Lenny Rachitsky’s Post

As much as I'd like to buy a tool like this for my growth stack, I've found it easier and less of a hassle to build this scrappy workflow with Google Sheets <> Zapier <> ChatGPT Plus. I use this for experiment analysis + copy. 1. Train a custom GPT on brand guidelines & examples. 2. Google Sheets -> Zapier -> GPT connection. When I enter in my control copy, GPT will create 2 variants, explaining their hypothesis for each. 3. GPT -> Zapier -> Google Sheets connection. When I enter in experiment results, GPT will hypothesize why X variant won. 4. Upload a CSV of all experiment results to custom GPT again, so GPT can learn from winning experiments. Challenges: when analysing experiment results, it lacks context on the product & other external factors. But can't feed that to the GPT out of data privacy issues.

At Bayzat we've built an AI Insurance assistant that explains complex insurance policy terms and processes. For example, if someone is experiencing back pain, all they need to do is chat with their personal AI assistant to understand what their exact policy covers. The experience is seamless, real-time, intuitive, AND multilingual. Customer support and AM teams used to spend a lot of time answering these questions every day, now they can focus on more complex work. Also, customers get instant answers without waiting for someone to respond. https://www.linkedin.com/posts/tarek-bayaa-87523537_the-medical-insurance-experience-has-historically-activity-7085253505866735617-G83F?utm_source=share&utm_medium=member_desktop

For me, GPT revolutionizes sales! Our most impactful application of GPT lies in summarizing customer interviews in sales. We used to spend a ton of time conducting interviews and synthesizing findings to improve our sales process, find innovative feature ideas for R&D, and enhance marketing collateral. It's all history now. With GPT, we: ↳ Identify recurring pain points across all interviews. ↳ Find compelling customer testimonials for marketing materials. ↳ Pinpoint nuanced critiques points that we may have missed. ↳ Generate action items for every individual customer interaction to hand over to CSM/AE The benefits to our customer experience (CX) efforts are gigantic.

We've created a custom GPT that has our schema, company overview, and some of our product requirements in markdown format. We are in the middle of an intense push over the next 6 weeks to deliver a fairly substantial feature set. This custom GPT is fully armed to act as both a source for requirement truth, and brainstorming technical approaches. If you are a PM and want to take it to the next level, look at delivering not just the usual PRD, but also an internally shareable link to a custom GPT that has those same requirements added to its context. (Don't forget to disable the training toggle in setup). Your devs will love you.

We're using GPT, other LLMs, and RAG to search for, organize, and generate new files across SaaS apps (Gmail, Zoom, Slack, Telegram, Dropbox, Notion, etc). Like "gather all the documents for Glendale house sale", "get all X product docs, exclude financial and share with <email>", or "search for docs required for a French visa and fill out an application based on them".

We have one for interview synthesis (so much better now with actual quotes rather than hallucinations) - and an affinity mapper (linked). https://chat.openai.com/g/g-kkCWQtO4F-propel-affinity-mapper. So far it works better than the ones built into miro/dovetail (in our opinion) because we can refine the prompts.

We developed an AI-powered assistant for frontline agents that enables them to offer fast, precise, and personalized support to our customers by leveraging the power of LLMs. Tool sources information from internal knowledge repositories, Slack, etc. It then summarizes and displays it on a single screen. However, human validation of answers remains crucial. Agents love it as it reduces their effort, leading to excellent customer experience.

Our company builds a Statbot that can answer non-trivial statistics questions. The bot also retrieves our data to enhance the answer This is a post about another effort but in similar spirit: https://statsig.com/blog/askse-solutions-engineering-ai I use ChatGPT analytics to do at least half of my analytics work. Basically, it's like pairing with an intern or a junior with good coding skills but needs checking and going back and forth with answers. Sundas Khalid did a great interview with our engineers and PMs on how they use ChatGPT daily: https://youtu.be/i8Y71gtThkA. The video will be published sometime this week.

today I uploaded my lead magnet and asked CharGPT to write me a summary for the landing page. Was quite happy with the outcome 🚀

I’ve made a CEO coach gpt that I’ve trained on podcasts and books from my fav business leaders so I can ask it questions and advice. Works well!