---
title: 'This AI Tool Does Keyword Research For You'
source: 'https://youtube.com/watch?v=3n-24cTaE0Y'
video_id: '3n-24cTaE0Y'
date: 2026-07-14
duration_sec: 1560
---

# This AI Tool Does Keyword Research For You

> Source: [This AI Tool Does Keyword Research For You](https://youtube.com/watch?v=3n-24cTaE0Y)

## Summary

The video demonstrates how to build an AI-powered keyword research tool using Agent A that automates the entire process from seed keyword generation to content briefs and internal linking maps, all from a single input word.

### Key Points

- **One-Word Keyword Research** [00:00] — The tool generates a complete keyword roadmap for any niche in about 20 minutes, including clusters, content briefs, and internal linking maps.
- **Prompt for Building the Tool** [01:27] — The creator used a single prompt: 'Build an automated keyword research tool using Opus 4.7. User inputs a niche, tool generates seed keywords, clusters them, and produces a comprehensive report with action plan.'
- **Fine-Tuning Cost** [02:09] — Over $200 in API tokens were spent fine-tuning the tool over several days.
- **Agent A Platform** [02:25] — The tool is built on Agent A, an AI agent with unrestricted access to Ahrefs data, enabling real SEO and marketing data retrieval.
- **Keyword Research Process** [05:13] — Steps: find seed keywords, generate keyword ideas, filter by questions/modifiers, assess SERP competition and search intent, cluster topics, and plan internal linking.
- **Problem: Overly Specific Seeds** [08:14] — Initial seeds like 'golf shoe reviews' yielded only 90 keywords. Solution: use broader seeds like 'golf' and mine competitor keywords via Ahrefs top pages report.
- **Problem: Intent Misclassification** [10:58] — The tool initially flagged branded navigational SERs (e.g., Topgolf) as winnable. Fixes: added branded SER detector, publisher dominance rule, and free tool identifier.
- **Problem: Irrelevant Keywords (Homonyms)** [15:20] — AI filters out homonyms like 'mini driver' (golf vs. car) automatically, removing 2,456 off-topic keywords in one run.
- **Content Briefs Feature** [16:46] — Generates briefs with target keyword, SER analysis, competitor weaknesses, title suggestions, outline, key entities, and FAQ mined from Reddit.
- **Hub and Spoke Map** [20:10] — Visual internal linking map showing pillar and spoke pages with anchor text suggestions and rationale.
- **House Plants Demo** [21:22] — For 'house plants', the tool found 6,000+ queries, identified top opportunities (e.g., Monstera varieties and care), and generated a full action plan.

### Conclusion

The AI keyword research tool automates hours of manual work into 20 minutes, but requires human oversight for intent and strategic decisions. It is available for one-click install on Agent A.

## Transcript

What if you could type just one word and have all&nbsp; of your keyword research done for you? And I mean&nbsp;&nbsp; everything. You can see the keywords that you&nbsp; should go for. You can see them organized into&nbsp;&nbsp; clusters. You can generate content briefs at&nbsp; the click of a button. And while those briefs&nbsp;&nbsp;
are generating, visually see exactly how these&nbsp; clusters should be internally linked throughout&nbsp;&nbsp; your site. There's no tool that does this well&nbsp; and fast until about 2 days ago. I built an AI&nbsp;&nbsp;
tool that can do all of this, hand you the entire&nbsp; keyword road map for any niche, hundreds of hours&nbsp;&nbsp; of work packed into about 20 minutes. And today,&nbsp; I'm going to show you how to build it yourself.&nbsp;&nbsp;
All you have to do is just type in one word. So,&nbsp; I'll do house plants and run the research. And&nbsp;&nbsp; right now, it's looking for seed keywords. It's&nbsp; going to go in and expand those keywords. And now,&nbsp;&nbsp; it's filtering the keywords. And this is where&nbsp; the bulk of the time goes because AI is actually&nbsp;&nbsp;
processing every single keyword as it goes through&nbsp; to make sure that they're relevant first and then&nbsp;&nbsp; it's running different at calls right here. It's&nbsp; at 19% and it's going to keep going to make sure&nbsp;&nbsp; that we're getting the exact keywords that would&nbsp; be relevant to in this case the niche of house&nbsp;&nbsp;
