How to use ChatGPT in your PM work

Real-life examples (and actual prompts) of how PMs are already using ChatGPT day-to-day

👋 Hey, I’m Lenny and welcome to a 🔒 subscriber-only edition 🔒 of my weekly newsletter. Each week I tackle reader questions about building product, driving growth, and accelerating your career.

---

Are PMs actually using ChatGPT/GPT-4 for getting real work done, and if so, how?

Let this idea sink in:

Siqi Chen @blader

product development is basically just prompt engineering the engineers to rotate words into the right shapes. soon it’ll be prompt engineering all the way down.

Though I don’t believe PMs will be displaced by GPT-4/5/6/n, I do believe that, as with most knowledge work, learning to work alongside AI will quickly become table stakes. Just like Grammarly became for writing, Copilot for engineers, and Firefly, Runway, Midjourney, and others are becoming for designers. It’s scary and confusing, but also exciting. When was the last time there was a transformative new tool for product management? Never?

Shreyas Doshi @shreyas

The year is 2027. GPT-7 does most of the work on product teams: mocks, code, research, data analysis. There are highly leveraged engineers, designers, analysts working across 5-10 products. Every product still has a PM, who has 55 GPT tabs open. PM now means Prompt Manager 😁

deepu @pradeepb28

I wonder if @lennysan @shreyas have thoughts or predictions about what the product management would look like in next 5 years ?

Importantly, as Dan Shipper suggested in his recent guest post on building the Lenny Chatbot, “the best way to prepare for this fast-approaching future is to dive in and get your hands dirty.” So to help you roll up your sleeves and get started, I polled Twitter for concrete examples of how PMs are already using ChatGPT day-to-day. Honestly, I was shocked by how common (and useful) it has already become for people. Below, I’ve collected a dozen of my favorite use cases of how you can integrate ChatGPT into your work today—with actual prompts you can play with.

To follow along, go sign up for ChatGPT (seriously, go do that now), and try at least one prompt you find below. Even if you don’t find it useful today, it’ll plant a seed.

Also, a pro tip: You can ask ChatGPT for advice on how to best phrase your prompt. It’s very meta and very cool.

For example:

Barada Sahu @mercurialsolo

@lennysan @laurynisford P1: "Imagine you are a Product manager making a prompt to improve the onboarding for Airtable. Please write the prompt" Fed the prompt and came up with ideas ============================ P2: These are pretty common and already done, can you come up with more creative ideas?

Here’s what it suggested when I did this:

The future is coming fast.

12 ways to use ChatGPT in your PM work

1. Collect and summarize user feedback and usage data

Synthesize survey results:

Lucas Didier @lucdid

@lennysan Something like: "I'm going to send you a list of X survey responses to the question "...." Can you group these into buckets of insights, with their associated weight (in %), and list some examples of associated responses for each insight?"

Find feature ideas and bugs from app store reviews:

Max Schubert @MaximalTweet

@lennysan Another one from yesterday evening: _________ I want to scrape App Store reviews of our app using Javascript. Walk me through it step by step. Then, I pasted all the reviews into the chat and asked GPT-4 to find the top 3 most requested features and complaints.

Extract insights from raw usage metrics:

RossMcLoughlin @Dotnetster

@lennysan Converting our product usage metrics data to text and then just asking GPT questions about the data.

Do sensitivity analysis:

Mark Rogers @tlonuqbar

@lennysan You can paste in the data in a table and ask for the conversion percentages. And then ask scenario questions. "What is the volume of raw leads that would be needed to generate x conversions?"

2. Come up with product name suggestions

Bailey Jennings @Bailey_Jennings

@lennysan Product/Feature naming: prompt: I am going to give you a description of a new software product I am planning to launch. I'd like you to return 10 product/feature name ideas. Try to avoid using cliches like "Product 2.0". Do you have any questions?

Jillian Funes @jillianfunes

@lennysan @angara_ps2c I wish I had saved the prompt, but essentially I use this format: "You are a PM at a company that does X. You are launching a new product that does Y. Come up with 10 catchy, brandable names for it." It's come up with some pretty good ideas! 🙂

3. Strengthen your argument

Come up with critical questions your audience may ask:

Vybhav Ramachandran @vybhavram

@lennysan PRD Review: "Assume you are the CTO, review this PRD and give me critical, but fair feedback) Before meetings : "I'm meeting the VP of data science for a 30 minute call to discuss X. What should I ask her?" After meetings : "Summarize my notes into minutes, action items"

Identify gaps and hidden assumptions in your thinking:

Rishi Kar @rishi_kar

@lennysan I really like how helpful GPT-4 is when trying to avoid getting blind sided. My favourite prompt when thinking about GTM or a feature: “What am I missing here?” “What am I being overly optimistic about?” “What is a macro event that can totally reverse the outcomes of this test”

Kranthi Kiran @kranthitech

@lennysan Identifying assumptions in any hypothesis. This is an example of our visual mapping tool. Trying to uncover assumptions because of which a user experience may not be as expected.

Highlight edge cases and counterarguments:

oliver @olivercitrin

@lennysan Incredibly helpful w removing personal bias and/or understanding edge cases. 1- ‘Tell me 5 reasons this feature won’t work as intended’ 2- ‘Tell me 5 unintended consequences of this feature’

adit @dagaadit

@lennysan Steelmanning the other side of an argument when drafting a proposal (“why shouldn’t we do this?”)

4. Inspire roadmap ideas