Transcription
Imagine logging into Chat GPT and having your very own team of SEO experts telling you exactly what you need to do to rank number one, without having to log into another SEO tool like Ahrefs, Semrush, Keyword Tool, or even something like Jasper AI. You can do all of that now within Chat GPT.
In this video, I'm going to show you how to connect real-time SEO data with Chat GPT, which makes it an incredibly powerful SEO tool. I'm also going to give you some incredibly useful use cases that you can start using today to improve your SEO with this Chat GPT connection.
And I've got a quick question before we get started. Have you ever had a bunch of SEO data right in front of you and thought, "Great, what do I do now?" You have the click data, the appearances, and all the errors that you need to fix, but you don't even know where to start. It's a bit like having an IKEA flat pack without the instructions. Sure, it's all technically there, but it's kind of difficult to even know where to get started. Well, this is the solution for that problem. This connection with Chat GPT is going to allow it to do all the search research for you and also tell you how to fix stuff in a very understandable way.
Let's get started right away. You need a couple of things to do this. One, you need an account with Data for SEO. They're going to be the API or the data providers for this. You can sign up and create a free account, and they'll give you a couple of dollars free credit to use it, which goes a surprisingly long way when you're using it in this matter. It's also a pay-as-you-go kind of subscription, meaning if you don't use it, you don't pay for it, which is two thumbs up in my opinion.
You need a link to this GitHub repository. I'm going to leave this below, don't worry. This is going to give us access to all the different API calls that we want to connect the GPT with. And we also need a base64 encoder. This is going to encode and protect your login details, which you need to give the GPT, allowing you to share this GPT with anyone else and you don't have to worry about your login credentials.
So the first thing we need to do is make sure we go to the GPTs section, and we're going to create a custom GPT because it is with the custom actions that we access the Data for SEO data. And if this gets you a little bit annoyed because you thought you could do it directly from your chat in Chat GPT, don't worry. You will be able to do that. I'm going to show you a trick how to call all of these custom GPTs into the one conversation.
You need to name your GPT. I'm going to name it "Data for SEO example YouTube". And what I want to do is go to the code, the capabilities, and I want to select "code interpreter" and "data analysis" on, because we're going to be doing a lot of data analysis. And in actions, we want to click this "create new actions" button. This is where the magic begins to happen.
You want to go to the GitHub link that I've provided below. And these little folders tell us each different API call we can make to Data for SEO. Now, we can't place all of them within one GPT. So, we need to create custom GPTs that are experts in keyword analysis, for example, another one in Google My Business, another one in backlink data, domain analytics, you get the idea. But the more versatile out of all the calls is this one here, the Data for SEO labs. It has so many calls that with this one piece of code, you can practically do an entire SEO strategy.
So, we're going to copy this raw file. We're going to go to the custom GPT that we're building, and you're going to paste that code there. You're going to give it a minute. And now you're going to see that all of a sudden you have all of these available actions. These are all the API calls or the data calls we can now make within Chat GPT. This gives it incredible power.
Now, we're not done yet. We need to go to the authentication settings and we need to go to the API key. This is where we need to encode our login details and our password. So, if you've made an account with Data for SEO, there are two things we need here. We need to go to API access and you need your login details and your password. And now you're going to place your login details and your password like this. So your login details first, or your email. Your email, a colon, and then the password, and then we're going to click "encode to base64", and it's going to give you this little URL or this little code string of codes.
Now I want to make this very clear because I see a lot of people having trouble with this. So let's go through this again. So, this isn't actually real, but this is how you need to encode your credentials. You're going to do your email. So, I'm going to say example@gmail.com: like that. And then your password. And your password is going to look something like that. That is what you need to place in here in the base text. And then click "encode base64". And it's going to give you this an output. Yours will be different. That output is what we need to copy and paste into our custom GPT as an API key that is hidden. And it's hidden because we've encrypted it essentially. And now we need to save. You need to give this a while to load a little bit.
Now that it's loaded, we're going to test it to see if it's working. So let's go, for example, "keyword ideas". We're going to test one of these, and it's going to talk to the connector. We're going to confirm. Give it a while. And if we've done everything correct, it's going to give us some data. Now, it's just chosen a random keyword to test here, but already we see the incredible applications. So, we've got, for example, "kids electric bike" in the United States. It's found out the monthly searches for this. Already, this is so powerful.
So, now that it's working, all you need to do is go back to actions and we can add a set of instructions if you want. You can leave it empty. In our community, we've got a pretty thorough prompt that will turn this into a competitor analysis or a competitor analyzer, I should say, rather. You don't have to use this, but it makes things a lot better and a lot more efficient. Um, I'll leave a link to the community where you can get this here.
