If you're sick of repeating yourself every time you open a new chat with Jenai. You know, Here's who I am, here's my tone, Here's what I do, then this episode is for you. Neo Applan, Inventium's Jenai Guru is back to walk us through one of his favorite tricks for getting consistently great results from Jenai, and it's all about having a briefing document. Welcome to How I Work, a show about habits, rituals, and strategies for optimizing your date.
I'm your host, doctor Amantha Imber. Two years ago, I completely overhauled how I work with Jenai and I'm now saving over forty hours every single week. That is no exaggeration, and that's exactly why my company, Inventium created the Jenai Productivity Upgrade. It's a twelve whe course designed to move you from AI doubler to productivity machine. No fluff, just practical strategies that will pay off from week one, saving
you at least ten hours every single week. You'll learn how to make AI sound exactly like you, use AI as your second brain to excel at your job, and so much more. Whether you're a complete beginner or already dabbling. We've got you covered, starting with prompting fundamentals and going all the way through to advanced automations and agentic AI. We kick off on July fourteen, and spots are limited. Visit inventium dot com, dot au, forward slash Genai hyphen
cohort to secure your place now. There's a link to that in the show notes, and you've got nothing to lose because there's a seven day money back guarantee, So head to the link in the show notes and check out the program today. So, Neo, what's the actual problem that we're trying to solve for When we talk about using briefing documents.
JENAI generally doesn't know everything about you, and so every time you open up a new chat thread, you need to give it the context about what you're trying to achieve, where you're working, potentially, your industry, all those kind of things, and so building up that context can take a fair bit of time. We've talked in previous episodes about getting
it to ask you questions and things like that. This is all about building context so it understands a lot more about you and it can give you a personalized result because knowing more about you and what you're trying to achieve. Turns it from being a pretty good result engine into giving you an engine which produces an excellent result. And so if you're going to give this context all the time, it can be a little bit painful every
time you've got a new chat thread. So briefing documents are a cheat fast way to get that context into the GENAI to get a better result.
Okay, so what exactly does a briefing document look like?
Like?
What should go into one?
Okay, Really it's up to you and may some people might have many briefing documents, but at its heart it would be things like your role, what you do, the company you work for, maybe your team structure, maybe your KPIs. It might have some things about how you want to work with jen AI. It might be your voice and style prompts and how you want to have different tones in there. So pretty much everything that AI would need
to know so it can do a great job. So you don't have to give that information every single time. So to see this as a living, breathing document of here's me, here's what I need to do, Here's how I need to work with you AI. Here's the briefing document for you, and so you'll have this, you just upload it to a new chat thread and then it has a fair bit of background in what you're trying to achieve.
So do you recommend having the one mega briefing document with basically a full history of your life and everything about it, or like, in all seriousness, I should we be splitting them out and having different categories of briefing documents.
I do recommend that, Yes, for most people, you might find that one briefing document is fine and enough. But you might find that you might have maybe there's a project at work and you need to have a lot of background for this particular project. The people as different stakeholders, different needs, different goals, different communications you need to build for it. In that case, I would have a separate briefing document to say me my normal job and certainly
one from my own personal life. So I would recommend most people would have two to three briefing documents, one for each kind of facet on why you would want to use GENAI. The fewer the briefing documents is better because you're not going to confuse the AI. It's going to know if you're talking about project day with a project day briefing document that it's only going to have all information for that and it doesn't need to know
anything else, so it won't get confused as much. So the other thing about that is it's a smaller document, which means it won't fill up the context window, which is its short term memory. Gives it a little bit more memory there for it as well.
Okay, so when you say shorter document, I mean how many pages in a word document should it ideally be kept under.
I'd say under ten pages for sure. I'd say five pages is probably a good number for a briefing document. So if you're going to blow over five pages, I'd say look at maybe splitting it up into two or maybe three briefing documents.
Okay, so let's imagine that I've created a briefing document. How does that actually change how I'm prompting the AI to do the task that I want it to do.
