Depesh Mandalia presents a practical, case-study driven approach to improving ad performance by using custom GPT models and a BPM-method framework. He contrasts generic ChatGPT outputs with personalized custom GPT prompts that produce more actionable, account-specific recommendations and promises a walkthrough of interpreting AD account data to generate improvements.
Chapters
[DEPESH MANDALIA ]: For me, my experience is things that do work, things that I've actually spent money on and seen working. So if you go back to some of the examples, perhaps from just a general chat GPT response, these are more generalized. For me, I've created something more personalized. These are based on exact examples and case studies and things that we've done in the past. So the reason I'm sharing this to you is, is this is kind of how to build something that's more relevant for you. So if you go into chat GPT and just have a general conversation, it doesn't matter what it is, you're going to get generalized responses to an extent. If you can build a custom GPT where you can actually pre prompt and provide information and I'll show you how I've done that as well, it allows you to get better tailored responses. So I'm not going to worry too much about the CPM info here. So this is kind of based on BPM method framework and case studies and things like that. But what I wanted to show you here is how you can take an AD account, interpret the data and get some really good actionable steps to improve that data.