Everyday AI Podcast – An AI and ChatGPT Podcast - EP 554: No-Code Interactive AI: How to use Canvas, Artifacts for business growth

Episode Date: June 25, 2025

No coding skills? No problem. When using ChatGPT's Canvas, Google Gemini Canvas and Claude Artifacts, you can create interactive dashboards and bespoke web apps in a flash. Ready to put AI to work for you this Wednesday? Let's get it. (Oh, and we'll be able to break a lil news midway through!) Newsletter: Sign up for our free daily newsletterMore on this Episode: Episode PageJoin the discussion: Thoughts on this? Join the convo.Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: [e-mail] with Jordan on LinkedInTopics Covered in This Episode:No-Code AI Dashboard Creation TechniquesGoogle Gemini Canvas vs. ChatGPTClaude Artifacts Interactive AI ToolsBusiness Analytics with AI DashboardsData Visualization in AI PlatformsReal-Time AI Tool Live DemoComparing AI Dashboard FunctionalityEnhancements in AI Tool ResponsivenessTimestamps:00:00 Everyday AI: Leverage for Business03:25 "Exploring AI Modes and Examples"09:11 Ready Check Before Starting10:37 Boost Productivity with Custom Tools15:23 Leveraging Public and Licensed Data17:49 "Efficient Debugging Solutions"21:01 "Spotting Blind Spots with AI"26:17 "SONNET Over Opus Frustrations"27:24 "Optimizing B2B Prospects with Opus"30:33 "Growth Lab Insights: Content Amplification"34:06 Evaluating AI Platforms37:44 Podcast Update: Video Available Online40:34 Growth Opportunity Radar Challenge43:55 Analyzing Albatross Data Effectively46:57 Upload Error with Fifth File50:32 "AI as No-Code Development Partner"Keywords:No code interactive AI, Interactive AI tools, Canvas mode, Google's canvas mode, Chat GPT canvas mode, Claude artifacts, Business analytics tools, Data privacy, Data security, Large language models, Business growth, Custom software, Data visualization, Interactive dashboard, Cross channel trends, Actionable growth plays, Lead ID software, Marketing analytics, Content amplification engine, Search optimization, Content pillars, Open source AI, Coding tools, Data analysis, Machine learning models, AI-driven insights, AI development, Open-ended instructions, Business use case, Operational efficiency, User interface, Automated processes, Growth dashboard, Data export.Send Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info)

This is the Everyday AI Show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business, and everyday life. Meet Firefly AI Assistant, now live and Adobe Firefly, the All In One Creative AI Studio. Just describe what you want to create and the assistant handles the rest, orchestrating multi-step workflows across Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome. The assistant accelerates execution. Did you know that without writing a single line of code,

you can create custom pieces of software, dashboards, business analytics tools that you can use to grow your company and your career. So not only that, but it's extremely easy. And the learning curve is actually so short. I'm going to show you how to do it today live on the Everyday AI show. All right. I'm excited for this one. I hope you are too.

What's going on, y'all? My name is Jordan Wilson. And welcome to Everyday AI. This is your daily live stream podcast and free dealer newsletter, helping everyday business leaders like you and me, not just learn AI, but how we can actually leverage it to grow our company and our career. So if that sounds like what you're trying to do, awesome. You're in the right place.

It starts here with the unedited, unscripted, sometimes longer than advertised podcast and live stream. But if you really want to take it to the next level and be the smartest person in AI at your company, that's where you go to our website. Your EverydayAI.com. There, you got to go sign up for the free daily newsletter. We're going to be recapping the highlights and the most important takeaways and some we could even get to from today's show, as well as keeping you in the loop with everything else you need to know happening in the world of AI. All right. So if you want the AI news, that's going to be in the newsletter as well.

So let's quickly talk about these no code interactive AI tools. So essentially, we're going to be looking at Google's Canvas mode, ChatGPT's Canvas mode, as well as Claude Artifacts. Okay, so I want to tell you this. I'm going to try to do my best for the podcast audience. I know not all of you go and check out the videos afterwards. This might be one of those where you might need to. But I'm going to try to do my best to describe because what we are ultimately going to be doing is taking data that I have. Right.

And always, right, always, always, always know what data you should be sharing and shouldn't. I'm a business owner. So I don't care what happens with my data. I'll give it to you all, right? It's, it's all, you know, data. I'm going to go through and tell you exactly what it is. But keep that in mind, right, depending on what type of plan that you're on. You know, if you're talking about chat, TBT, Gemini, Claude, co-pilot, et cetera, always understand your data.

But that's what we're going to be doing a lot of right now, is taking data and creating useful tools and dashboards that can help us really just grow our companies in our careers. So we're going to be looking at those different ones today. So we're going to start live, all right, and then we're going to backtrack. And I'm going to better explain kind of what these three different modes inside Gemini, Canvas, Chad GPT, Canvas, and Claude artifacts, what they all do, some of the pros and cons, and some of the differences, as well as hopefully get some live examples. But that's actually where we're going to start is with some live examples. All right.