plants. And after this happens, the clusters are&nbsp; going to form. It's going to generate an executive&nbsp;&nbsp; summary and tell you exactly what happened. So&nbsp; while this is happening, let me show you the&nbsp;&nbsp; prompt that I used to build this. So I started off&nbsp; this chat session by saying, "Build an automated&nbsp;&nbsp;
keyword research tool using Opus 4.7. The user&nbsp; inputs a niche. The tool generates seed keywords,&nbsp;&nbsp; clusters them into groups, and produces a&nbsp; comprehensive report with an action plan. In terms&nbsp;&nbsp; of output, all I wanted was keyword clusters with&nbsp; search intent classification. Super important. A&nbsp;&nbsp;
competitor gap analysis and a visual map showing&nbsp; internal linking structures between clusters.&nbsp;&nbsp; And this part is usually the part that takes a&nbsp; really long time. So I said I also said that I'm&nbsp;&nbsp;
open to additional suggestions during planning and&nbsp; that's the only prompt that I used. Now this tool&nbsp;&nbsp; didn't come from one hero prompt. I spent days&nbsp; fine-tuning this and over $200 in API tokens,&nbsp;&nbsp;
which is really small in the grand scheme of&nbsp; things considering what it's able to do. Uh,&nbsp;&nbsp; so rather than just reading my entire agent&nbsp; a chat to you as a really bad bedtime story,&nbsp;&nbsp; I'm going to highlight some of the biggest&nbsp; problems and I'll explain exactly how I fix them&nbsp;&nbsp;
so you can build it, too. If you run into the same&nbsp; problems, some of them you might, you can build it&nbsp;&nbsp; with the exact same specs that I did here. Now,&nbsp; the way that I built this tool is with agent A.&nbsp;&nbsp; And if you don't know what Agent A is, it's an AI&nbsp; agent with unrestricted access to HF's data. So,&nbsp;&nbsp;
it can actually pull real SEO, Google, and&nbsp; marketing data, which is obviously super&nbsp;&nbsp; powerful for any marketer. And you'll see why&nbsp; that the unrestricted access part matters so much,&nbsp;&nbsp; especially for things like SER data as we go&nbsp; through the build. So, it starts by asking me&nbsp;&nbsp;
where I want this tool to live. And so, I wanted&nbsp; it as a console app. And so you can see that I&nbsp;&nbsp; built it here. Um, and we can access this. I can&nbsp; access the keyword research tool just like that.&nbsp;&nbsp; And then it asks where we want to source the data.&nbsp; So I said keywords explorer as well as HRF site&nbsp;&nbsp;
explorer. So we can actually find uh keywords that&nbsp; aren't necessarily so obvious but competitors are&nbsp;&nbsp; ranking for. Then it asked me about the competitor&nbsp; gap analysis like how should the competitors be&nbsp;&nbsp; picked? We're basically just taking data from&nbsp; the organic competitors report in hrefs and&nbsp;&nbsp;
how big should it be? Uh, not that important. um&nbsp; beyond what you listed which of these are useful.&nbsp;&nbsp; So it kind of gave me this checklist of different&nbsp; things and I told it the prioritization score per&nbsp;&nbsp;
cluster is important and the way it's scoring it&nbsp; is based on volume intent and difficulty. Quick&nbsp;&nbsp; wins is always good because especially when you're&nbsp; starting a new site, you don't want to compete for&nbsp;&nbsp; things that are out of your league. And so getting&nbsp; a few quick wins is always a good thing to do. So&nbsp;&nbsp;
uh low difficulty is one that I flagged. And then&nbsp; I want to be able to export this as PDF and CSV.&nbsp;&nbsp; What style for the cluster link map? And I chose&nbsp; the hub and spoke diagram. And this is something&nbsp;&nbsp;
that was popularized by HubSpot many years ago&nbsp; where we have, you know, a pillar content in the&nbsp;&nbsp; center and then supporting pages around it. So&nbsp; that's basically this one. Yeah. So this is the&nbsp;&nbsp;
whole idea of clustering. So, we have something&nbsp; like this where we have a pillar page about&nbsp;&nbsp; putters and then we have spokes that come from it.&nbsp; So, you know, best putter, hockey stick putter,&nbsp;&nbsp;
Tiger Woods putter, Scotty Sheffler putter, some&nbsp; different ideas that we can go after there. All&nbsp;&nbsp; right. And then it takes the plan, it goes,&nbsp; it tells you everything that it's doing. It's&nbsp;&nbsp; basically like somebody who's thinking out loud.&nbsp; And it'll go, it'll build the UI. It has more&nbsp;&nbsp;
