Executive Summaries Exercise

[00:00:00]
>> Using that same formatting exercises memo that we have back on this page. So everything's starting down be here. I want you to go and create an executive summary, a bluff, TLDR. Basically like a little blurb that if someone read it and they agreed with it, they could just skip reading the rest of this entire doc, right?

[00:00:21]
So give you a bit of time and you can press pause on the video and then come back here after you're done and we'll talk about it. The other thing to try is I want you to at least try writing yourself once I find it's useful for people to kind of take large amounts of information and crystallize it down to your point.

[00:00:40]
That's what the entire exercise of this is for. But then try putting it into some sort of no chat CPT or barred or something like that and see what they give you back as well. Then you can kind of compare and contrast you can see that AI is useful but also it's probably not gonna get you 100% of the way there.

[00:00:57]
It's gonna get you like 80, 90% of that, and then you're gonna have to use your squishy human meat to get it all the way to 100%. We are gonna take this formatting access and we're gonna write a TLDR about it. What is the point here? So okay, before you might ask what the point is like who's reading this?

[00:01:19]
Well, if you remember up here at the top, it says that is the lead senior leadership team. Let's pretend that's six people. The CEO, the CPO, the CTO, VP of this, VP of that and your mother. I don't know whoever the last one is, that's the other last person there.

[00:01:36]
Okay, and then we're gonna write a update for them. So, we set up here that our point is the project is going well but it is behind schedule and you wanted to you want to communicate the delay, but how it's not that big of a delay. So that's really what we want to call out here is, there is a delay in the project, it's because we discovered new user research.

[00:02:00]
The project will still be delivered two weeks behind on this date. That's essentially what I would put in the bluff. So that anyone reading is like it's delayed, this is why it's delayed, this is when you can expect results. And I think most senior leaders would be, great, that's what I wanted to hear, right?

[00:02:21]
And then if they want to get more context of why, what the moving pieces are, who can I talk to about this, how can I dive deeper? The rest of the memo is there for that, but the bluff gets mostly there. So how did this go for you?

[00:02:35]
Is that close to what most of you landed on? Is that close to a chatGPT told you to do.
>> It maybe was the user didn't help me that way. [LAUGH]
>> Yeah. You'll find like the prompt engineering here, and there is [INAUDIBLE] course on how to prompt engineer well.

[00:02:58]
But really driving in to the point helps chat GPT get to the point that you want to get it to. And it's almost to the point where write the bluff yourself, to get to give you the bluff that you want, right? At which point, what are you doing?

[00:03:13]
That I don't know? It's actually kind of interesting, I did use ChatGPT for this course to write all the SEO functionality of this. So let me show you. I'm quite proud of this. So I'm going to show you anyway. If you look down here in the head, and there's this description, I'll make this bigger for you.

[00:03:32]
But we don't need all the rules right now, just make that smaller. So you can see here all this stuff, I actually had chatGPT right the summaries which is the ones that have passed us to Google, right? These are the ones that Google would show underneath it, and it's what Google uses to index it.

[00:03:50]
And so I actually wrote into this project the ability to go through and write all of these descriptions cuz I hate writing [LAUGH] them, right? And I didn't even really care how good they were, but these actually ended up being pretty good, right? And you can see here also these keywords.

[00:04:06]
These ones, which ones? Keywords, these ones were all pulled out by chat GPT as well. And I can show you here, I am not that talented with It's not going to be in here, it's going to be the next core starter. I'm not that talented with AI. It's not something I use consistently every day.

[00:04:33]
And there's people that are much better that are able to get much better results. I think Scott Moss, another teacher here on front of masters, is leagues a bit better than me at it. But I'll show you the prompt here that I was able to get fairly useful descriptions out of this, It's here, this is not great to read.

[00:04:54]
I do have it open. So I can open get prompt, this is what I wrote here. So, I mean, this is obviously working for trying to get it to rank well in Google, but you could easily modify something like this and giving it a bit more context is like, hey, you're writing this for executives.

[00:05:19]
They want to understand this at a glance. How would you use this? Using as plain language as possible so that it's, you know, very easy to read and very direct to the point. And any like tidbits and tips that you can give the agent here will bode well for however your prompt is.

[00:05:33]
So you can see here, I wrote all more English than the content that are then the summary actually expected was and that's gonna be pretty typical, right? So yeah Justin.
>> Just a piggyback on that. Yeah Scott moss does have just updated his full stack Next.js course that incorporates AI and some prompt engineering into it.

[00:05:58]
And then also Max Firtman just has released a course on, first Look, ChatGPT API for web developers that kinda looks at using the API in general if people wanna see a little bit more how to use that.
>> Awesome. Very applicable to product management it is coming for our jobs.

[00:06:14]
I'm just kidding, I mean, maybe but right now I think it's more help than harm to the product management profession.