So I have kind of some longer prompts ready. So you don't have to see me type. All right. But essentially what I'm doing, if you want to. watched what episode was that? Was that last weeks? This is our new, you know, putting AI to work on Wednesday. So if you go back and look at episode 549,

we talked about custom GBT's gems and projects. So in some of these cases, I'm actually using the projects, at least in chat, GBT, and Claude that we created last time that already have all of these files in there. So I don't have to really reproduce anything. But I do have 13 different files in this chat GPT project. So if you want to know more about how that works and the pros and cons differences,

go check out episode 549. So I am in this chat chviti project. There's some custom instructions in here as well as project files. All right. And these actually, let me just go ahead, put the prompts in because I want to do all this live. And we're going to do six total prompts. And let me first just go ahead, put them all in. There's a couple of things that I need to.

change here. And don't worry, I'm going to be explaining this as we wait, because some of these, and sorry, I got to update like one or two things in each of these. They're going to take a while because I'm using the most powerful models. I'm using these thinking models. All right. So I just sent the one inside chat GPT using 03. I'm going to go in Claude now in Claude projects. It also has access to all of my files. I'm going to put in this long prompt, and I'm going to go back later and explain exactly what's going on. I'm going to use Claude Obis 4. I wouldn't be surprised, y'all, even in the course of this live stream, if I hit my rate limits, hopefully not. But yeah, if you listen, you know how much I despise Claude's rate limits. You can't even

use the thing. So we'll see if we can actually get through it. And then I'm also going to put this same prompt in Google Gemini. So if you did watch that episode number, 549, you know that unfortunately you can't use Gemini's canvas mode inside its gems functionality. So unfortunately, I'm just going to have to put in this prompt. I'm going to change this here to say Gemini Canvas. And I'm going to have to add these files manually. That's okay. Just one extra step.

So bear with me, y'all. All right. So there we go. we have our first. Let me make sure I click Canvas mode here. Let me make sure I did it in the others. All right. Sure enough,

hey, this is why I love doing these things live. All right, I forgot in my first one here in Chatchip-T to click Canvas mode. That would probably help. It might say write or code. So, yeah, that would probably help to make sure that I'm using the correct mode. And if you're in Claude, you just need to tell it to use Claude artifacts. There's no, you know, UI button or anything like that. And then in Google Gemini, I need to click the Canvas mode.

All right. So we have our first prompt going. Let's do our second one. All right. And then I'm going to explain kind of, kind of the differences here. Nothing crazy. In one instance, actually, let me just tell you this. And I'll probably read a little bit of the specifics. In the first one that I'm doing, I'm telling it all the information. I'm saying, here's the data that I have. And I'm more giving it like open-ended instructions.

I'm saying, here's what I have. Here's what I'm looking for. Go make me something insanely useful, right? And make it work all inside here. And then in the second one that I'm running, I'm asking for something a little more specific. I am asking specifically for some sort of tool that can help me sort, prioritize different things for everyday AI. All right.

So let me go ahead. Let me copy this longer one. in, we're going to go to our second group here. All right. So I am in my chat GBT folder. I have Canvas mode enabled and I'm using the O3 mode. That is the best mode that uses Canvas inside chat GBT, the O3 Pro mode. It does not or cannot leverage that unfortunately.

All right. Let me just make sure I change the one thing in here. I think my copy and paste version said Claude. Artifax. So I just need to change that to chat GPT Canvas. There we go. And then I'm going to do the same thing over here inside Google Gemini. Google Gemini using the canvas mode. It uses the most powerful mode.

Gemini 2.5 Pro. All right. There we go. I'm going to have to add these manually my 10 different files inside Google Gemini to use it in canvas. And then last but not least, let's get test number two going inside Claude projects. And this one, let me do this one. All right.

I think we should be good. I think we should be good there. Let me switch this on to Opus. And we should be off and running. All right. Before I stop sharing this screen, I'm actually just going to first check in on all of them, make sure they're working because we're going to give them five. to 10 minutes to work and we're going to come back and check on them in a little bit.

So I just want to make sure they're all, they're all at least starting. Okay, it looks like for the most part they are. Okay. We got the hard part off, y'all. We're doing this live. So let me explain a little bit what's the best mode to explain this in. It might be clawed here. Let me zoom in and give you just a brief example of what I'm doing.

So the first one, all right, I am telling. each of these large language models to create an interactive dashboard that takes advantage of your project files, right? So in the project files case, I'm saying the files are in there. And in Google Gemini, I'm uploading the files. So as I'm doing this, this is a real business use case that I do all the time, right? People always ask me, well, Jordan, how are you using AI? Right?

