Google's Gemini 3 Just Prepped My $200K Sales Call in 15 Minutes

Speaker A: Gemini 3 was released today and I am absolutely blown away. I'm going to walk you through three use cases. One where it helps me prepare for a $200,000 sales call in minutes. On top of it too, creates a real time dashboard and even a $100,000 website from scratch live. So I'm going to walk you through all this. I have some examples built out, but let's get right into it. Okay, so I think like step one is I want to give you like context of why this is such a big deal. I don't want to be one of those hype men who talks about like, this is the most amazing change ever if it isn't. Okay. However, I want to show you some stuff. I just dropped this post on LinkedIn and what you could see are these are the benchmarks where it's significantly exceeding what's available in the market today. So there's a lot in here. I think you could get lost in these details if you're not a technical person. The part I want to point out the most is if you look at the screens. Screen Spot Pro. Okay, so this is screen understanding. It, it happens in basically lands at 72.7% whereas GPT5 is 3.5. Cloudson, it's 36 and 11.4. Okay, this is what, this is important because of computer use, um, and basically agentic capability. And this is why a lot of people are most excited. But what I wanted to do is show how we could use it for you in real world examples right now. So the one I want to share with you is a $200,000 sales call prep. So as you can see here, one of the things that I did is I had this set up and I have a pretty strong prompt, right? There's everything from research to preparation to logistics. And you know, I asked it to present in a visual layout. So I didn't even ask it to go into deep research. And it's like, huh, huh, this is pretty thick. So I'm going to go into deep research now. What you're going to see here is basically I obviously changed the names and how I'd approach it for the sales call. But I said, all right, what if we were trying to get into Google AI Studio? And I just kind of guessed and said it was a $300 million tech startup with an Alphabet, right? So within Google, the Google AI Studio, I thought it was kind of met up. So what it does is it breaks down all these different areas about different focus parts of the company with, um, intrinsic Innovation, sandbox, aq and then one of the other things that goes into financial and operation, leadership, competitive intelligence and market position. As you can see, it's got OpenAI in there. Um, legacy OEMs, key leadership profiles, industry challenges. And as we go down, I'm not going to go through this whole thing for you right now because you're going to be bored. But I'm going to show you the goods which are coming up. Right. So it's got a lot of details which I thought this was really good. A 12 slide pitch deck narrative of, uh, what could be possible. And the speaker notes are really impressive with what it has. So that was pretty solid. It's got follow up email templates, logistics and action items. Okay. And then concluding insights. Now one of the things that you could take to make this lawn text file, there's a little tab over here called Create. So what I did is I created an infographic from it. I'm going to show you the result of that because I think you'll really appreciate it. I was really impressed with it. So this is what it built literally in minutes. And so it built a whole sales call prep dashboard was like, holy cow. And I've used this before. But uh, this is really, really dialed in. I mean there's a couple areas like here, like key call key contacts, right. Doesn't have the names in there. But what I would say is got key discovery questions, pitch strategy. Um, it's got uh, objection responses. And check this out. It even clicks down, right? So you could even like go into these, make these bigger, smaller. So it's truly an interactive infographic. I can even check these boxes over here. Um, and then it's got a mapping of the 12 slide pitch deck with different follow up email templates that I could click on as well. All right, so overall I was really impressed. And then I even add Gemini features to it, which is, as you can see here, I could now ask it to cover, um, an objection like we don't have budget, right. And then what it'll do is it'll generate a response. And this is really, really freaking impressive in terms of that. Um, and some of these, you know, like for one shot, this isn't bad. Uh, I thought this was really insanely good. Now focus of manufacturing, that's a little off, right? It's not manufacturing unless you're talking about like the Waymo card. So that's example one. Okay. Example two. I created a gem before where it helps me design effectively. Like um, it takes from a text post and it helps me create a visual. And so what you're going to see here is it has like a hand drawn like post, if you will. So this is what I created with one shot. Um, I basically just gave it the post. What I said is a cover Gemini 3.0top 7 features. And as you can see, it's really good in here. There's this little kind of like markup here, which I don't know what that is. Same here. But the rest of it is pretty dialed in. So I'm sure if I took a little bit of time editing like this one. Deal. Slippy, Slippy. Okay, so that was a little bit of a mess, but it's got my name, it's really clean and crisp. Um, so from an image perspective, I was pretty impressed with that. Obviously there's more work to be done, um, so that we can get it one shot where you don't have like kind of the typos or other areas. So that's part two. Okay. Now this one I was really surprised with. Okay. And I've heard things in terms of ui, um, and improvements. And this is my. So what I had to do is check this prompt out. You're going to love this. Okay. So I said I need you to rebuild my website so it looks like it cost $100,000. Because I'm now 100 focused on AI. Ah, transformation for sales teams, marketing C suite teams. Right. I typically work with CRO or cmo. Please design a beautiful one. This is my current website. Okay. So check this out. Like, this is freaking impressive. And like, the beautiful thing about it is I could build this out for you in real time and you'll be. Actually, I'll do it with the next one. I won't do this one. But look at this. Like, super sharp. This, this spot needs a little bit improvement. But no editing, no back and forth. I thought that was super solid. So that's like upgrading an existing website. Now if we went to build a website from scratch, okay, this is another example. So I had to build a fake AI prompt, uh, website. And then here's what I said. I said create a million dollar AI prompt generator. Sassy front end, fully responsive. Don't use tailwind css. Give me a code preview. Okay, so