questions as it's going because it's processing as&nbsp; it goes. I want us data. I want uh competitors to&nbsp;&nbsp; be from the top five and then we're going to cue&nbsp; it and notify when ready. Then it goes it builds&nbsp;&nbsp;
me the first version of the tool and then that's&nbsp; when we run into real problems. So before we get&nbsp;&nbsp; into the problems I need to explain the keyword&nbsp; research process to you because everyone does it&nbsp;&nbsp; slightly differently but there are certain pillars&nbsp; that need to be done. So the first step is finding&nbsp;&nbsp;
seed keywords. So for a niche like golf that&nbsp; might be something like golf, putter, irons,&nbsp;&nbsp; driver and we can run a search. So this is what&nbsp; happens. Now from this search we need to go and&nbsp;&nbsp;
generate a whole bunch of keywords. So a keyword&nbsp; research tool will give you a whole bunch. But the&nbsp;&nbsp; thing is these are not going to be the queries&nbsp; that you're actually going for. So you need to&nbsp;&nbsp; actually filter these down. There's 4 million&nbsp; keywords in here. That's crazy, right? So you&nbsp;&nbsp;
might do that by looking at questions related to&nbsp; these queries. So, like what is a golf handicap?&nbsp;&nbsp; How do I That's irrelevant because of driver,&nbsp; a golf driver versus Door Dash driver. We'll&nbsp;&nbsp; questions or you might actually use modifiers. So,&nbsp; a modifier is just an add-on to a base keyword.&nbsp;&nbsp;
So, if we have driver golf irons, if we type in&nbsp; best as a modifier and we click show results,&nbsp;&nbsp; then we'll have things like best golf balls,&nbsp; best golf courses near me, you know, queries that&nbsp;&nbsp;
people are actually searching for. So, once we've&nbsp; filtered this down, we've taken all the different&nbsp;&nbsp; um queries like from the questions, different&nbsp; modifiers, manually searching through some,&nbsp;&nbsp; we're going to end up with a huge list of&nbsp; keywords. Now, after you've taken these four&nbsp;&nbsp;
million keywords and and narrowed it down to a few&nbsp; hundred or a few thousand, you have to actually&nbsp;&nbsp; look at the search results for all of these&nbsp; queries because you need to figure out if you have&nbsp;&nbsp; a chance at competing against these competitors&nbsp; in the search results. And B, you have to assess&nbsp;&nbsp;
search intent. And search intent represents the&nbsp; reason behind a searchers query. So, can you&nbsp;&nbsp; actually match the reason for the intent? So, if&nbsp; we're looking at best golf balls, we're looking&nbsp;&nbsp; at the search results here. Okay, Reddit's ranking&nbsp; here. Today's golfer is is pretty authoritative,&nbsp;&nbsp;
too. Uh, Outbound Golf, this is a DR36, so it's&nbsp; not that authoritative of a website. DR is not&nbsp;&nbsp; the be all end all of things, but we see Amazon&nbsp; here. And we can see that the intent overall is&nbsp;&nbsp;
is quite commercial, but we can see that blog&nbsp; posts are ranking here. So, for a content site,&nbsp;&nbsp; of course, you can rank with with a blog post. So,&nbsp; we would decide, yeah, you know what? we will go&nbsp;&nbsp; for best golf balls and target that. Now imagine&nbsp; doing that hundreds or thousands of times, right?&nbsp;&nbsp;
Super long process and we're not done here. Then&nbsp; you have to figure out how these topics all kind&nbsp;&nbsp; of blend and mesh together like which posts should&nbsp; be linked to and from and we create this internal&nbsp;&nbsp; web structure so that search engine crawlers&nbsp; can come and access all those sites easily&nbsp;&nbsp;
understand which ones should be connected to each&nbsp; other. So keyword research is just a much bigger&nbsp;&nbsp; process than just finding keywords. And as I was&nbsp; building this, I ran into problems pretty much at&nbsp;&nbsp;
every stage. And so I'm going to highlight some&nbsp; of those for you and how I was able to fix them.&nbsp;&nbsp; The solutions that that uh AI and I were able to&nbsp; come together to solve those problems. Okay. So&nbsp;&nbsp; the first thing that AI did was it had to come&nbsp; up with seed keywords. And it's super important&nbsp;&nbsp;
