Summarify.net

New ChatGPT Model Just Made This AI SEO Method BETTER

h86I1NfUCoM — Published on YouTube channel Income stream surfers on September 16, 2024, 5:28 PM

Watch Video

Summary

This summary is generated by AI and may contain inaccuracies.

Here is a brief summary of the key points from the transcript: The transcript discusses using GPT-3 to automatically generate data visualizations for blog posts. The speaker takes an existing blog post and feeds the text into Claude and ChatGPT to generate data visualizations. ChatGPT produces a better infographic than Claude, with real-looking data rather than made up data. The speaker then takes the ChatGPT-generated infographic and easily adds it into an existing blog post on Shopify to improve the content. Key points: - Using GPT-3 models like Claude and ChatGPT to automatically generate data visualizations for blog posts - ChatGPT produces higher quality infographics than Claude with real-looking data - Easily take GPT-3 generated infographics and add them into existing blog posts to improve content - This approach allows quick and easy content upgrades without much effort - Having access to powerful models like ChatGPT is beneficial but usage is currently limited - Services like Harbor provide automated content generation like this as a service

Video Description

We’re using gpt o1 preview today to re optimise old blog posts using a method I’ve been talking about for a while with automated graphs and automated data visualisations

In this video we’re going to take a deep dive into how I reoptomize blogs and add new data visualisations to them using AI to speed things up

I was interested in how gpt o1 preview would perform with this new method and was surprised to see how good the results were

Thanks for watching

Hamish

#openai #gpt #o1 #seo #aiseo

Try our SEO tool: https://harborseo.ai/
Work with us: https://incomestreamsurfers.com/

Transcription

This video transcription is generated by AI and may contain inaccuracies.

There are a lot of videos on GBT zero one preview right now, but I wanted to make a video with a very real use case for this amazing new model which has the ability to think you're going to want to watch this video. If you want to add data visualizations to blog posts that already exist, or if you want a quick and easy way to rank better on Google using the new GPT model. Let's just get straight into this video. So on my screen here we have dapafam. Dapafam is one of our clients. I'm gonna show you how we can improve their content very quickly and very easily. So the first thing you wanna do is you wanna drag down to the bottom of the article like this, so that you've highlighted everything. Then you wanna go on Google and type paste to markdown and then just Ctrl V to paste and then control a, control C. This will give you all of the content, all of the links, everything that you need. Okay, now what we're going to do is I'm going to do Claude and also chat GPT at the same time. So I'll put this in Claude and then I will also put this in chat GPT. And then the prompt is just very simple. It's make a beautiful, modern, interesting data visualization using only HTML and style tags. This is the article. Do not change the article overall, only make a data visualization to add to the article. Don't give unnecessary HTML tags as this is going on. Shopify. So we'll send that to Claude 3.5 and we'll also send this to chat GPT zero one preview. As you can see at the top, it is literally called chat GPT zero one preview. Even though there's a bit of confusion with the name, some people call it zero one preview. Some people call it GPT zero one preview. Okay, so this is what we got from Claude. This is, you know, this is fine, but this is actually not really based on anything. This isn't real. There's no evidence that 92% or 9.2 comfort ratings. It just, it doesn't really make sense as a data visualization. So we'll see what chat GPT zero one preview came up with instead. So we'll go to w three schools, try its editor control a, control V, and then run. This seems to have made up images, which is kind of annoying. So I'll just say, don't make up images. And then I'll say to Claude, don't make up data. Just to show you I ran this before we started recording this, the exact same prompt. And this is what GPT zero one preview came out with. So you can see this is a really nice and also mobile friendly infographic that could fit into basically any article. So this is a really. This is exactly what we were looking for. This kind of stuff is amazing for articles. Just so you know, you can actually just take a picture. Like take a picture of this. If you're on WordPress, for example, it might actually just be easier to press print screen and then take a screenshot and then control v onto WordPress. That's probably the better way to do it. So we'll see here what this says. Again, this is just. It's gone back to the same thing. It's pretty cool that it's animated, but yeah, it's really not the best. It seems to be obsessed with this comfort thing and with making up data, which is not what I wanted at all. I didn't want made updates. So one of the biggest issues with GPT zero one preview right now is, is the lack of usage that you're allowed to do. If you don't know. You can use the playground as well if you have access to this model, which not everybody does, but we do. We actually have two tier five development accounts. We've probably spent $20,000 on OpenAI in the last year or maybe a year and a half. So we'll see what this came out with. Let's just see if it's made up any images. It doesn't look like it's made up any images. So we'll run here. This is fine, perfectly good. This is a nice 12345. I wouldn't call this an infographic, but like a step by step maybe it's mobile friendly again. So one cool thing about GPTO preview, it does seem to give a shit. You can see here about mobile friendliness, which is not something that Claude does often. So let's just see if this is mobile friendly. We'll press copy here. Same thing. Run. If I do this, what happens? Oh, it is mobile friendly as well. So they are both mobile friendly in this case, which is great. So what we can do here is, let's say we liked this one here, this is perfectly good. But I actually prefer this one here. This is much better. So what we can do is we can easily put this into an article. So we'll go to shopify. It's going to be on a different shopify account. It's not going to be on Dapfam just because I don't want to log into dapper fam right now. So let's say we wanted to add this to one of our articles. So we'll click online store, click blog post, let's find an article about shoes. There we go. So this is bespoke kit on sneakers, blah, blah blah. So we'll click in here, we'll click the HTML thing and let's say we want it. Let's just put it at the end of the article because it's probably easiest. So we'll just paste that there and then press show. And then if we scroll all the way down, we should see our beautiful infographic right here. Let me just see if this mentions. Okay, no, this is actually fine. We can save this. I've just added that to this article just because it actually fits with this article. And then let's just scroll down and there we go. Beautiful infographic added to the article without a lot of effort whatsoever. Now, harbor does do this automatically. If you're interested, you can just go to harbor SEO AI. We're improving things every single day. If you want this to be built into your articles automatically, then use harbour. If not, then use the methodology that I have just shown you. Two men is at an all time high of 847 clicks. It's never been above 847. We're almost at a thousand clicks a day. And the most recent date, you know, 755 ish clicks. So two men's going extremely well. We're going to keep posting to two men, showing it as a case study. Thank you so much for watching, guys. I hope that this was very, very helpful. This is a huge upgrade to the data visualization thing because it has the ability to reason before it creates it, which means it creates much better things out of the box without you doing a lot of complicated prompt engineering. Thanks for watching. If you're watching all the way to the end. Thank you. You're an absolute legend and I'll see you very, very soon with some more content. Peace out.