Pipeline

Episode 5 · Oct 23, 2026 · 27 min

Do the Thinking Before You Go to AI (with Sanjana Balaraman)

Prefer audio? Listen here
Chapters8 chapters

The more you lean on AI for your posts and cold emails, the closer they drift to the median.

Everyone in sales and marketing is using AI now, and that's why so much of the copy reads the same. On this episode of Pipeline, Joseph Lewin talks with Sanjana Balaraman, Founder at ChapterOne Creative, about using AI without letting it do your thinking. Sanjana spent two and a half years at an AI startup. Her rule is to form a point of view first and go to AI with it. She brings conversation transcripts into her prompts, directs the prompt instead of chatting with it like a colleague, and tells it to verify its sources. Her blog writing went from half a day to half an hour. But 25 minutes of that half hour was her editing the draft until it sounded like her.

What you can use by Monday:

Write out or voice-note your own point of view before you open ChatGPT or Claude, then give AI that point of view to draft from.

Add customer call transcripts or notes from a feedback conversation to the prompt, and tell AI how to lay out the draft and what to take from each one.

Put one line in every research prompt telling AI to use verified sources and say where each number came from.

Keep a list of words that read as AI and check every LinkedIn post against it before it goes out.

Before a cold email goes out, add something that's very you, like a bit of humor.

Chapters:

Why AI pulls content toward the median

Do the thinking first, then let AI draft

Where AI does the grunt work and where it doesn't

Why every piece needs your lens on it

Get ideas from your head and from customers before prompting

Tell AI to use verified sources

Editing the first draft to sound like you

What ChapterOne Creative helps with

Pipeline is short conversations with GTM peers who are actually filling pipeline right now, so you leave with something you can use by Monday. New episodes weekly on Apple Podcasts, Spotify, and wherever you listen. Produced by Podcast 2 Pipeline.