that you get your seed keywords right because&nbsp; everything else in keyword research that comes&nbsp;&nbsp; after are built around the seed keywords. And so&nbsp; the first run at this, it came up with seeds like&nbsp;&nbsp; golf shoe reviews or striicks on versus titleless&nbsp; golf ball. Like this is super specific. And the&nbsp;&nbsp;
and the problem with this when you use these as&nbsp; seeds is when you go to generate keyword ideas,&nbsp;&nbsp; you're not going to come up with any. So in this&nbsp; case, these two seeds only have 90 keywords.&nbsp;&nbsp; And if you look at the keyword list, like there's&nbsp; no volume behind it. There's no demand. So,&nbsp;&nbsp;
it's not about going for like these high ultra&nbsp; high search volume keywords, but there just isn't&nbsp;&nbsp; any demand around these. And sure, you can create&nbsp; content around these, but is it worth the time&nbsp;&nbsp; and effort? Probably not, right? So, I had to tell&nbsp; it specifically to go much broader. So instead of&nbsp;&nbsp;
using golf shoe reviews, golf ball reviews, golf&nbsp; hat reviews, just go with golf and we'll figure&nbsp;&nbsp; that stuff out later. And then go with instead&nbsp; of strixon versus titalist golf ball. Like that&nbsp;&nbsp;
is one query that we might actually target,&nbsp; not a seed. Let's go with titleist different&nbsp;&nbsp; manufacturers and then go and expand from there.&nbsp; Now, the problem with this approach though is&nbsp;&nbsp; that if you're only going for the usual suspects&nbsp; like golf and putter and putters, everyone else is&nbsp;&nbsp;
doing keyword research that same way. And so, it's&nbsp; tough to find lowhanging fruit that people aren't&nbsp;&nbsp; necessarily targeting, but there is still a lot of&nbsp; demand around those topics. And so, what I had the&nbsp;&nbsp;
AI do is to actually mine competitors keywords&nbsp; using the HF's top pages report. So it would&nbsp;&nbsp; find the competitors. It would mine those queries&nbsp; to see what they're ranking for where there might&nbsp;&nbsp;
be a pattern with certain seed keywords. And so&nbsp; was able to actually find stuff that you normally&nbsp;&nbsp; wouldn't be able to find or that wouldn't be&nbsp; super obvious to find especially in a first&nbsp;&nbsp; round of keyword research. And so agent A was able&nbsp; to get the seeds down perfectly and it was able to&nbsp;&nbsp;
go through to the next stage. So to show you an&nbsp; example, I'm looking at the cluster right now for&nbsp;&nbsp; golf apparel and shoes. And it was able to group&nbsp; all of this stuff together. So like golf shoes,&nbsp;&nbsp; golf shorts. Why is it telling us to go for the&nbsp; head term? Well, it's because there's a listical&nbsp;&nbsp;
that's ranking there in the top five, right? So&nbsp; it's actually looking at the search results here&nbsp;&nbsp; and it's able to determine that. It's able to&nbsp; look at all these different apparel type stuff,&nbsp;&nbsp; all these apparel type queries, best&nbsp; golf sunglasses, and telling us, yeah,&nbsp;&nbsp;
this is probably something that you should go&nbsp; for as well. And there's so many different things&nbsp;&nbsp; in here. And obviously, you don't have to create&nbsp; content around every single topic in the cluster,&nbsp;&nbsp; but it's meant to be here where you get everything&nbsp; shortlisted for you. Now the next problem that&nbsp;&nbsp;
actually came is related to this and you can't&nbsp; see it here but the first round of creating this&nbsp;&nbsp; keyword research tool. There were some serious&nbsp; issues with understanding intent. So the way that&nbsp;&nbsp; AI made this system is based on like a verdict&nbsp; engine. So it decides whether you should go for&nbsp;&nbsp;
them. So these are queries that fit a golf content&nbsp; site. There are a group of may where it's like you&nbsp;&nbsp; could go for it, you don't have to. Um, but yeah,&nbsp; these are things that you can look up manually&nbsp;&nbsp;
if you want to expand your keyword list further.&nbsp; There's ones that lack data. So, those ones I just&nbsp;&nbsp; skip over because we have plenty to go with. These&nbsp; ones are definitely skips. And so, for example,&nbsp;&nbsp; Topgolf, you're not going to rank for it. And&nbsp; even if you do, people probably aren't going to&nbsp;&nbsp;