And they're like, you must have a team of like 20 people doing this. And I'm like, no, there's like two and a half of us, right? And this is one of the reasons because I'm obviously using, you know, the latest and the greatest AI all day, right? From 5 a.m. till 10 p.m. I'm using things like Gemini Canvas, Chad, CBT, Canvas, Claude artifacts. I'm building myself little tools and processes just to make long, manual, tedious work much, much faster. All right. So it also, as I read this out, don't just think of what I'm doing. I want you to think of how you can use these three things, right?

You may not think that you need an interactive dashboard, but how many times think, how many times have you, or how much time have you spent, right? Maybe you're in a certain piece of software that your company uses, or maybe you're looking for a tool online that helps you do something and you spend, I don't know, two or three hours researching and then you're comparing. It looks like our, our Claude one is already finishing here. And you spend hours and then you're comparing different tools and then, oh, you go in there and you find out it's not working how you want it to, right? I can't.

I'd be embarrassed to say how many times that's happened to me. Live stream audience, do you find yourself doing that all the time? Right. And now all you really need to do, right, you can create. I don't know. do you need a simpler CRM with just certain things that you need for your sector? Maybe you don't need a whole, you know, Salesforce or pipe drive or whatever CRM people are using these days. You can just create a very simple one yourself just specifically for the areas that you need it, right?

These aren't straight up vibe coding tools. Let me just say that right now, right? For the most part, you can create a simple like one page app. You can create a very oversimplified piece of software with your data. So you're not going to be creating a full-blown, you know, CRM with pricing and login and all of those things that you would launch to the public. But you can literally create pieces of software that you would just use on your own pretty much instantly.

So a lot of these, you know, I have some examples here that are done. This is like a cooking show here, right? I already did this just in case because all of these. files and let me just go ahead. Let me go ahead and read this to you. All right. I'll probably just open a Google Docs here so we don't get too distracted with all of the work that's being done in the background. All right. So just like I did last week, I'm sharing 10 different files and it's hundreds of thousands of cells of data. Okay. So this is information from Google. analytics from the everyday AI website, right? So we have hundreds of pages on there,

thousands of data points inside Google search counsel. So how people are discovering everyday AI online. So when they're Googling something, you know, I don't know, chat, GPT free versus paid, you know, maybe one of our podcasts might pop up as an example. So that's the Google Search Council data. YouTube, we used to post a lot more on YouTube. Maybe I'll get that thing going back up. I don't know. Should we revive the AI in fives? I don't know if any of our live stream audience would ever listen to those, but let me know if I should keep doing those. For time consuming, I don't know, just say A on five. I always go back and look at the comments, y'all.

So Beehive, that's our email newsletter program. So it's every single email we've ever sent, which is, I don't know, like 700, but there's thousands of data points in there. Probably 100, no, tens of thousands at least. Buzz Sprout, that's our podcast. So it has all of our podcast information, downloads for each locations, all of that. And then Albuross, that's a piece of lead ID software we have on our website. So those are the different. We are uploading all these files and right.

Here's the thing. This, we have so much data for being a small little business, right? Because everything we do, I put out there on the internet. And that's another reason you need to do your best with data privacy, data security, all those things. When I'm working with clients, I'm not this lax with data, obviously. But when it comes to my data, it's like, I don't necessarily care. So I'm feeding it to large language models. You always need to say what you can or cannot feed to a large language model.

But think, any information that is publicly available, you probably don't even know. Like if you work at a big public company, all the financials are online, right? Your company's 10K, any, you know, financially like financial filing. So if you work at a big public company, chances are, there's so much information online. Or maybe if you're selling to huge companies, there's a good chance just about all of their information is public anyway. So keep that in mind.

Take advantage of all public data or data that you, you know, have the okay from who's ever in charge in your organization to, yes, go ahead and leverage this data. Obviously, if you have like a chat GPT enterprise license or a Claude Enterprise license, Gemini, same thing.

Right. there's different levels of data security and privacy, the further you go out, the further you go up. So essentially, I'm dumping all of my information into these two, or sorry, into these three different modes. Google, Google Gemini Canvas, Chad, GBT Canvas, and Claude Artifacts. There's different things they can do. They're not just strictly these, you know, these coding, kind of coding tools, right?

That's really what they are. even though all of these companies have their own kind of coding platforms, right? Google Gemini has a couple of them, actually. Claude has ClaudeCode. OpenAI has codex. There's actually one checking my watch here. I'll be able to talk about in about 15 minutes. So that's not, these aren't replacements, right?

So when you're talking about canvas and artifacts, these aren't like full-blown, you know, vibe coding platforms. They're not, right? But this is just such an easy, way to be able to write and render code. That's the other cool thing is it's not just going to spit out a bunch of code. It is going to create a working in many times a working program, a working dashboard that you can share with others and you can take advantage of right away.

