As an SEO content strategist, I’ve spent several years of my career helping SaaS and B2B companies, and even small local businesses, create impactful SEO content strategies.
In 2025, my own website brought in $45k in revenue from my SEO content strategy that I put together last December. If interested, I’ve shared all my journey, step by step, with all the details in my monthly recaps and affiliate reports.
$45k is a massive jump compared to 2024, when it only made a little over $10k. I followed the exact same steps I normally use for clients.
But my process of SEO content strategy creation back then was still mostly manual.
In a nutshell, I logged into Semrush for a technical audit, pulled competitor reports separately, ran keyword research on its own, cross-checked indexing in Google Search Console, and then stitched everything together in a doc. The result was usually solid, but the process was not scalable.
If you were juggling multiple client websites, you simply could not do deep work on every single one. At least, that’s been my experience managing client accounts for years.
With AI SEO tools in place, my SEO content strategy creation process has changed completely.
I now connect Semrush MCP and Google Search Console directly to Claude, and the entire strategy creation process happens inside one conversation. I describe what I need, Claude pulls live Semrush data, cross-references it against my GSC indexing and performance numbers, and hands back a plan I can actually act on.
In my experience, this turns a one-week strategy project into a one or two-hour conversation. And it’s a real strategy, not a generic AI output, because every recommendation is based on data from Semrush and GSC.
If you are still treating keyword research, competitor analysis, and content planning as three separate tasks, you are doing more manual work than you need to.
Let’s dive into how my AI SEO content strategy generator works.
P.S. Some links in this post are affiliate links, including the Semrush links. This means I’ll earn a small commission if you decide to try Semrush One. I’m sharing this because you deserve to know. Even if you don’t sign up through my link, I’m happy to keep recommending Semrush, since I’ve been using it myself since 2019.
My tool stack for AI SEO strategy generation
Before showing you my workflow, let me quickly describe the tools I use. It doesn’t mean you have to use exactly the same tools, my AI SEO content strategy process will work if you use other LLMs, or any other SEO tools.
Claude is my main LLM for managing my website. I use Claude Pro to have dedicated projects and skills for all my needs, both work and personal life-related. To use Claude effectively for all my AI and SEO needs, I’ve been feeding it with lots of personal information about me, my writing style, my work approach, and more.
Semrush One is my go-to SEO tool for keyword research, competitor analysis, technical site audits, AI visibility tracking, and pretty much anything related to SEO. I connected it to Claude through the Semrush MCP connector, now built directly into Claude’s connector directory. If you wonder how it works, check my Semrush MCP review.
Google Search Console is a source of the organic performance data about my website. Unlike Semrush, GSC does not have a pre-built connector. So I set GSC up manually through a custom MCP connector in Claude Desktop. It’s an extra step, but it’s worth it.
My 5-step SEO content strategy with AI
This is the exact sequence I run inside one Claude conversation, using the Semrush MCP connector and my custom GSC connector together. I’ve created an AI-generated SEO content strategy for my website with a 12-month content plan this way (see below).
1. Technical website audit
I start by asking Claude to pull a technical SEO audit through Semrush Site Audit, then prioritize the findings by organic performance impact instead of just listing every issue it finds.
A raw Semrush audit can return hundreds of line items, and most site owners have no idea which ones are actually important. I ask Claude to sort tasks into tiers: critical issues with high organic impact first, then moderate-impact fixes next.
In one recent audit, Claude flagged four orphaned pages with zero inbound internal links and three revenue clusters that needed stronger internal linking. Each task had a specific description, which helped better understand the issue and why it had to be addressed.
The value here isn’t that Claude finds issues that Semrush wouldn’t already show you. It’s that Claude interprets the input data and turns a wall of technical data into an order of operations you can start working through today.
2. Competitor analysis
Next, I ask Claude to pull a competitor’s organic performance and compare it directly against mine, using Semrush’s domain and traffic data: monthly traffic, organic keyword count, traffic value, domain rank, and six-month growth trend.
