Most AI demos skip the hard bit. They show you someone typing a prompt and getting a beautiful output, and you're left wondering why yours never look like that. The difference isn't the tool. It's what happens before you open it.

In this session, Jake Bird - founder of JIM AI and a consultant who's spent four years helping B2B teams actually use AI, not just talk about it - pulls back the curtain on what good AI work really looks like. Using a live client example, he'll walk through the methodology behind effective prompting: the role definition, the format structure, the layered instructions, and crucially, the source information that most people never think to include.

You'll see the full prompt architecture before it's run, understand why each element matters, and watch the output that results. No magic, no jargon - just a repeatable framework you can take back to your desk and use on Monday. If you've ever felt like AI should be working better for you than it is, this session will show you why - and what to do about it.
 

Learn

• Why the quality of AI output depends almost entirely on the work done before you write the prompt
• A practical prompt architecture: role, format, instructions, and information categories
• What "prepared" looks like versus "winging it" - and the difference in results
• How to structure source information so AI actually has something useful to work with
• A repeatable framework for building prompts that deliver consistent, usable output

The prompt is the last 5% of the work - the other 95% is preparation, and that's where most marketers are falling short.