click it because it's a very much a navigational&nbsp; search, which it shows right there. And so, in the&nbsp;&nbsp; first run, this keyword research tool was very,&nbsp; very bad at identifying intent. It told me that I&nbsp;&nbsp;
should try and rank for Topgolf. Can you rank for&nbsp; it? Yes, this person does because of an editorial&nbsp;&nbsp; site. And this is where human overlook is actually&nbsp; quite important. And so I was able to train the&nbsp;&nbsp;
AI to almost think and act like a human as it&nbsp; goes through this and make that decision like,&nbsp;&nbsp; hey, probably not going to make sense to go for&nbsp; Topgolf or golf or golf courses near me. So yeah,&nbsp;&nbsp;
it was it was waving through branded SERs like&nbsp; Topgolf or Scotty Cameron and they're saying that&nbsp;&nbsp; these are winnable content opportunities when&nbsp; in fact they weren't. The first example that&nbsp;&nbsp; I actually went with was related to the coffee&nbsp; niche. So it's like um it was basically waving&nbsp;&nbsp;
through things like espresso machine brands. It&nbsp; was telling me to go in and rank for like brevel&nbsp;&nbsp; and I had to correct it and teach it. So look,&nbsp; it's telling me that the pillar keyword here was&nbsp;&nbsp; Breville espresso machine. Like with a content&nbsp; site, you're probably not going to rank for that&nbsp;&nbsp;
one there. it's probably not worth the effort in&nbsp; that immediate moment. So I was saying, shouldn't&nbsp;&nbsp; it be something like best espresso machine? So&nbsp; the thing with AI is that if you ask it to do&nbsp;&nbsp; something broad, it's going to do something broad&nbsp; and it's not going to be very good. So that's why&nbsp;&nbsp;
we had to go through this entire process, but&nbsp; I will tell you the main things that I changed&nbsp;&nbsp; here. So there were a few major adjustments that I&nbsp; used to fix this. So first, I added a branded SER&nbsp;&nbsp; detector. So, if there was a brand that was just&nbsp; constantly listed in like the top three results,&nbsp;&nbsp;
like if it was telling me to rank for Topgolf, you&nbsp; know, you would see Topgolf, their homepage, you'd&nbsp;&nbsp; see their Twitter, their Instagram, and just a&nbsp; whole bunch of things related to Topgolf. You know&nbsp;&nbsp; that it's a branded navigational SER, and so it&nbsp; would know to tell me, let's skip over this one.&nbsp;&nbsp;
Okay. Okay. So, the second thing that I made was&nbsp; a publisher dominance rule where it would look for&nbsp;&nbsp; all these mega sites like like Forbes, USA Today,&nbsp; and if the SERs were completely dominated by them,&nbsp;&nbsp; it would just kind of skip over it or add it to&nbsp; the maybe pile. Um, the next thing was able to&nbsp;&nbsp;
identify free tools. So, when it's looking for&nbsp; keywords, it's actually looking not just for&nbsp;&nbsp; content keywords like blog posts, but it's also&nbsp; looking for video topics and it's also looking for&nbsp;&nbsp; free tool opportunities. And so identifying which&nbsp; ones are free tools was not as straightforward as&nbsp;&nbsp;
it might seem. And so I had to train the tool to&nbsp; actually identify which ones were free tools. So&nbsp;&nbsp; if we look here, I can go to recommended type&nbsp; and it's has a classification here called free&nbsp;&nbsp;
tool landing page. Let me just go for the goes.&nbsp; So there are 342 free tool landing pages. Right?&nbsp;&nbsp; Look at all this. The golf handicap calculator.&nbsp; We can see that these are all golf handicap&nbsp;&nbsp;
calculators. A golf club length calculator. These&nbsp; are all tools. Golf swing analyzer. These are all&nbsp;&nbsp; different tools. Basically, I was going through&nbsp; this loop of looking through data. So, it would&nbsp;&nbsp;
basically do, you know, the more technical stuff&nbsp; of coming up with ideas and the logic to make this&nbsp;&nbsp; work. And I would use kind of my SEO brain to be&nbsp; like, "This one doesn't seem right because of X,&nbsp;&nbsp; Y, and Z." And then we were able to come up with&nbsp; ideas to come to a pretty good solution. So,&nbsp;&nbsp;
and AI will actually go and grade the keywords&nbsp; one by one. So, yes, there are still some that&nbsp;&nbsp; we have to filter through, but in my experience,&nbsp; it's very, very little across three different&nbsp;&nbsp; niches that I tested. Now, the biggest problem&nbsp; I ran into was identifying irrelevant keywords,&nbsp;&nbsp;