All you need to do is, in each new thread, you upload the briefing document and then you ask your question, but also get it to refer to the briefing document. So you might upload the document say I need you to do this task, refer to the briefing document for the tone, or for the background, or for the context and things like that, and then Jenna I will get all the information from the briefing document and put it
in the outcome. So here's an example. You might say, I need to write an email to a stakeholder on a project. Here's the project briefing document. It will then know your name, it'll know what you're trying to achieve, it'll know who that stakeholder is, and will then be able to craft the email in a right way for that project, for you and for that stakeholder. So the first draft is a lot closer to what you need rather than just a generic first draft of an email
to any kind of person. This one will sound like you. It'll fit with your needs and your context and what you're trying to achieve a lot faster, so it's less tweaking for you to have to do later on.
Okay, so let's imagine I'm in a thread. I've prompted the AI to do the task with the briefing document within that thread, do I need to keep reminding the AI to refer to the briefing document or once I've uploaded it, does it just remember and it knows always refer to this briefing document for this thread.
There are a couple of reasons why you would need to remind it to look at the briefing document. One of them is some of the genais they want to draw information from their own memory rather than from documents who have uploaded, and so sometimes they forget that they need to refer to this briefing document. They call it
prompt adherents. So all you need to do is, if it's not referring to it, all you need to do is write the words, make sure you refer to the briefing document, and then it all redo its last response and give you a much better response from there. Other things you need to have to use with the briefing document is if you're having a really long conversation with the genai, it might run out of its short term memory that context window, so you may need to re upload it and re remind it if you need to.
But most of the time they're pretty good in being able to recognize that you have this briefing document and you continually refer to it, and so therefore it needs to continually refer to it as well.
Okay, are there any simple tests that you have built in to just check, oh it is the GENAI. Still referring to the brief email documents.
Yeah, you will know that. Most of the time, I've got a little cheat's way, which is, whenever it writes emails, I hate that GENAI loves to write, I hope this email finds you well. I'm sure people have seen that. It's a common sign for this is written by AI. It's the same thing as the m dash being in there. So I've got a line at the bottom of my briefing document. So for any email, never write I hope
this email finds you well. And if I find that indeed, I'm getting it to write me an email and it does say hype funds you well, then clearly it's not reading everything in the briefing document or it's not taking that into account with its output. And so that's a great tell for me to see that it isn't reading the briefing document. Same thing you might want to use is Australian English right in there. Use Australian English spelling,
but not aussieisms, otherwise it'll go guy mate. And if it starts talking in America, and then you'll know that it hasn't actually read everything in the briefing document, so you might need to re upload it or re reminded of the briefing document. So there's two great techniques there.
Okay, So one final question, I mean, is this the same as building a custom GPT in chat GPT or a gem in Gemini? Like, are we talking about the same thing here?
Your kind of It's the document itself can be used in so many different ways. So the first one is some people like to use more than one Genai, so they can say, oh, I got one output from a Jenai, I don't want a second opinion from another Jenai. In this case, I can upload my briefing document to both of them and then I can get different responses. So that's one of the great ways. The other is some people don't have the paid plans, and so if you don't have the paid plan, you cannot build your own
custom GPT, so therefore there's not available for you. There's also another one which is in the Microsoft ecosystem. Their agents are slightly different to the GPTs, and I've found they sometimes work differently with briefing documents. So it's great to have these documents as a core tool. But yes, if you do upload them into a GPT or a projecting Claude or a gem in Gemini, then these are
the core source knowledge for each of these. So whether you're using these great tools, GPTs or gems, this briefing document is a great tool you'll need anyway.
Nio, thank you so much. I know that for me creating different briefing documents has been such a time saver, So for anyone listening, I strongly encourage you to try this out today. If you like today's Joe, make sure you follow on your podcast app to be alerted when new episodes drop. How I Work was recorded on the traditional land of the Warrangery people, part of the Cool And Nation. A big thank you to Martin Immer for doing the sound mix.