In a recent comparison, Claude showed me a competitor pulling over 80,000 monthly visits against roughly 11,000 of mine, with 16,347 organic keywords compared to my 3,149. The part that actually impacts my strategy is their top pages, ranked by traffic, with a note on whether I already cover that topic, have a weaker version, or haven’t touched it at all.
That “not covered” column is where new content ideas come from, and it’s a much faster way to spot content gaps than clicking through a competitor’s top pages one by one in Semrush.
3. Keyword research
This is the step most people think of as “separate” from strategy. I don’t treat it that way anymore. Keyword research should be part of a larger prompt, with the goal of building an AI SEO content strategy tied to your business goals, not just a list sitting in a spreadsheet.
Here’s how I do it:
I ask Claude to pull keyword data through Semrush’s Keyword Magic Tool and Keyword Overview reports, then group results into clusters based on volume, difficulty, search intent, and how well they support what I’m already trying to rank for. I also ask Claude to flag emerging or “exploding” keywords, terms where the trend line is climbing fast but competition hasn’t caught up yet.
Below, you can see keyword suggestions for one of the content clusters on my website focused on Semrush. I share my hands-on experience using different Semrush toolkits and compare them with other SEO tools I’ve tested. Since I monetize these articles with affiliate links, they generate a solid stream of income for my website each month.
Claude is also aware of my goal to reach $100k in annual revenue by the end of 2026, so the keyword recommendations for this Semrush cluster, and the other content clusters on my site, are designed to help support that objective.
Overall, these are the main content clusters on my website for which Claude generated keyword suggestions:
4. SEO content plan
Once Claude has the technical audit, competitor gaps, and keyword clusters in the same conversation, I ask it to turn all of that into a content plan, built around my actual publishing capacity. I can only create 1-2 new posts per month and update up to 3 posts a month. That’s my limit considering my demanding full-time job and other duties.
As per my request, Claude maps out a quarter-by-quarter plan, including existing pages, which new posts to create around which keyword cluster, and the reasoning behind the sequencing.
My Q1 plan, for example, was update-heavy, because updated pages tend to re-rank within weeks while new posts take three to six months to gain traction. Prioritizing updates meant traffic and revenue gains arrived faster, with new posts published alongside them so they had room to mature by Q2.
5. Content production
The last step is turning the plan into actual published content, and this is where I still do the most hands-on work myself.
Claude handles the supporting pieces inside the same conversation: structure, keyword placement, image suggestions, on-page optimization notes, and FAQ generation. But my job is creating a rough draft itself, the part that needs my actual experience, opinions, and voice.
To facilitate my writing work, I have skills built for every article type that I create. For example, I have a skill for tool comparison posts that I write as well as listicles. This makes my job way faster and easier since Claude doesn’t need to guess the outline, headlines, and the post format. It’s already and it mostly has to fill out the gaps based on my rough draft and instructions.
Before this workflow, I used to spend 10 to 12 hours per content piece on my site, even on topics I know inside out. After this workflow, I’m at four to five hours for a high-quality, deeply personalized piece. That’s roughly half the time per article, with the same quality output.
AI can speed up everything around the writing. It should not replace the writing itself, especially if your differentiator is first-person experience and real data, which mine is.
How much does this workflow cost?
I use the Claude Pro plan, which costs $200 per year. I upgraded because it gives me more usage capacity and access to features like custom connectors (including Google Search Console), unlimited projects, Claude Code, Claude Cowork, and more.
For what you get, I think it’s reasonably priced, especially compared to some other AI SEO tools. For example, I previously paid $249/month for the Writesonic AI SEO platform, which felt far too expensive for the value it provided.
Google Search Console is free to use and can be connected to any LLM with a custom connector. Here’s the GSC to Claude connection guide that I used. The process is a little technical, but it’s definitely worth the effort.
I also use the Semrush One plan, which costs $199/month and includes both the SEO and AI Visibility toolkits. It’s not cheap, but I’d recommend trying the 14-day free trial before deciding whether it’s a good fit for your workflow.
With Semrush One, you get 50,000 API units per month for LLM connections like Claude, and Semrush pulls from that balance every time Claude queries your account.