especially with homonyms. So, this is where AI&nbsp; is absolute money. So if we go to the overview&nbsp;&nbsp; report, we look at the executive summary. You can&nbsp; see here that these are the seed keywords that&nbsp;&nbsp;
were used. So like golf, callaway, tailor made,&nbsp; these are all brand names, by the way. Puttern,&nbsp;&nbsp; irons, driver. So you can see like iron can mean&nbsp; many different things. Driver can also mean many&nbsp;&nbsp;
different things. So, it actually tells us here&nbsp; every single time it runs through this process,&nbsp;&nbsp; when it goes through that filtering stage, the&nbsp; part that I said takes the longest, it's actually&nbsp;&nbsp;
making sure that all the keywords are relevant.&nbsp; So, it's actually filtering out here 2,456 off-topic keywords. That's crazy. So, for example,&nbsp; Spark driver, Door Dash driver, Mini driver, the&nbsp;&nbsp;
actress, Baby Driver, the movie. It's constantly&nbsp; looking at these these keywords and making sure&nbsp;&nbsp; that you don't have to actually filter them out.&nbsp; And I think that was the best part is that AI has&nbsp;&nbsp; all this knowledge already. So as it goes through&nbsp; it, it's able to say mini driver, nah, not nothing&nbsp;&nbsp;
to do with that. And it'll just take out anything&nbsp; that's related to that. So every single keyword&nbsp;&nbsp; runs through AI and it removes the keywords&nbsp; that are irrelevant to the niche of golf. Now,&nbsp;&nbsp; not everything was bad. There were some things&nbsp; that were actually great uh out of the box. And so&nbsp;&nbsp;
here, one of those things was content briefs. So&nbsp; if we look at like I don't know, best golf pants,&nbsp;&nbsp; this is one that I generated. Um when I click&nbsp; those buttons to to generate content briefs,&nbsp;&nbsp; it'll do things like this. It'll show you the&nbsp; target keyword. It'll tell you about the SER.&nbsp;&nbsp;
So you know, your content writer can go in there&nbsp; with, you know, some basic knowledge and know who&nbsp;&nbsp; they're competing against, the types of pages.&nbsp; It actually goes and it reads the different pages&nbsp;&nbsp; that are ranking in the top 10 and it's able to&nbsp; identify weaknesses, areas that you can capitalize&nbsp;&nbsp;
on to create better content. So, it says here that&nbsp; for best golf pants there's a weak editorial that&nbsp;&nbsp; there's a weak editorial landscape and only one&nbsp; true listical which is kind of interesting to me.&nbsp;&nbsp;
And we can see here that the KD is one the keyword&nbsp; difficulty that there's weak top results that the&nbsp;&nbsp; single editorial content is a DR51. It's doing the&nbsp; analysis for you kind of of what you would do when&nbsp;&nbsp; you look through everything, but you don't have to&nbsp; look through all these SERs at once. You just go&nbsp;&nbsp;
through them one by one and it's all listed here.&nbsp; It'll give you suggestions for titles. So, this is&nbsp;&nbsp; telling about intent. It's telling you to create&nbsp; a listical here clearly. Uh the metad description,&nbsp;&nbsp; metad description is not so important anymore.&nbsp; It'll tell you the URL slug to use a word count&nbsp;&nbsp;
an exact word count here. Um, but it's, you know,&nbsp; giving you the median for the people who do care&nbsp;&nbsp; about word counts or people who are paying other&nbsp; writers to hit certain word counts. And then it'll&nbsp;&nbsp;
do the entire outline for you. And the outline&nbsp; is actually decent. So, it's already looked at&nbsp;&nbsp; the different listicles that are ranking and it's&nbsp; providing suggestions to you based on that. So,&nbsp;&nbsp; it'll tell you, you know, we've seen affiliate&nbsp; sites do this before is they always have this&nbsp;&nbsp;
thing of like how I tested these golf pins or&nbsp; whatever it is. I think that's actually quite&nbsp;&nbsp; good for the reader as well to know that you have&nbsp; actually reviewed these, you've touched them,&nbsp;&nbsp; you've worn them and tested them. So, we want to&nbsp; give people things at a glance because especially&nbsp;&nbsp;
with blog readers today, everyone is skimming.&nbsp; So, make sure that it's skimable. We provide&nbsp;&nbsp; all that info for them. Do you have to label&nbsp; them like this? No. You can give your honest&nbsp;&nbsp;