So again, in this first prompt, uh, I'm saying essentially here's all of this data. You know, I want you to essentially, I've looked at all this data. I want you to unearth insights that I'm not seeing. I want you to connect the dots that may be misaligned or things that I'm not seeing. And I want you to just create something of immense business value that is going to help me grow the everyday AI show. And then the second prompt. So again, we're doing three different things, two different prompts. And I don't think they're all going to work right away, right?

But we're going to find out which ones work and which ones don't. Because also at the very end of these prompts, I am saying, you know, ensure your creation renders in this given environment. Right. So I'm saying, hey, the thing that matters most is like try to make something that works with one shot, right, that we don't have to debug. But the good thing is all of these systems, if they don't get it right, there's a lot of times a one click solution where it says like, you know, fix bugs or, you know, render, render code, right? You get a little different message. And sometimes you can just click once to fix it. Or if something's broken, you can just take a screenshot of it and say, hey, this isn't. working or this isn't rendering correctly and it'll just go through and update the code and then in the canvas or artifacts window it will go ahead and render it so then in the second one i am specifically telling it that i need a fully functional interactive tool uh versus in the first one i'm just giving it open ended instructions so i'm giving it in the second one i'm giving it five functional uh requirements so i wanted to see how the three of the three of

of them could take the same data, this same prompt. They all have this functionality, right? It is, but here's the thing. It's extremely difficult. I am asking for a ton of these three different modes and these different models. So I don't expect them all to work right away. And that is why I did spend a little bit of time before, right, in cooking show style, you know, saying like, all right, hey, if it doesn't work right away, I'm going to have one

that I've spent maybe two or three follow-up prompts, and they're done and ready to show. So before we get to those, let's take a quick break. Word from our sponsors. This podcast is supported by Google. Hey, everyone. David here, one of the product leads for Google Gemini.

Check out VO3, our state-of-the-art AI video generation model in the Gemini app, which lets you create high-quality, eight-second videos with native audio generation. Try it with a Google AI, pro plan or get the highest access with the ultra plan. Sign up at jemini.com to get started and show us what you create. All right. Let's get right into it. Let's look at results as they come in. All right. So we're going to start with the general prompt. Again, I uploaded all of this data. I gave a more general

prompt or, you know, without a specific output in mind. And I just said, yo, I can read spreadsheets. Don't just create a visual that shows me this data in the spreadsheets. I want something that helps me grow this thing that helps me spot blind, you know, helps me find blind spots, you know, find trends, et cetera. All right. So inside my chat chpte project, I, again, I use the canvas mode. So in chat GPT and Google, it's a button you click, artifacts inside Claude, you just sometimes need to tell it, use Claude artifacts.

So you'll see on the look, left hand side, all right, podcast audience. On the left hand side is my extremely long prompt. And then I get a response from Chad ChbT, but then I get my, my canvas on the right hand side. All right. So right here, all it looks like is a bunch of code. It's about 296 lines, which isn't a ton. All right. And I'm using the O3 model. So let's just see. Let's just see what, what we get here. All right. So I'm going to click preview your output. So that's in the upper right hand corner. All right. I'm going to try to zoom out a little bit and maybe resize my windows so we can get a little more. All right. So the first one, okay, this one's interesting. It didn't really do what I wanted it to.

So it's telling me to upload fresh exports from your six data pillars below. The dashboard mines cross-channel trends and surfaces, in surfaces actionable growth plays automatically. So first, let me see if it actually works. Okay, so when I click that, it's not showing on my screen, but it's happening. It is allowing me to upload something. So let me just try this. So let's see. It looks like, okay, it looks like this one here in the upper right hand corner is for uploading YouTube data.

So let me just go ahead and upload that. And let's see if it does anything. Okay, so instantly when I uploaded something, it put in like, my gosh. okay like a hundred little boxes immediately and it's just all these things retention fix retention fix right so i don't see any data that it filled in though so in this case didn't do a great job i'm going to click this charts button nothing there so all right first one that we gave to chat chp t not great but again keep in my y'all generative i i is generative it is different literally each and every time.

All right, because the first time I did this inside chat, GPT, I actually got a much better and fairly usable output. All right. So let me, I love just having to jump around to 20 different screens here. I do just want to show you this one because the first one, it was fairly, it was fairly clean.

Let me do this. Okay. I got to stop the screen shift. start another one. This eight, love doing these things live. You know what I'm actually going to do? I'm going to just share my entire desktop. So then that way I can more quickly bounce between these things. So apologies in advance because you might see a lot of a lot of things happening here on the screen. So let's go ahead and do that. We're going to share my entire screen. So for our live stream audience, the first.

time that I did this more general prompt inside chat GPT canvas, I got some pretty good results, right? So at the top, it created this growth dashboard. It has a little very nice, minimalistic, lots of white space. So it has these things, these little icons with sessions, podcast plays over the last seven days, although that is that number is not correct. It has the email open rate. And then, you know, a certain email that. that performed well. It has a site to podcast correlation, which I'm not even sure what that is. And then I have this kind of nice, nice little bar chart or line graph underneath. So there's different tabs I can click. So it says Google Analytics and I can kind of scroll,

Google Search Council. It gives me a little list, right? So I can click on these different things. So pretty good. However, all this really did. So I would say in the first instance, chat GPT failed because it didn't, it just visualized data, which is what I really didn't want, right? I didn't really have a need for that. And then the one I just did live for you all didn't really work. All right. So let's look at Claude. All right.