In my experience, a single keyword research session through the connector consumed 2,440 of my units, leaving me with 47,560. Building a full strategy afterward, including competitor analysis, a technical audit, and the 12-month content plan, consumed another 7,190 units in that same session.
Credits reset every month, which helps. But I’d recommend keeping manual approval enabled for anything beyond keyword research, organic research, and domain overview reports. That way, you can see exactly what’s using your API balance before Claude runs more advanced tasks.
If you’re curious about the exact usage, you can find more details on unit consumption in Semrush’s official MCP documentation.
What I learned from using this AI SEO content strategy generator workflow
Creating the strategy isn’t the hard part anymore. Training your LLM is.
Once I connected Semrush MCP and Google Search Console to Claude, tasks like technical audits, competitor research, and keyword analysis became much faster. The real challenge shifted to teaching Claude everything about me.
If your LLM doesn’t know you, your writing style, your SEO process, and the tools you use, the output will feel generic. That’s why I rely heavily on my Claude projects and skills instead of starting a brand-new chat every time. The more context Claude has about me and my business, the better the results.
Real data beats prompting.
The reason this workflow works so well is that it uses real data.
Keyword ideas come directly from Keyword Magic Tool. Technical recommendations come from Semrush Site Audit. Indexing and performance insights come from Google Search Console.
Instead of asking Claude to guess what my SEO strategy should look like, I ask it to analyze data I already trust and turn that information into actionable recommendations. In my experience, that produces far better results than relying on prompts alone.
One person can now do work that used to require a team.
Not long ago, building a solid SEO strategy required hours of manual work from an experienced SEO specialist or several people working together.
Today, with the right setup, a solo content creator can accomplish much of that work in just a few hours.
AI tools can’t replace SEO expertise, but they make it possible to scale that expertise much more efficiently.
Who this workflow is for
This AI content strategy for SEO workflow isn’t for everyone.
But I believe it will be particularly helpful for:
- Solo website owners who don’t have the time to manually audit their site and create a new content strategy every few months.
- SEO content strategists and small agency teams managing multiple websites where a fully manual process simply doesn’t scale.
- SEO agency owners looking for a repeatable, AI-assisted workflow to onboard clients and build SEO strategies more efficiently.
Personally, I use this workflow in three situations: when I build a new topical cluster around a seed keyword, update my website’s SEO strategy each quarter, or onboarding a new client and creating a content plan for them.
If you’re interested in the foundation behind this process, including audience research, topic clustering, and topical authority building, I covered that framework in my topical authority guide before AI became part of my workflow.
FAQ
What is an SEO content strategy with AI?
It uses LLMs like Claude connected to live SEO data sources, such as Semrush and Google Search Console, to automate the research, analysis, and planning stages of a content roadmap, instead of pulling reports from multiple tools manually.
Is an AI SEO content strategy generator accurate?
I’d say it depends entirely on the data feeding it. A generic AI prompt with no data connection will produce generic output. When Claude is connected to Semrush MCP and GSC, every recommendation comes from real keyword volumes, real audit results, and real indexing data.
Do I need a paid Semrush plan to use Semrush MCP?
Yes. You need an active subscription, such as Semrush One, SEO Pro, Guru, or a Trends API plan, plus a Semrush API key from your account settings. The app is free to connect inside Claude, but it draws from your existing API unit balance.
Can Claude really build a full SEO content strategy on its own?
Claude can build a solid strategy only when it’s connected to your real performance data. It cannot replace the judgment calls around business goals, audience understanding, and the actual writing. I treat Claude as the strategist’s assistant, not the strategist.
Join the newsletter for actionable SEO tips, case studies, and a behind-the-scenes look at my affiliate marketing journey.
No spam. Unsubscribe anytime.
3 Responses
Great read. Such valuable insights. I also appreciate the effort it took to compile this information. Many thanks.
Such a helpful guide, Victoria! Just shared it with my team 🙂 Have a great Christmas, and hopefully I’ll see you back on Medium. (I took a break this year but may jump back on in 2026!)
Thank you, Scott! 🙂