The Seotistics newsletter is written by Marco Giordano, a Data/Web Analyst with the goal of combining business and web data. Tired of the usual boring Analytics content without any business impact? Seotistics teaches you how to use Analytics, web data and even content in your workflow while helping you with Strategy.
Share
An LLM Content Workflow That Works
Published 4 days agoΒ β’Β 6 min read
β
Use Data Or Be Used By Data!
The February 16 issue of Seotistics is here for you!
Automating content is not what they tell you...
you can still do it without sacrificing quality.
The price to pay is that you will manually review all and still work like a human.
If you ask me, this is ideal and how it should work.
Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by accident. Gmail users can read this tutorial to do it.
My new course about Analytics can be preordered in like 1-2 weeks from now.
Those who voted will receive special benefits.
This will be for Analysts or everyone wanting to explore Web Data in general, not only SEO.
The topics I am planning to cover are:
Coding in general (Python, SQL, R - all aided by LLMs)
LLMs and how to use them properly
Auditing websites and deeper business considerations (so not only SEO)
The topics covered in this newsletter
If you forgot to vote and want to be included in future communications, please contact me.
βοΈ I am also actively working on new content and topics for my dear audience.
After all, this issue is about content.
The Problem / End Goal
As you know I am a content creator too and creating new B2B content is extremely challenging.
If I were only producing tutorials, this would be trivial... it's not my case though.
Most of my content starts from atomic social media posts (LinkedIn).
Given this piece of info, I'd love to prototype and write faster for my personal website.
AI was supposed to replace content according to many, the opposite happened!
What I teach in my dedicated course and what you also need for life.
So... how do I even help myself to produce more?
The biggest people do with Content is to assume machines can write it.
βΌοΈAbsolutely not! LLMs should give you inspiration or drafts, don't hit publish!
I am surprised I have to explain that but not few people in SEO fall for this.
This is why if you want to learn either Content Marketing or even Analytics, you need to get practical.
SEO experience alone is the worst way to learn.
The Overview
What I did looks like this:
β
It's not overtly complex but we'll go over it.
The Data
The must have data to get here is:
your content (website, newsletter, owned assets)
3rd content (e.g. competitors' content)
It's that simple, really!
There are multiple ways to extract this content, based on your scenario and content maturity.
I prepared a list of websites to scrape and those act as the backbone for the content I miss (or what I deem interesting).
To inspire you, this is what you can scrape π:
Competitor websites (of course, you need to check content gaps)
documentation (e.g. Google Cloud docs)
social media content (incl. Reddit)
newsletters (golden content)
YouTube transcriptions (optional but broken)
The opportunities are endless and this is all text data, meaning LLMs excel at it.
Crawl4AI
In my case, I scraped some other websites with the aid of crawl4ai, a Python library that is my new baseline.
I've been recommending trafilatura for ages but now it's been dethroned π
Antigravity did pretty much all the job + Claude Code.
So in 2026 you don't need to be super technical, you can get away with a lot less... use this opportunity!
β
To run the scraping, I created a CLI app (aka you run it from the terminal).
This is the best way to build app for yourself, give it a try!
Content Storage
As discussed in many of my previous issues, Obsidian is my go-to for storing content.
You can create a vault aka a folder on your laptop containing .md files and whatever you like.
The usual picture I share to show you how my Obsidian looks like.
The tool is free if you don't want any synch feature or early access to updates.
In my use case, I was that lazy to use another folder inside Antigravity.
(Remember that to connect Claude to Obsidian you also need an MCP).
qmd
As covered in my previous issue, this is like having Google on your laptop.
The entire goal of this is to make easier for LLMs to understand your content.
Apart from that, it's a nice way to have a local search engine for free.
β
The only annoying part is that the conversion of your content can take a LOT of time.
You are converting content into machine-readable stuff, after all.
Claude + WordPress MCP
Now this is the cool part.
Claude or any LLM can actually interpret your .md files and give you advice or even write new content.
The thing is... Claude burns a lot of tokens for this simple task.
This is why we used qmd one step before!
I recently installed the WordPress MCP to have proper integration with my CMS, this was huge!
The steps are easier than you think:
From WordPress: Users > Profile > Scroll down to Application password and create one
Use the WordPress MCP Server from the connectors and put the password
Then, you can simply ask Claude to check your WordPress to extract content (and save it) or just search for topics.
Yep, one of my next articles, I show you how it looks like in Cowork.
It can also search inside drafts which is lovely.
My only complaint is that it's using WP's classic mode, whereas I use GeneratePress blocks.
Understandable, so remember to always adapt your content and inspect the generated code, if needed.
Output
Unlike most of the slop we read online, the goal of this automation isn't to spam content or mass publish.
This is B2B we are talking about, it must be original and strongly positioned.
I was drafting an article and noticed that Claude got every single number wrong.π
Well, depending what you ask, you get a different output. In this case it was a comparison with a published article (ignore the GEO crap proposed by Claude).
So humans like you should be deeply involved in the content creation process.
The real pros of this automation are the cost and flexibility to do it all.
A paid tool would ask several 100s just to generate some text.
The Business Impact
Well, for my specific case the impact is that you will get more cool articles to read :3
My goal this year is to scale my website and the Discord community.
Organic Search is still the superior way, social media are necessary for visibility/discoverability but have terrible algorithms.
I am already drafting like 5 articles that I hope to publish this month.
I will keep you posted because I wanna have a case study on Seotistics... β your own assets > any client.
You can take my framework to the next level and add the following:
check your queries from GSC + BWT to detect even more topics
check Reddit for topics
give instructions on how to write posts in WordPress
add a styling guide (this is actually did lol)
make it automatic and less "human-involved" like this one
I prefer to keep it simple and more "manual" so I can prompt Claude and get inspired.
The ideas are endless so before you venture into the unknown define what you need.
And if you want to think like a true analyst, this is the way:
The Seotistics newsletter is written by Marco Giordano, a Data/Web Analyst with the goal of combining business and web data. Tired of the usual boring Analytics content without any business impact? Seotistics teaches you how to use Analytics, web data and even content in your workflow while helping you with Strategy.
Read more from Seotistics - Web Analytics + Business + Strategy
Use Data Or Be Used By Data! The February 9 issue of Seotistics is here for you! Needless to say, 2026 started with a blast with more LLM news. I am here to show you why they matter and how I am using all of these new things. Well, not all of them are new but I will show you anyway! P.S. Most of them are about Claude because it's superior lol P.P.S. Yes, ChatGPT started testing ads in the USA, this is unrelated to Analytics though. Please move this email to your Primary inbox or reply to it....
Use Data Or Be Used By Data! The February 2 issue of Seotistics is here for you! Let's talk about career today. Seotistics is big on upskilling and understanding how professionals can improve and I want to tell you my key lessons. After 7 years as an SEO and around 5 years as a Data Analyst, this is my time to shine. I wanted to make this issue different so you can also go beyond data. Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by...
Use Data Or Be Used By Data! The January 26 issue of Seotistics is here for you! What if I told you that not using data can be better than data? Yes, because the goal is optimizing decisions, not using data! You probably heard someone sound smart with "data-driven" and other words... in reality it's harder than that. Please move this email to your Primary inbox or reply to it. This is to prevent Seotistics goes into spam by accident. Gmail users can read this tutorial to do it. Read this in...