Ah, you got to love it, y'all. I did this live. Look at this. I already hit my Macs. It looks like I, oh, this is fun. Love doing things live. You guys saw this. literally first time it died right it couldn't i paid $20 a month i couldn't do one prompt i couldn't do

one prompt got to love it let's see if it actually finished this though uh but i did last night build this because i'm like yeah this is going to happen um and sure enough it did all right so let's look at what uh claude did uh last night and i think what i had to do in this is i had to use i had to use Sonnet 4 instead of Opus 4. Because apparently, even though I paid $20 a month, I can't run a single prompt in Cloud 4 Opus when I'm uploading all of these documents, which, yeah, you got to fix that anthropic. No one's taking you seriously. Like enterprises aren't.

Yeah, if you're a developer, if you're coding, if you're using, you know, curse around the back end, sure. But people who are using front-end large language models as their business as their AI operating system, people aren't using Cod for that exact reason. I hear from people all the time. They're like, you can't even use the thing. But I did use Sonnet, which is the less powerful model. So I used Sonnet for this is from last night. And it actually created a pretty nice looking at dashboard. So for our podcast audience, similarly, we have these little tabs here. And it's super slick, right? It's super slick. You know, it has this kind of dark, uh, gradie.

I have an AI intelligence tab. I have a B2B prospects tab. I have a content performance and audience analytics across platform in a growth lab, which sounds pretty sweet. All right. So I can't imagine like if Opus, if I could actually run this once with Opus, presumably it would be even a little better. And then there's also some overall stats at the top,

some B2B prospects, total downloads, website views, and AI insights. All right. So there's some filters. So I'm curious if these things are actually going to work. All right. So it says some AI powered strategic insights. So, okay, this says enterprise revenue opportunity, content performance leader. Let's see.

There's some little things. It says take action. So I click on it. Nothing happens. Okay. So, okay, some helpful insights, but not that I did know these things. So it says tutorial dominates with 40. thousand downloads across six episodes.

International market penetration. It says the UK market shows 173,000 downloads across 1339 cities for a that doesn't seem right, 197 percent market share. So some of these stats, not sure. It looks like some of these things are clickable. When I click, take action, nothing happens. But it looks fairly slick, right? So that's nice.

Let me go back up, see if these things work. So I'm going to click seven days. Some of the stats at the top change, but the insights don't click all time. So it looks like just that changes the stats. There's a filter for industries. So maybe this is it categorized the podcast episodes by industries, maybe. I'm not sure. I'm clicking law practice.

Okay, this is actually, it looks like a lot of the B2B lead ID information. So this is the type of companies that have visited our website at your everyday AI.com. All right, B2B prospects, same thing. So I can sort that by all industries. I did a nice little dashboard. So, okay, this tells me. Okay, this is something I didn't know. So it looks like the most companies that visit our website are higher education universities.

There's been 346, looks like universities that have visited our website and average. about 4.7 views. So that's helpful, right? That's something, okay. And then I can sort by country as well. I can click US and that adjust. It says I can search companies. Let's see if that actually works.

I'm just going to put in, oh, no, okay, so I type in one letter and then it stops. And I have to re-click. So the search functionality doesn't work. The drop downs do. It's pretty, right? Overall, pretty cool. One thing that I'm interested in is cross-plocel. So I'm not even sure what this is.

It says cross platform content correlation matrix sounds really cool, but what is it? So it says strongest cross platform performer. So this is like B2B page views, podcast downloads, and episode counts. So it's not super clear what this even pertains to. I'm sure if I spent a little bit more time, I could get a little bit more value out of this. But let me go to the growth lab. All right. So in growth lab, this is what I wanted to know.

So it looks like it's giving me some ideas. So it says content amplification engine and then it has a medium ROI. It says AI agent content performs three times better than average scale winning teams or themes. All right. So it says, you know, the performance looks like average performance for these are 4,378. I'm guessing podcast downloads. I'm not sure. So it says create AI agents master class series.

which is funny. I literally manually planned this like six months ago. I already have it planned. I got to update it now. So all right, pretty pretty Jim, a pretty good job there, Claude. Launched YouTube channel with video versions. We already have that. Develop interactive web experiences. Create downloadable resources. So nothing earth shattering, right? But if I'm ever just stuck, right? And I'm like, all right, hey, how do I grow this channel? Do I want to open a spreadsheet with hundreds, literally hundreds of thousands of cells in open like 10 of them in bounce between. Or would I rather look at a dashboard like this? I'd rather look at a dashboard like this, obviously.