reviews like that, but if you're truly going&nbsp; for like content site type thing, you know,&nbsp;&nbsp; you at least have seeds and suggestions of what&nbsp; you might want to do. This part is pretty cool.&nbsp;&nbsp; It tells you key entities and topics to cover&nbsp; basically to have, you know, full relevance&nbsp;&nbsp;
and topical coverage here. So, you know, Lulle&nbsp; Lemon, um, Travis Matthew, Bonobos, Nike, Adidas,&nbsp;&nbsp; etc. gives you an FAQ and says that it's mine from&nbsp; Reddit threads in the SER. I did not know this,&nbsp;&nbsp;
but I think that is pretty cool is that it's&nbsp; picking up things like people also ask questions&nbsp;&nbsp; and it's bringing all of this together. And&nbsp; this is where the money is, right? The in the&nbsp;&nbsp; internal linking part is it's telling you that&nbsp; from the target page, so the golf apparel shoes,&nbsp;&nbsp;
which the bigger part of the hub, it's telling&nbsp; you to link from that page, it says this is a&nbsp;&nbsp; spoke page. So this is one of those branches,&nbsp; right? It links up to the pillar and across to&nbsp;&nbsp; relevant siblings. So, it's telling you to link&nbsp; from this page. Uh, tells you the anchor text. It&nbsp;&nbsp;
tells you why you should do this. And so, all this&nbsp; stuff is super super important. And I think that,&nbsp;&nbsp; uh, you know, most content tools are are&nbsp; missing these kinds of features. Again,&nbsp;&nbsp; you need to have some kind of SEO knowledge to be&nbsp; able to make good decisions in this. Take things&nbsp;&nbsp;
with a grain of salt. Now, another thing that it&nbsp; did well out of the box is the hub and spoke map.&nbsp;&nbsp; So right now I'm just looking at, you know, the&nbsp; go keywords here and there are there is a lot of&nbsp;&nbsp; stuff going on here. So let's say that I wanted to&nbsp; just look at the quick ones. Then we can actually&nbsp;&nbsp;
see the clusters. So we can see the different&nbsp; pages here. Um like what makes a golf cart street&nbsp;&nbsp; legal, how to test golf cart batteries. You can&nbsp; tell that these are going to be low competition,&nbsp;&nbsp;
but these are meant to be quick wins. You&nbsp; basically have options is my point here. So&nbsp;&nbsp; the final result, you know, we went from entering&nbsp; a niche. I entered just golf. It came up with a&nbsp;&nbsp; seeds and went through over 10,000 keywords,&nbsp; probably closer to like 400,000, and it would&nbsp;&nbsp;
bring it down to like 10K. It would go through a&nbsp; relevance filter. It would cluster stuff. It would&nbsp;&nbsp; pull all the live SERs. It would provide all&nbsp; the classifications and do the competitor gap,&nbsp;&nbsp;
create the hub and spoke map. It would do per&nbsp; keyword briefs which is on demand because we're&nbsp;&nbsp; not going to generate all of them obviously&nbsp; because that's a waste of API tokens. And you&nbsp;&nbsp; have the option for CSV PDF exports. Now let's go&nbsp; back and see if our house plans thing is done. Oh,&nbsp;&nbsp;
it's done. Literally done right then and&nbsp; there. So gez, it went through. In fact,&nbsp;&nbsp; 11,800 keywords. And this is on house plants. And&nbsp; for the record, I know nothing about house plants.&nbsp;&nbsp; So leave in the comments if this did a terrible&nbsp; job. We'll go look through some of the keywords&nbsp;&nbsp;
and see how well it did. So, the overview here, it&nbsp; found over 6,000 queries related to house plants&nbsp;&nbsp; that we could potentially go after. And this is&nbsp; going to be a real test because with house plants,&nbsp;&nbsp; I would assume that the majority of queries are&nbsp; going to be commercial where, you know, you have&nbsp;&nbsp;
to go and there's going to be e-commerce product&nbsp; category pages, service pages. Um, but yeah, so&nbsp;&nbsp; we can see here these are the seeds that it came&nbsp; up with like a monstera paos. I don't know how&nbsp;&nbsp;
to say that. orchid, snake plant, peace lily. So,&nbsp; it went out and it found the seeds that are like&nbsp;&nbsp; popular house plants. I don't know about that.&nbsp; I've heard of I've heard of these some of these,&nbsp;&nbsp;
