Keyword research usually starts the same way no matter which tool you use: pick a seed keyword, pull a list of related terms and metrics, and then figure out what to actually do with all of it. The tools that dominate this space, like Ahrefs and Semrush, are excellent at the first part and leave you on your own for the second. That gap between a keyword list and an actionable plan is where most of the manual grind in SEO still lives.
Keyword Research in Ahrefs
Ahrefs has a strong keyword database and is notably more affordable than its main competitor, Semrush. Running a seed keyword like "agentic marketing" through Keyword Explorer returns keyword-level metrics such as search volume, global search volume, and CPC, plus expanded ideas including People Also Ask questions and matching terms that contain the seed keyword.
A search on "agentic marketing" pulled a solid list of 652 keywords, loaded quickly, and covered the bases you'd expect. The catch is that it's still just a spreadsheet. To turn that list into something actionable, the data has to be exported, or connected through something like an MCP into a tool like Claude, before any real analysis or planning happens. Ahrefs is a good, affordable starting point with a manageable learning curve, but it still leaves the bulk of the manual work in the researcher's hands. Semrush, by contrast, has a much larger keyword database, but at a higher price.
Building a Keyword Research Agent in Saaga Solve
Saaga Solve is an agentic marketing platform built for SEOs to automate the repetitive parts of the job without taking over the workflow itself. Every SEO has their own particular process, and legacy tools like Ahrefs and Semrush force everyone into the same workflow, even when they bolt on AI features. With Saaga Solve, agents can be built around a specific way of working, so the process stays personal rather than dictated by the platform.
Instead of typing a seed keyword into a search box, the request goes to a custom-built keyword research agent in natural language: give me 30 high intent, low difficulty keywords around and including agentic marketing. The agent pulls from Semrush's database automatically, recognizing it as the stronger keyword source, without needing a separate Semrush or Ahrefs license. The output includes the same metrics you'd expect: volume, keyword difficulty, CPC, and intent.
So I'm actually paying less for Saaga Solve to get access to the same data and have it manipulated and served to me in my particular style.
Turning Keywords Into a Pillar and Cluster Content Plan
Because the request happens in natural language, the next step is just asking another question. After the initial list of 30 keywords, the agent offers to map them into a pillar and cluster content plan or check which competitors currently rank for them. Choosing the content plan sends the agent off to pull SERP data for the top keywords, then reason against it, work that would be painfully slow to do by hand.
The SERP analysis surfaced a clear gap: the keyword "agentic marketing agency" returns zero existing results, meaning nobody has built a page for that exact topic. From there the agent proposed a hub and spoke plan: a pillar page built as a complete guide targeting "agentic marketing," representing 640 combined searches, with supporting cluster pages targeting smaller, lower-intent keywords like "best AI for marketing agency SEO guide" all linking back to strengthen the topical pillar.
Memory, Model Choice, and Cost
Once the plan is built, the agent saves it to memory. Reopening a new chat about the same topic later means the research doesn't need to be repeated, which saves time and avoids the cost of re-pulling data from tools each time, a common problem when working through MCPs that constantly refresh the same queries. The data can still be refreshed on request when it's actually needed.
Saaga Solve also gives access to a range of models to run these tasks, letting the choice of model match the cost and reasoning needs of the task rather than defaulting to a single frontier model for everything. That's an additional layer of customization on top of the custom workflow itself.
Getting Started
Saaga Solve offers a free trial with $5 in usage, typically enough to run at least 10 workflows like the one described here. The platform isn't limited to keyword research; any SEO workflow can be brought in and tested. Agentic tools are becoming part of how SEO work gets done, and sitting out that shift means falling behind teams already using them.
Transcript
Let me show you how much easier it is to do keyword research in Saaga Solve versus ahrefs. Now, if you're not familiar with Saaga Solve, Saaga Solve is an agentic marketing platform built for SEOs like you and I to automate the grunt work of SEO. Not to replace us, but to give us superpowers, if you will, so that we're not bogged down using legacy tech and jumping around spreadsheets to get the data and insights we need. And, and instead we are quick to action and working alongside agents, um, to future proof our strategies. So let's get right into it. First in ahrefs, let's start with the fan favorite, right? Ahrefs has a bunch ah, of fantastic database. It's especially compared to its competitor Semrush, it's a lot more affordable. So to do keyword research, we're going to pop into Keyword Explorer. Let's put in our seed keyword, let's say agentic marketing. Now, seed keyword is a keyword that's, you know, kind of the root, the start of all the research. Right. What do you want to anchor around? As you notice, when I type in that keyword, I get the metrics, uh, on that keyword level. Search volume, global search volume, cpc, kind of all the things I want to see. Down here though is really where I can expand that research the keyword, uh, ideas. So I get questions that comes from Google's people also ask questions that are also associated with that keyword. And then I also get terms, um, match or keywords that contain that keyword in it. So pros to Ahrefs, um, giving credit words due. This is a, you know, solid list, 652 keywords off the riff load super fast. Um, but, you know, it just feels like another spreadsheet, right? In order to, um, manipulate this data or get down to the actionable insights, I have to export it. Maybe I connect the uh, um, MCP to Claude and I can interact with the data there. Um, I will say though, yes, ahrefs is way cheaper than semrush, um, especially for