Right. So overall, overall pretty pretty impressive. Right. And I could obviously make this better with a little more iteration. So unfortunately, first try live. Chat DBT kind of failed. Uh, hit our, hit our token limit. Uh,

in Claude. Although when I did refresh it, oh, I lie here. When I refreshed it, it actually showed up. But funny enough, it made something very similar to the last time.

But when I'm clicking around in it, it looks like there were some errors. All right. So if I can reload it, all right. I'm not going to give it any more shine. But overall, overall,

all pretty okay for one shot from Claude, but it looks like I just hit my message limit and things crashed and burned. All right. Let's see if Google inside Google Gemini gave us something we could use in one shot. Bam, we did. All right, here we go. So Google inside Gemini using the canvas mode. I got this, your AI growth consultant. Okay, so it looks like a simple little web app, which I like. On the left, there's some little tabs that I can click. So it says growth insights, topic performance. Oh, this is nice.

This is slick. Audience and acquisition and B2B lead center. All right. This looks nice. Okay, there's like some actions here, which I'm curious. One thing I wish in Google Gemini in their canvas mode, I wish I could move the window a little. bit more. So I just kind of have to zoom out and I'm old and I had bad eyes. So I can't see.

So I'm even curious. This like B2B lead center. So it looks like these are, I don't know, just companies that have been to my website a lot. But it has this little outreach button. And I'm wondering if it doesn't eat. So I'm going to click it. All right. It didn't do anything. So I'm not really sure what that outreach button is. But again, this is one shot. I would probably have a conversation. So what I would do is I would go through and I would separately write down what I like, what I don't like, how I wish it could. work better. And I would send that in a prompt back to, you know, Google Gemini or Claude or chat GPT. And there's a good chance that after maybe two or three messages, you could literally

get something that looks like this, but is fully functioning. So at least the Google Gemini canvas version off the bat, it looks better. For the most part, things look like they're working. So I'm looking at the main page here. All right. So your AI growth consultant. So this is good, because this is the only one that didn't just like essentially take my data and display it, which is what I explicitly told it not to do. So as an example, it says cross platform winner. So I have these little these little kind of hoverable like icons it looks like. So it says the topic AI tools is your strongest content pillar.

So it looks like it took all my content, created content pillars and then looked at performance. And then it says driving significant traffic. across YouTube and your website. It's a proven winner with your audience. And then it gave me a little suggestion. So very nice, very slick, again, dark mode with this, you know, bright colors. And then it said double down on this topic, create a deep dive series, a dedicated podcast episode, or a premium guide to capture this high intent.

Let me look at one more thing here. It says hidden SEO opportunity. You are highly visible for the search term AI for marketing. All right. Didn't know that, but I could have found that out pretty quickly. But your click-through rate is low at only 1.7%. So it's telling me to optimize this page, title and meta description for this term, try including numbers, questions, or stronger value props to entice clicks. For instance, it's telling me to create content.

That's 10 AI marketing tools that generate ROI in 30 days. It also looks like there's something that says explore the topic. Let's see if this one does anything when I click it. it. Hmm. It actually did something. Oh, interesting. Let me see if this works again.

Okay, so let me go to that same hidden SEO opportunity. Explore topic. Okay, explore topic general. Very sweet. Okay. And then it takes me to another portion, topic performance deep dive. Okay, this is very impressive. There's a couple like UI, UX things I might change.

But this, every. everything in here except for this B2B lead contact center outreach button. Everything else, it seems like it's working, right? So I'm in here in this top performance. I can filter by topic. I can go to AI tools. It brings up everything with performance metrics on the number of views. So it looks like these are, okay, it looks like these are suggesting different pieces of content and it's projecting the performance metrics based on, uh, my current metrics. So that's really, really cool. Uh, so that is the, uh, topic performance. There's the audience and acquisition. So this is the top acquisition terms showing me how many

clicks that I get for different things. All right, as well as an interactive chart. So it says your podcast audience is showing strong, accelerating growth month over month podcast audience. Thank you for sticking in, even though some of these things are a little harder to explain, but we always do get the video. So, you know, check, check. Check. the show notes on our website, your everyday AI.com, this same exact podcast. If you want to see the video version, it's on our website. Like I said, the B2B contact center, a couple things weren't working. But overall, the one shot general ran out of messages on one shot on Claude. Chad GPT didn't work very well, although the one I did yesterday worked a little better. And then we had, Gemini did a

Fantastic job. Fantastic job. All right. So let's move on to our second version. All right. Actually, no. You know what? I'm looking at my clock.

I can break a little bit of news that is kind of related since I just talked about Google and Google Gemini. Google is as of this very minute embargo lifted. So I can talk about it. They just launched their version of, you know, if you've used Claude Code from Anthropic, if you've used Open AIs codecs. So now Google has their own free open source AI agent that is a code, a CLI, right? So super, super exciting and impressive. So we'll have more info on that in our newsletter.

