Free Tool

The Claude AI SEO Walkthrough: What I Actually Use It For

By John Akande · 8 min read · May 13, 2026

TL;DR

Claude is useful for schema generation, content brief creation, and title tag testing at scale
AI cannot replace competitive analysis, actual ranking data, or client-specific strategic judgment
The workflow that works: human strategy, AI execution of repeatable tasks, human review of output
Using AI for all SEO output without editorial judgment is how you produce content that ranks for nothing

What I actually use AI for in my SEO workflow

I run SEO engagements for US service businesses. I use Claude regularly. And I want to be honest about where it is genuinely useful and where it is not, because the hype about AI replacing SEO work is not grounded in how search actually works.

Here is my actual workflow, not the idealized version.

Schema markup generation

Work With John

Your site should be your best salesperson. If it is not, that is a fixable problem.

I work with US service businesses and B2B brands to build SEO systems that produce consistent, compounding leads. I will tell you exactly what is broken. No pitch.

Book a free strategy call

Schema markup is structured data code that tells Google explicitly what your content is about. Writing it manually is tedious and error-prone. I use Claude to generate valid JSON-LD schema from a brief description of the page.

The prompt I use: "Generate JSON-LD schema for a [LocalBusiness / ProfessionalService / Article / FAQPage]. The business name is [X], located at [address], serving [service area], offering [primary services]. Include all required properties and common recommended properties."

Claude produces clean, valid schema in seconds. I review it, add any client-specific details, and validate it in Google's Rich Results Test before deploying. This saves 20-30 minutes per page on schema implementation.

Content briefs at scale

A content brief defines what a piece of content needs to cover: the target keyword, the search intent, the key questions to answer, the recommended structure, and the competitive context. Writing briefs manually takes 45-60 minutes per piece.

I use Claude to draft the structure of a brief once I have the keyword data and competitive research. The prompt: "Create a content brief for an article targeting [keyword]. The target audience is [description]. The search intent is [informational/decision/local]. Key competitors ranking for this term cover [X, Y, Z]. The brief should include suggested H2 headings, key questions to answer, and a recommended word count."

Claude drafts the structure. I review and adjust based on client-specific context the AI cannot know. The brief goes to the writer with 80% of the structure already done.

Title tag testing

When I have a page that is ranking but underperforming on CTR, I need to test title tag variations. I generate 10-15 variations quickly with Claude, then select the best two or three to test.

The prompt: "The current title tag is [X]. The target keyword is [Y]. The page is ranking at position [Z] with a CTR of [%]. Generate 10 alternative title tags that are more compelling, match the search intent, and stay under 60 characters. Do not use clickbait. Do not use generic phrases like 'best' or 'top.'"

The AI produces variety fast. I evaluate each against the specific intent and pick the strongest candidates. Human judgment selects; AI generates options.

Where I do not use AI

Competitive analysis. I need real ranking data, real domain metrics, and real traffic estimates. AI cannot provide current data — its knowledge has a cutoff. I use Ahrefs and Search Console for this, not an AI.

Strategic prioritisation. Which of 40 identified technical issues to fix first is a judgment call based on the specific site, the specific competitive context, and the specific business goals. AI will give me a ranked list based on general principles. That is not the same as the specific judgment call a client is paying for.

Client-specific content. I know my clients' businesses, their voice, their specific client outcomes, and their positioning. Content that represents them accurately requires that context. AI writing without that context produces generic output that ranks poorly and converts worse.

The workflow that actually works

Human defines the strategy. AI executes repeatable tasks within that strategy. Human reviews and edits all AI output before it touches the live site or goes to a client.

This workflow saves me 6-8 hours per week on administrative and structural tasks. It does not replace the judgment, the competitive analysis, or the client-specific work. Those take the same amount of time they always did.

If someone tells you AI can replace the strategic layer of SEO, they have either not tried it or they are selling AI tools. The tactical layer is where AI adds time value. The strategic layer still requires a human who knows what they are doing.

Stay Sharp

Guides like this, twice a month.

No agency talking points. No keyword stuffing. Just the frameworks I use on paid engagements, free.

No spam. Unsubscribe any time.

Frequently asked questions

Google does not penalise content for being AI-generated. It penalises content for being low-quality, unhelpful, or lacking expertise signals. AI content that is accurate, specific, demonstrates first-hand knowledge, and serves the reader's intent performs the same as human-written content with those qualities. Generic AI output that says nothing specific performs poorly regardless of how it was generated.

For content and writing tasks, Claude and GPT-4 are the most capable general-purpose options. For SEO-specific research, traditional tools like Ahrefs, Semrush, and Search Console remain more accurate because they have current ranking data that AI models do not. The best workflow combines traditional SEO tools for data collection and AI for content production and structural tasks.

AI can draft articles that rank when: the human provides specific expertise and real-world examples that only they could supply, the AI structures and polishes the draft, and a human edits for accuracy and voice before publication. AI writing from scratch without expert input produces generic content that competes poorly in any established topic area. The expertise signal has to come from a human.

Paste the existing content and ask AI to: identify sections that could be more specific, suggest additional questions to answer that the content currently ignores, and propose a more compelling opening that better matches the search intent. Then review each suggestion and implement only the improvements that make the content genuinely more useful, not just longer.

It can, if you publish AI output without review. It does not have to. AI is a drafting tool. A rough human draft reviewed by a second person is better than either alone. AI drafts reviewed by an expert are often better than unassisted human drafts because the AI catches structural gaps and the human catches factual and strategic errors. Quality depends on the review process, not on whether AI was involved.

Competitive strategy decisions, keyword prioritisation based on current market data, client-specific positioning recommendations, and any content that relies on first-hand experience or specific client outcomes. These tasks require current information and specific context that AI cannot access. They are also the tasks that clients are actually paying for — the judgment layer, not the production layer.

Yes. AI models confidently produce plausible-sounding but incorrect ranking data, domain authority scores, and traffic estimates. Never use AI for competitive data — use Search Console for your own site data and Ahrefs or Semrush for competitor data. AI is useful for tasks where the output can be verified before use: schema code can be validated, title tags can be length-checked, content structure can be reviewed. Ranking data cannot be verified and should not come from an AI.

John Akande

SEO specialist and growth operator working with US service businesses and B2B brands. I build search infrastructure that produces compounding leads — not just rankings.

Book a free strategy call More guides →

← Back to all resources