but that's it. Top content opportunities, Mona,&nbsp; varieties, and care. Okay, that's cool and helpful&nbsp;&nbsp; to know. Share a voice. This is interesting.&nbsp; the share of voice in this niche. The majority&nbsp;&nbsp; for house plant care advice and it's giving us a&nbsp; recommended action plan. So ship the Mona pillar&nbsp;&nbsp;
first. It has the highest opportunity score and&nbsp; supports a deep cluster of variety of listicles,&nbsp;&nbsp; care how-tos, and problems solving guides. Cool.&nbsp; Let's take a look. So we're just looking at go&nbsp;&nbsp; keywords here and let's look at the clusters. So,&nbsp; it's said to look at the Mona varieties in care&nbsp;&nbsp;
uh or Monstera toxic to cats. This is considered&nbsp; a quick win and a go. So, let's let's generate a&nbsp;&nbsp; brief on that for now. So, basically, we have all&nbsp; the clusters here. We have the competitor gap and&nbsp;&nbsp; we can see the the websites that are getting the&nbsp; most traffic for these queries. So, the spruce&nbsp;&nbsp;
that makes sense. Lively root sounds like it makes&nbsp; sense. Uh and the hub and spoke map. This one I'm&nbsp;&nbsp; a little nervous to see. I'm not looking yet. But&nbsp; yeah, that's crazy. Oh man, like these these hubs&nbsp;&nbsp; and spokes. You know what? Because it's limited to&nbsp; I limited to I think 15 to 20 for medium searches,&nbsp;&nbsp;
it's actually quite reasonable to look at. And&nbsp; being someone who knows almost nothing about house&nbsp;&nbsp; plants. I can come here and I can see how some&nbsp; of these should be connected. Saying that there&nbsp;&nbsp; should be reciprocal links here because this one&nbsp; links to here. So this one on alocasia varieties&nbsp;&nbsp;
and care should link to airplane care and vice&nbsp; versa. We can see the different ones that content&nbsp;&nbsp; that we should create. We can see the supporting&nbsp; content here. This is all go content. But this is&nbsp;&nbsp; all done for me and I think that it's done a very&nbsp; very good job. Uh the content briefs they're all&nbsp;&nbsp;
done uh generating now too. So let's look at the&nbsp; one for cats. We have the title. So are monstera&nbsp;&nbsp; plants toxic to cats? What every owner needs to&nbsp; know. Vet reviewed. Okay. Well, that's not true.&nbsp;&nbsp; But, you know, we could get a vet to do that. I'm&nbsp; sure it's done the competitor analysis for you.&nbsp;&nbsp;
Like, I like I'm actually genuinely impressed&nbsp; with this as I'm as I'm going through this. I&nbsp;&nbsp; don't know how accurate this stuff is and how how&nbsp; factual it is. I'm not a vet either, but you can&nbsp;&nbsp; see here what to do if your cat shoot a monster&nbsp; leaf step by step. It's telling you everything to&nbsp;&nbsp;
do. Call your vet or ASPCA poison control. Will my&nbsp; cat is just answering questions really of someone&nbsp;&nbsp; who's searching for this. But from what I'm seeing&nbsp; here, I can see that it does a lot of really&nbsp;&nbsp;
really good work for us. And all I did was type&nbsp; in one word, house plants, and it came up with all&nbsp;&nbsp; of this, which is super cool. Now, the the crazy&nbsp; thing about this all is in agent A, if I wanted&nbsp;&nbsp;
to make a different change to this, let's like I&nbsp; don't like the way that let's go back to our house&nbsp;&nbsp; plans. Let's say that I wanted to have I didn't&nbsp; want to have that limit of like 15 to 20 clusters,&nbsp;&nbsp; but I wanted to have as many clusters as as&nbsp; needed. It would go and do potentially an infinite&nbsp;&nbsp;
number of those, right? Um the hub and spoke map.&nbsp; Let's say I don't like the colors or when I click&nbsp;&nbsp; on one of these, let's say I want there to be&nbsp; like a little box that pops up and I can generate&nbsp;&nbsp; the content brief directly from the hub and spoke&nbsp; map. All I have to do is go in here, type that up,&nbsp;&nbsp;
enter it, and it'll make those changes. Now,&nbsp; if you want this exact keyword research tool,&nbsp;&nbsp; you can build it yourself using the steps that I&nbsp; walk through. You know how to solve the problems,&nbsp;&nbsp; too. Or if you've got an agent a workspace,&nbsp; you can oneclick install it from the app&nbsp;&nbsp;
store. Technically two clicks uh and have&nbsp; it installed directly into your workspace.&nbsp;&nbsp; And if there's anything that you want me to&nbsp; build for you, tell me in the comments below.