But hey, a little treat there for our live stream. and so you are kind of going to be the first to hear because we can just talk about it like literally as of like three minutes ago. So yeah, make sure to check that out in the newsletter. Essentially, their version of a codeline interface AI coding tool looks pretty impressive. It's free and it's open source. All right, let's get to the rest here. We're going to try to go quickly. Our second prompt, let's go ahead and look at the results.

See how they did. All right. So again, this one, I gave it more special. specific. I gave it five, and I'll just kind of read those, five functional requirements. I said it needs to run fully in the chat window on the first attempt. It needs to produce actionable output. As an example, ranked recommendations. Three, it needs to offer interactive controls, such as a slider, drop down, or search. It needs to surface non-obvious patterns, and it needs to return exportable results. So the first one, I was open-ended, and I said, hey, who can render something of

immense business value right here in the, you know, canvas or artifacts window. All right. And the second one, I gave a little bit more details. So, uh, let's go ahead and click the preview button here for our chat GPT canvas mode. So it created a growth opportunity radar. So, uh, it did work for a shot. Again, I will say in my experience, which is a pretty decent amount. I was actually surprised that the first one in Chad ChbT didn't work because normally I think what happens

is chat ChbT's version in Canvas, whatever it's creating from a business dashboard little tool. It's usually not as impressive, if I'm being honest, but it usually works more times than it doesn't, which is why I was surprised on the first iteration. It just didn't really work because usually it's more consistently working, whereas sometimes Claude and Claude artifacts and Gemini canvas don't always work on the first one and you have to use them a little bit more. All right. So Chad GBT and the second attempt spit out this growth opportunity radar. So there's a drop down with different strategies. There's something here with minimum, search volume, search weight, YouTube weight, podcast weight, and Google Analytics wait.

So I'm not really sure what this is. And then we have a little table down here that has different topics, scores, search volume, clicks, YouTube views, podcast downloads, and Google Analytics views. So it looks like it took common themes or threads that showed up in all those different areas. And it kind of gave them scores. So I'm not really sure how I would use this. So it says drag the sliders to reweight channels. Gap highlights unmet demand, double down, boost proven winners. So, okay, fill the gap. That's one of the things. And then double down. All right. So let's see. Search wait, YouTube wait, podcast wait, Google Analytics weight. So, okay, so it looks like I'm saying, okay, if I have something that I really

want to more focus on Google Analytics, I could crank this slider up. And I'm not as concerned about it ranking on YouTube. I could crank it down or search crank it down. Okay. So everything works. I'm not really sure what business value this provides because it's a little unclear because the topic is just one word. Okay, so do I just talk about chat GPT more? So leaves a little to be desired, although it is working at least according to the five functionalities that I said. Again, I would follow up with this one multiple times, but I wanted to give you all the one shot. All right. Let's go into Claude.

So even though I hit my usage limit, it looks like it at least rendered this. So it looks like we have a B2B content intelligence engine from Claude. Discover which content topics attract high value enterprise visitors. All right. So if I look down, I see a couple categories and segments. So pretty good. It gave me five. It looks like my top five are LLM.

implementation, AI agents, content automation, AI ethics and bias, and prompt engineering. All right, pretty good. I probably could have figured that out with myself, just dropping some spreadsheets and a large language model and asking it to categorize different topics. But this did a pretty good job here, and it gave me a momentum score, estimated B2B visitors, right? So it actually did a good job of looking at my AlbuCross data, seeing what companies visited, what pages. It sorted those. And then it gave me a return on investment score. So as an example,

I don't know where it came up with this arbitrary number. I'm sure I could read and it would tell me. I'm not going to take time to read because all of them did also give me a prompt or a response back. And it said like, hey, here's what I built. But it's kind of interesting. So I see, right, some companies from Oracle, SAP Service Now went to these different websites on our or these different pages on our website. And it's telling me to do this as a podcast series, LLM implementation, what every $1 billion plus company must know in 2025. So pretty good.

It looks like some key insights out here. All right. Nothing crazy. Let's see if the export button works. All right. The export button doesn't work. I forgot to check that in the last one. Let's see.

I don't see an export button in the, chat GPT one. And then there's also some filters in this one from Claude in the B2B content intelligence engine. So I can go to technology. Let's see if it brings up fresh insights. Looks like it doesn't. It just brought up the AI agents one.

And then I can also sort by company size. So if I want to do something for medium size enterprises, it looks like AI agents, content automation. So three of the five original ones, it said, are good for those medium-sized companies. So overall, pretty good, pretty good, right? I'd say Claude, at least in the second one, did a much better job than chat chbt. All right, let's look at Google Gemini. All right.

So first, not as good looking as normal. All right. So it says B2B growth engine. Unlock content opportunities from high value website visitors. All right. It looks like I have some hover sections. These look like just the files I already uploaded. So when I click them, it just brings me to a screen to download more.