this is what it created and it actually works. Um, so let me do this. I'm going to bring out a new chat so you could see what this looks like in real time. Now if you're doing this, you want to click on canvas. We'll put this code in or I should say this prompt, not this code. You got to make sure it's on. Thinking right fast isn't going to do it fast as the old model. And so what you're going to see happen here is as it builds this out, you're going to see it's going to build out all the code first, and it might be like four or five hundred lines of code. And then what happens after that is it's going to shift and it's going to give you like a visual. So this is the code build out on what's happening. And the thing that's really impressive about this is exactly like, it's really freaking fast. Like, I mean, like insanely fast. Like, I've been talking for 10 seconds and we already got almost 200 lines of code to build this out and it's still flying. Look at this. Insane, right? I am not a coder by any means. I just gave it a simple command instruction to do that. And what you're going to see is literally, I think if we timed it with a clock, it'd probably be about 30 seconds by the time that this was all finished. Uh, which blows me away, right? Not just the speed, but also the quality. And the look of this is really, really good. This is testing very, very strong in terms of what I'm seeing. And look, it's still building this out. It's going about 600 lines of code. And it came up with the name Lumina AI Sass. And then what you're going to see is it's going to shift to preview. I tested. This is one of the first tests I ran. As it came out, I was like, okay, this is pretty freaking cool. So this looks like it's going even in more depth than it did before. I think yesterday's the one I showed you had about 500 lines of code. So why don't we do this? I'll show you one more example and then we'll come back to this. So this was another one that blew me away. Um, what I asked for to do is to create an interactive Q4 sales dashboard. I gave some like, kind of fake metrics with wind cycles. And I said, you know, make it interactive. Design it like a modern sales CRM. Okay, look at this thing. Same thing. Like, if you look at the code, this is kind of like what it's doing on the other page. It created 576 lines of code. But then look at this. Like, you could toggle the win rate, the average deal size. And as you can see the active pipeline number increases and decreases with that. Okay. Like you could extend the quarter. You could toggle different segments. Um, it's got like fake refs. Lewis Lift, Harvey Specter. Um, fantastic. It's got top performers, active opportunities, all these areas. Right. And I was like shocked. Like I didn't have to go back and forth with this at all. This is, um, what it did. Now I did kind of like. Look, let me fast forward that. I have multiple versions of this. I think I did. I think this is the only one. Okay. No, I think I just used the first one. Right. So I think that was the first one that I just showed you. So let's go back to our other example. All right, here's the prompt website. Um, Lumina AI. So just to kind of give you a recap, what we did is it spit out 921 lines of code in about a minute, which is bonkers. And then stop wrestling with words. Lumina Transform. Big ideas and highly engineered prompts. Um, using enterprise nlp. Right. So it's got this in here. Let's see if it works. 10 million prompts like this is pretty solid. It's even got a pricing page and you could start a free trial. Okay, so let's just see if it works. I wonder if it works this time. Um, why don't we do a killer prompt for a front end website design? Actually, let's do something that you would probably use to, um, A marketing campaign for a new prompt generator company. All right, let's see if. Let's see what happens here. See if it spits it out. Oh, there it is. All right. System role. Here's the optimized marketing campaign for a new prompt generation company. Award winning photography, depth of field. Okay, let's try this. Let's. Let's just try this prompt and see how it works. It's pretty simple. It's nothing, Nothing super crazy. Um, and it's pretty generic too. But let's see what it says. Marquee campaign for a new prompt generator company. Award winning photography, depth of field. Okay, I probably should have put this in code, but that's okay. So now what it's doing is it's building this out. And you could see like this is the thought process of this model, which is insanely high. So I think this is probably going to be a little too long for me just to have you sit and wait on this. Okay, here it goes. No, it was fast. Right, so this bigger prompt over the shoulder. Okay. The creative flow, the tack. Okay, so this is all. It basically gave three image prompts to generate that. You want to do it? Okay, let's heck, let's do it. Um, close up, over the shoulder shot. Okay, let's just see if we could switch it right in here, create an image. And this is probably going to take a little while. I'm going to switch to fast, so you can see, see how this works. Uh, actually this might take a minute or so. So while that's working, I'm going to stop sharing and we'll come back to that. Just to kind of summarize things. So all in all, like I said, oh, it finished already. It's even faster than I thought it was going to be. Okay, here it is. So this is what I came up with, which I think is. Okay, Magic type prompt generator. Okay. I would say this is pretty solid, you know, from, from that super quick example. Uh, okay, so just to summarize this. Like, I wanted this to be quick, but I wanted to get it out to you. So in less than 15 minutes you had live reviews and examples of what Gemini 3 was possible of. Um, I have different access levels across different licenses. I have, I should say, with Google, and this is really, really, really impressive. Um, I think this has massive potential to basically catapult Google past OpenAI for the first time in terms of like their overall model performance. Uh, and this is across all modes, which, which surprises me because, you know, if you look back about a year when they had Bard, it was so far behind. So it just shows kind of all the hard work that Google's done. I think this is a tide that lifts all boats. Right. So I think this is gonna, you know, accelerate the, the speed to outcome that open AI will have probably with GPT6. Um, as long as lawn is like Claude 5, Grock 5, you name it, it's going to continue to level up and these are going to get better and better. But I'm absolutely super excited to do more follow up details on this. Wanted to share this with you as soon as it came out. Say, had access. All right, appreciate you joining me on this video and we'll see you on the next one.