things like keyword research. Semrush's database is far better than ahrefs they have, I frankly think billions more keywords indexed and in SEMrush's database than Ahrefs. So I'll just say pretty good start if you're looking for an affordable tool that, you know, doesn't have a sharp learning curve. Um, but there's still a lot of manual work for me to be able to do. Something with this data and get it actionable. Now let's look at Saaga Solve. So if you haven't used an agentic marketing platform before, this is going to really put you on game, so to speak. So one thing I love about Sega Soft off the jump is I've built my own keyword research agent. So every SEO has particular flavors as I like to describe it. They have their own unique process and I'm sure you do too, of how you like to do things Specifically, unlike Ahrefs, Semrush or any of the legacy techs, they force you into their workflow, right? Even when they're using AI, it's their prompting behind the mcps and you are control and, and your special flavor is kind of subdued, right? And you're forced into their workflow. With Saaga Solve, I can actually use agents to build agents based on my workflow. So I like to say if you could dream it, you can build it in Saaga Solve. I've already built my own keyword research agent. So all I have to say, speaking in natural language to my agent, give me 30 high intent, low difficulty keywords around and including agentic marketing. As you can see, it's going to pull up SEMrush, um, because that is the better keyword database. So I don't even need to use AHREFS data. Although there is keyword, uh, uh, uh, integration of Ahrefs in Sega Solve. No license needed. Same with Semrush, no license needed here. So I'm actually paying less for Sega Solve to get access to the same data and have it be manipulated and served to me in my workflow in my particular style. A lot better, um, than I am getting through Ahrefs or semrush. And again for cheaper because I don't need an additional license like yourself comes with it built in. And then here's the list that spit out. So we've got agents for marketing, same kind of things I'm looking for as I was in hrs, right? My volume, my kd, my cd, my cpc, my intent, blah blah blah. And the good thing is, remember when we looked at hrs, right? Like how do I uh, actualize that data? What do I do with it? Because I'm doing this in natural language, I can just easily expand, ask more questions, etc. In fact, Saaga Solve is telling me want me to map these 30 into a pillar and cluster, uh, content plan or check which competitors currently rank. Let's say we want a pillar and cluster content plan. Foreign. Let's just see Where Saaga goes in that direction. By the way, you get access to a wealth of models. These are my top picks here. Um, my favorite is Geo GLM 5.3 Flash. I know it's not a Frontier model, it's not your chat GPT, it's not your claude. But the reasoning and analytical skills compared to the cost per 1 million tokens are far beyond any model I've seen. And that's another layer of customization you get with Sago solve, right? Not just your workflow, but how, which models you're going to pull into which tasks. So you're making an effective balance between cost and the reasoning and critical thinking needed for that task. From my ask, it's going to pull the actual SERPs again using SEMrush because it knows that is the better keyword research base rather than Ahrefs. Imagine taking all these SERPs manually, you guys. I mean that is such a pain. So it's looking at the top keywords, pulling the SERPs in and then, uh, reasoning against that, right? So SERPs tell a clear story. There's one big finding Agentic Marketing Agency returns zero results. Nobody has built a page for that shape. State plan. Here it is. So the plan is, um, making a pillar complete guide, um, to target, uh, the keyword agent marketing, 640 searches combined. And then within that. So this is the hub and spoke model, right? You have your main guide you want to drive traffic to, and then you build out these pages that target smaller keywords, um, or less intent keywords, and you link them all together to create a really strong topical pillar. So you got your cluster, one best AI for marketing agency SEO guide, so on and so forth. Right? Um, and then you got some notes here and it breaks it down again. Oh, this is a great example. It'll actually save this to the agent's memory so that as I open up new chats and I'm talking around this topic, um, or I reference this guide again or a topic that the guide has, it'll remember that in the future. Which saves you A a ton of time because it already has the research and B, frankly it saves you a lot of money in the long run because it doesn't have to repull those tools again. Uh, obviously if you want to refresh the data, you can ask for that. But what I see a lot, especially when people work in mcps, is that the agents they're using like a claw, just has to constantly refresh that data. Um, and you guys know that gets so exp. Anyways, if you want to save yourself a lot of headache at a fraction of the cost. I know that sometimes that sounds too good to be true, but don't. You don't have to take my word for it. There's a free trial with Sega Solve right Now. You get $5 in usage, which is typically enough to run at least 10 workflows. 10 workflows just like this one. So you can see the tool for yourself. It's not just limited to keyword research. Bring any SEO workflow you have in mind and and give it a try. Start today@, uh, saagasolve.com that's s a a g a s o l v e dot com. Thank you. My name is Laura. I'm a seasoned SEO. If you're not using agents, I promise you you're not going to be left behind. You're actively being left behind right now. Future, proof your strategy and give Saaga Solve a try. Have a good day.
Frequently Asked Questions
Yes, Ahrefs is described as noticeably cheaper than Semrush, especially for keyword research, though Semrush's keyword database is considered far larger and better.
No, Saaga Solve integrates keyword data from both Ahrefs and Semrush without requiring a separate license for either tool.
The free trial includes $5 in usage, which is typically enough to run at least 10 workflows so you can test the tool yourself.
Yes, after pulling keyword ideas and analyzing the SERPs, the agent can map the keywords into a pillar and cluster content plan and identify content gaps, such as a keyword with zero ranking results.