So that's interesting. Let me just go ahead and reupload this file. All right. I'm not sure what this is going to do. So it says upload the require CSV files to begin the analysis. The engine will process the data entirely in your browser. I'm not sure why I would re-upload this. All right.

So so far, we didn't really hit. the mark, but I'm trying to reupload these pages. It looks like it's having me, uh, re upload them and match them. And I'm not exactly sure why, but I'm going to try it. I'm going to try it before we, uh, write off this iteration. All right. Um, all right. I uploaded four or five. I'm trying to upload the fifth in for whatever reason. It does it look like it's working. Um, so unfortunately, uh, this version didn't work. And I'm also getting an error message. It looks like I've way too many LLMs open or something like that.

Maybe if I shut down some of these other ones. It looks like it looks like maybe this this Gemini window. I had too many things open. I kind of want to see now what happens. If I upload all of these, if I am going to get something amazing, or if it's just something in my, in my prompt, in my instructions that should have been a little more clear, although the other systems seemingly did a better job. So yeah, looks like it's just allowing me to upload four and then it's kind of crashing.

I will quickly show, though, since we didn't get one from the Gemini and the second one, let's see the one that I did yesterday. Yeah, still didn't work very well. The Gemini second version. Oh, yeah. So this is what I said before, right? Sometimes you'll just get something that says like, hey, something went wrong. There was a problem when running your code.

And I can click that fix error button and it'll go back and hopefully fix the errors. Right. So let's wrap this up. I actually had a whole, a whole stinking thing here. You know, what's best for what. You know, and we didn't even get into all of the aspects and power of Gemini's canvas. Chad GPDs canvas and clawed artifacts. This is just one of the things that they can do. And again, did I, did I were any of these good enough that I'm like, my gosh, I'm using them

right away with one prompt? Probably not, right? I did really like the first version here from Google Gemini. I thought this was really good. It gave me some really good insights. And then I did like the version, which one that was the, oh, I just had to close it. I liked the version that Claude did on the second, the second one as well. So some of these, I think were good enough that I'm actually going to go back and spend time maybe once a week, going back and looking at them. And these were just essentially one shot, right, which is usually not a good idea to do. You always want to be able to work with them and make them better. So I'll share this if you want to, just let me know.

I have a little actually used Google Gemini canvas mode to put together all of my notes to a sleek little overview here. But let's just wrap the show by going over this. All right. The results, even though there's some bugs, that's why I do these things live. With a little more work, they're going to be really good, right? As soon as I end this live stream, this podcast, I'm probably. I'm going to go back in and tweak these a little bit because I see a ton of potential here.

I see a lot of different opportunities, both that they found and some opportunities from what I see, uh, what these large language models created on the first try that I'm like, hey, if I tweak this just a little, this is going to be something that I could use daily, right? Uh, that's the thing. I use so many pieces of software and there is so much power in these no code interactive AI tools. if you are just creative, if you're patient enough, and if you work with them a little, it is literally like having a no code software development coding partner sitting by you,

right, and in helping you tweak and make things better as you go. So the next time you're stuck and you're like, man, I can't get this figured out in this piece of software or how will I ever? There's this task I'm having to do over and over and it's so manual. and it feels antiquated, jump in, have a conversation with chat GPT canvas, with Gemini canvas, with clawed artifacts, and see what you could build.

Because even if you are a non-technical person, there is a great chance that you can just go in, chat with one of these moves. It renders it in the screen, and you can even then share that with yourself or with members, right? So you don't even have to use it inside chat chbtee. You can get a public link and use that yourself

or share it with team members as well. So this is not just creating something for yourself, although I think that's the way to get started. These aren't fully software engineering platforms. It's for quick, simple little web apps, little business dashboards that can take some of the mundane processes out of doing your day-to-day resource-heavy, analytics-heavy tasks. So I hope this,

A.I. Putting AI to work at Wednesday was helpful. If it was, please let me know. Let me know what you want to see next. I work for you. Also, please go to your everyday AI.com. Sign up for the free daily newsletter. Do me a favor. If you're still listening to this, go read today's newsletter. Right. So if you're on the live stream, it'll be out in a couple hours. Reply to today's newsletter. Tell me what you want to see next week for our putting AI to work on Wednesday. So thank you for tuning. And I hope to see you back tomorrow and every day for more everyday AI. Thanks y'all. Meet Firefly AI Assistant. Now live in Adobe Firefly, the Allman One Creative AI Studio. Just describe what you want to create in your own words and the assistant handles the rest, orchestrating multi-step workflows across Adobe Creative Cloud apps,

including Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome while the assistant accelerates execution. Stand control with the ability to step in and run. refine at any time. See it today at firefly.adobie.com. And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going. For a little more AI magic, visit Your EverydayaI.com and sign up to our daily newsletter so you don't get left behind. Go break some barriers and we'll see you next time.