Build a custom GPT model

This content was originally featured in the November 15th, 2023 newsletter found here: BUILD YOUR DIGITAL TWIN WITH CUSTOM GPT

Build a custom GPT model

Did you know you can now clone yourself?

Well, sort of.

Over the weekend, OpenAI rolled out the ability to create your own custom GPT model. Pretty cool, right?

Let’s not get ahead of ourselves. There are still a lot of limitations and considerations. As an experiment, Chris and I (mostly Chris) built a KatieGPT, using newsletter content as the foundation for the database. Since I’ve been writing the newsletter every week for a few years, there is a decent amount of content to pull from.

Once we loaded the content, gave some guidelines, and most importantly, updated the avatar, it was time to see what KatieGPT was capable of.

You

What are Katie’s latest thoughts on management?

GPT

KatieGPT

I searched through the available documents for specific insights into Katie’s latest thoughts on management but couldn’t find any direct references or detailed discussions on this topic. Her writings cover a variety of subjects, including marketing strategies, productivity, and the impact of technology on our work and personal lives, but they do not explicitly focus on her current views or philosophies regarding management.

If you’re looking for general insights on management from Katie, it might be helpful to consider her overall approach to topics like data-driven decision-making, transparency in communication, and the importance of adapting to change, as these themes could potentially reflect her management style and philosophy. However, without specific references in the documents, I can only make broad assumptions based on the general tone and content of her writings.

Welp.

I’m underwhelmed by KatieGPT. For context we gave the system all my cold opens from 2022 and most of the opens from 2023. I talk about management a lot. That’s what I do. That’s my area of expertise. However, despite having content that covers that topic, my doppelgänger had trouble finding useful content.

A couple of things could be the problem here. First, it could be that I don’t talk about management as much as I think I do. But let’s be honest, that’s not the issue. More realistically, we haven’t given the system enough content to work with. We also haven’t given enough diversity in the content. While the cold opens are good, they are from one point of view. They are me, talking to you, about a topic. They aren’t give and take. They have a specific tone.

My next step to start refining KatieGPT is to pull podcast and livestream transcripts. Then I would look for transcripts from talks and webinars. Next I would look at online conversations (where appropriate) and load all that content into the custom GPT model. The more data you give the system, the better the output will be.

This is, admittedly, and over simplification of building a custom GPT model. I’d recommend you go through actual planning exercises, like user stories, to help figure out what you need and how you’ll use it. Experimenting with a custom GPT model should be just that, an experiment. If you decide to use your model publicly, you’ll want to be much more thoughtful with how you build it. Think about the data you should be sharing. Think about the end user and what they will get out of the custom GPT model. Think about the maintenance plan and how you’ll keep the system up to date.

If you want more details on building a custom GPT model, or how we set up KatieGPT, you can check out this week’s podcast In-Ear Insights: Custom GPT & Software Development Best Practices

Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.