Now, with that, I'm going to test it out now. And let's pick a website here. You probably seen us use this one here. So, Financial Advisor Pro, I'm still testing this GPT. And you can see that I can even select the model. If I want to use 4.1, I can. Another smarter model. This becomes now even a more incredible tool. And I can say, well, "um, what keywords is this site ranking for?" And you'll see that it's failed a bunch of times, but that's okay. It kind of kept trying until it got it right. And now I've got all these keywords that that website is ranking for, the positions, the estimated monthly search volume that that keyword is getting, and the cost per click. Cool.
Now, this is working. I can create this. Now, if you're just sharing this amongst your team, I would say "available to anyone with a link". I'm just going to go "only to me" cuz I'm the only one that's going to use this. And now, I want to show you a couple of important and incredible use cases. So, I'm going to view this GPT and cool. I'm going to test our brand new custom GPT out. I've changed the model for that. You can do that here. 4.1. I haven't tested that model before in here. I'm going to tell it the competitors for the website that we are trying to help. So, I'm just going to say here "our site" and then these are the competitors. We know that it will be working when it's talking to the connector. We're going to have to confirm. Cool. And it's already starting to give us an initial domain performance comparison. Wow. Okay. So, estimated search traffic Financial Advisor Pro 142. Everything else is in the thousands. Total ranking keywords. Top keywords ranking in the top 10 03. Perfect. So, it's already giving us a detailed insight into why they are ranking us. And it's even giving us some follow-on questions that we might be able to ask it to continue this strategy. So already competitor analysis-wise, we can see what we want to do.
So we've done a bit of a competitor analysis. Let's do a detailed keyword gap report where we know where the competitors rank and we don't. And then what we're going to ask it is for a keyword research report telling us what other keywords we should rank for. And they're getting smarter and smarter. You can see that it has an error cuz it got too much data in one point. But it found a workaround. Chat GPT is getting a little bit smarter these days. And it might be because I'm using 4.1 instead of 4 40. Highest valid keywords competitor outrank us on. Wow. Okay. So, I can already start seeing that the competitors probably have a bunch of specialized local services areas. So, they have a bunch of pages on retirement planning in Tuscan, in Kansas, in Mesa, in other sections of Arizona. It's given me the observations. What does this mean for you? The next step should be wow. Okay. "Would you like a priority table with recommendations and actions for the highest potential keyword plus on-page specific?" Now, let's go. "Tell me what keywords I might or I should rank for."
So now let's do a little bit of keyword research and see if it can give us a bit of a strategy here. Okay, perfect. Now it's given me some unique opportunities that would take me a lot of time to find because I need to understand the business and then the potential keywords. For example, not all of these are going to be good, but a lot of them are not bad. So "assets exempted from wealth tax," "wealth threshold," that's a good one, "wealth tax unpopular," "minimum net wealth tax," "lexenberg beneficiary strategies," "third party administrator to fiduciary responsibilities," "tax implications of trust distribution," that's a really good one I know that he provides those services, so already I'm getting quite a good ideas of keywords, fantastic. So this is all well and good, kind of three strategies that we just did; we did we did competitor analysis, we did content gap analysis and we did keyword research pretty easily.
But what happens if you're having a conversation with normal GPT? Let's go to for example, and let's ask it for example to take a look at the site. Now I'm not in the custom GPT and say, "give me some blog ideas for this website." Give me some blog ideas for this site. It's not calling the GPT now, but it's just giving me some ideas here. And what and the beautiful thing about this now is that I can start saying, well, "ah, I wish I had my custom GPT". And I can do that. I can go at and it's going to and it's going to search for all of the recent GPTs that I've been using. I can now pull that through. I can go "Data for SEO YouTube example". That's the one that we were just using. And I can now say, "okay, for the first six blog posts, can you tell me which keywords they'll be concentrating on as well as the search volume and the competition for those keywords?" So I can verify the keyword research within the conversation and I don't have to go to Ahrefs, Semrush, whatever other SEO tool. I can do it all within perfectly within Chat GPT. Target keyword "financial advice" of websites, search volume low, competition I mean perfect. And now when you're even writing a blog post with GPT, you can ask it to review your blog post, have a look at the keywords, see which ones I should be injecting that are really beneficial for me to have in the blog post, for example, a really simple but useful strategy.
Now, the one thing that this is going to lack is the ability to understand how to rank in the AI overviews, just because Chat GPT hasn't been trained on that because it's still very new. But what you can do is watch this video here that gives you a detailed seven-step guide on how to rank higher on the AI overviews. So you can use Chat GPT to create a bit of a strategy for you and follow those seven steps and you'll be ranking in the AI overviews in no time.