updated CSV where every post has its associated text content (stripped of HTML markup) in the post_content column of the final file as post_content.csv .
Try this in:
Sends the Original prompt shown above. ChatGPT and Grok open with it filled in, and ChatGPT may send it straight away. Gemini and Claude open empty, so we copy the prompt for you to paste. Every button also copies it.
Opens in a new tab. We don't see what you send.
These links are only a shortcut. We're not affiliated with OpenAI, Google, Anthropic or xAI, and your results may not match the source.
Optional stronger version
Our rewrite with extra checks. The poster didn't test it.
Before you start, ask me 1–3 short questions about the details below and wait for my answers. Once I've answered, confirm the plan in one line and run the whole task start to finish. Goal: Combine my newsletter export into one spreadsheet, then review my writing and give honest, specific feedback. Questions to ask me first: - Which platform is the export from? - Have you removed files with subscriber emails or other personal data? - What do you most want to improve: topics, hooks, calls to action or format? Inputs (fill in or confirm): - Posts list: [UPLOAD POSTS CSV] - Post files with no subscriber data: [UPLOAD COMPRESSED FOLDER] - My goals and audience: [GOALS / AUDIENCE] Rules: - First, match each post to its file and add the plain text in a new column; list any posts with no match. - Don't open or use any file with subscriber names or email addresses; tell me if you find one. - Base every finding on specific posts, and quote short lines as examples. - Give 3 keep doing and 3 stop doing points, ranked by impact. - Don't invent numbers like open rates; say what's missing if you need data I didn't give. Checkpoint: After building the merged file, show me row counts and any missing matches before the analysis. If you get stuck: If a file won't open or match, tell me which one and why, then continue with the rest. Verify before you finish: Check that the merged file has one row per post and no HTML tags in the text. Output: The merged CSV, then a short report with examples from my posts. Done when: I have one clean spreadsheet and a short list of changes for my next posts.
Customize the stronger version
Answer these questions and we'll put your answers into the stronger version. The original prompt above stays exactly as published.
Your file stays on your device. Nothing is uploaded to TriedPrompts.
Your prompt
Before you start, ask me 1–3 short questions about the details below and wait for my answers. Once I've answered, confirm the plan in one line and run the whole task start to finish. Goal: Combine my newsletter export into one spreadsheet, then review my writing and give honest, specific feedback. Questions to ask me first: - Which platform is the export from? - Have you removed files with subscriber emails or other personal data? - What do you most want to improve: topics, hooks, calls to action or format? Inputs (fill in or confirm): - Posts list: [UPLOAD POSTS CSV] - Post files with no subscriber data: [UPLOAD COMPRESSED FOLDER] - My goals and audience: [GOALS / AUDIENCE] Rules: - First, match each post to its file and add the plain text in a new column; list any posts with no match. - Don't open or use any file with subscriber names or email addresses; tell me if you find one. - Base every finding on specific posts, and quote short lines as examples. - Give 3 keep doing and 3 stop doing points, ranked by impact. - Don't invent numbers like open rates; say what's missing if you need data I didn't give. Checkpoint: After building the merged file, show me row counts and any missing matches before the analysis. If you get stuck: If a file won't open or match, tell me which one and why, then continue with the rest. Verify before you finish: Check that the merged file has one row per post and no HTML tags in the text. Output: The merged CSV, then a short report with examples from my posts. Done when: I have one clean spreadsheet and a short list of changes for my next posts.
Scroll to see the whole prompt. Your answers show in green.
3 blanks are still empty: Posts list; Post files with no subscriber data; My goals and audience. They'll stay in brackets in your prompt, so the chat tool may ask you about them.
What you type stays in this browser tab. We don't send it or save it, and it's gone when you leave the page.
AI tool
Outcome
Clint Cain of Tech Wing Management exported a Substack newsletter's data, then deleted the per-post deliveries and opens CSV files first, because they contain subscriber email addresses that shouldn't be shared with AI. In ChatGPT's Agent Mode, Cain attached posts.csv and a compressed folder of the post HTML files and asked it to pull the plain text of each post into a new post_content column, matched by post_id. In the same chat, Cain then attached the merged file and asked it, as a newsletter analyst, about the most valuable content, voice consistency, hooks, calls to action, 3 recommendations, similar publications and 3 keep doing and 3 stop doing items. ChatGPT 'almost hurt my feelings,' Cain wrote: stay on 1 topic, cut back on too many CTAs, ask real-world questions, restructure the articles and share the YouTube video.
Evidence: Described by posterThe person describes what happened in words, without screenshots or video.
Source
- Posted
- by Clint Cain (Tech Wing Management)
- Last checked
- Method:
- How we verify
https://techwingman.substack.com/p/i-replaced-my-substack-data-analyst
Context: Cain removed the files with subscriber email addresses before uploading anything; do the same with any export.
More in Dashboards & reporting
Graph monthly traffic deaths for two states with Grok
Grok graphed five years of monthly traffic deaths in Iowa and California, to check a roadside death counter.
WorkedGrokEvidence: Screenshots or logsView CaseBuild a support data pipeline and dashboard in under an hour with Devin
Devin's customer story says an Eight Sleep support manager built a data pipeline and dashboard themselves with Devin in under an hour.
WorkedDevinEvidence: Company case studyPoster says:Under an hour
View CaseSet up a Google Analytics dashboard with Claude in Chrome
Claude in Chrome spent 45 minutes on a Google Analytics dashboard, got stuck on one drag step and undid all its work: a failure.
FailureClaude in ChromeEvidence: Described by posterView CaseLet a whole team self-serve data via an agent-written schema guide
Claude Code wrote a guide to the company's data model so Hex's agent could handle the team's data requests.
WorkedClaude Code +1Evidence: Described by posterView CaseAutomate a daily analytics summary to Slack with Claude Code
Claude Code set up a GitHub Actions job that pulls PostHog data at 8am, summarizes it and posts to Slack; it worked the next day.
WorkedClaude CodeEvidence: Described by posterView Case