Use Data Or Be Used By Data!

The December 15 issue of Seotistics is here for you!

LLMs are everywhere and they have some interesting applications for our daily tasks in Analytics.

Yeah, they won't make you lose your job as the news claim...

I already discussed about LLMs in this past issue. Now we talk about other possible use cases and consequences.

There are too many things to say before 2026!

​Read this in your browser​

📣📣 Important Announcements 📣📣

Nothing new apart from my newly published article about Obsidian and LLMs for creating your personal knowledge base 😎

Preparing more content for 2026 now!

The Analytics Shift

Some years ago LLMs were cool but far from being used at work, at least for me.

Well, this is different now!

We have dedicated LLMs for coding and research and they make our lives much easier.

From visualization to diagramming and even generating mockups or playing with data, LLMs can do all.

I don't see the point of denying that these are essential tools for Analytics now.

I struggled with Javascript for years and now I feel more comfortable as I can write it more quickly.

The same goes for my infinite SQL code.

Oh, and I recently shared an example with my content production activity, Obsidian + Claude:

This is exactly what I predicted and said many many times to my dear readers 👀

The so-called Domain Knowledge makes a HUGE difference in the projects I worked on.

Unfortunately some outdated companies still require you to do live coding without LLMs...

needless to say, skip those companies, you won't actually work like that, we are not in school.

Back in 2020, companies were overpaying people with Python and SQL skills.

Anyone with a Udemy degree could get into a role and label themselves as Data Scientist.

This is no longer true (fortunately).

After millions wasted on pointless projects, companies have developed cost aversion and hire conservatively.

Specifically to Web Analytics, before BigQuery was a rarity and those who knew where heavy on code.

Now LLMs can make you focus on strategy and less on execution like before.

You still need to have marketing understanding, which is also but common based on what I read on LinkedIn. 😅

Capacity Multiplier

I was literally spending hours on non-value adding tasks like coding and now this is just a memory.

Now I can get my work done much faster and dedicate time on framing problems or building systems.

It's also true that LLMs will not make your work less because of Jevon's Paradox:

So you will be more productive but forget about delegating all to machines, it won't happen.

The Best Tools Out There

There are many LLMs or tools but so far this is what I feel comfortable recommending:

• Claude Code / Gemini CLI (terminal rules)
• Cursor or any other IDE

Everything else is bloat. It doesn't help you perform better.

I've recently tested Gemini CLI to retrieve BigQuery data in my terminal and ask for some advice:

I have to say the results are quite good! My only complaint with Gemini CLI is that it's slow.

For all the rest I rely on Claude because the performance is so good.

And you know what, you could even call Gemini CLI inside Claude Code 👀

This is often done to:

• increase your context window
• additional insights
• save tokens

What I like about LLMs is that they reward creativity and force you to think of possible solutions.

The picture above is a very simplistic example to explain it.

I will share more about this topic in my next posts, it's a useful application!

OK... But Google?

GA4 and GSC have added AI capabilities, but they're surface-level.

Asking questions in natural language doesn't fix the fundamental problems:

• Data sampling in general and partial data
• Attribution models that don't match reality
• Heavy limits on the tools that give you unusable data
• Missing storage of the data

This is why I still recommend using GA4 only for setup and basic concepts.

Then, exporting everything to BigQuery where you control the analysis.

I have talked many times about the fact that Google is doing all but useful stuff for us.

This is the foundations of data, you can't do any serious work if you don't store your data.

But... there is an actual great use of LLMs inside BigQuery, Gemini!

Yes, you can write queries with the aid of Gemini and this is what I use for debugging or when I feel lazy.

Areas Where I Am Not Fully Convinced

A lot of bad takes have been said about LLMs and here are some of the worst:

• AI will replace dashboards
• Analysts will be outdated
• Content is no longer needed

These are superficial takes from people that have 0 experience.

Actual dashboards are super complex and not something you can easily replicate with LLMs.

In fact, the problem is NOT visualizing things but all the underlying logic non-technies ignore.

This is why analysts won't disappear, our job is to optimize decisions, not writing code.

Content is so outdated that people still share blog posts and read a LOT.

At the start of 2025, you have been hearing wrong takes like the one above and many more.

Guess what, LLMs use grounding:

So they need external sources and most often it's websites and text material!

I am actually doubling down on my website and ignoring all the noise.

Websites matter even more than before if you have been creating proper content.

If you need to hone your thinking muscle and marketing knowledge, look no further:

Become a Top Data Analyst Now!

OK... Guess I Need LLMs

Yes, no reason to go against what works.

For sure, don't become overreliant on them, many problems we face require super simple solutions.

For example, a chatbot won't fix the issue of people not working with data...

a proper data product and onboarding will.

Don't expect LLMs to be that magic wand you needed in your life, it just won't happen.

👥 Join Our Community

Our Discord community offers a small place where we can talk business and web data.

If you hate all the noise of social media, then this place is for you.

I will start posting more there as we have a forum channel now.

This is the best way to stay updated in real time on Seotistics:

Join our Discord Community!

🔎 Analytics For SEO Ebook - Course / Ebook

You will:

✅ Use GSC and GA4 Data to their fullest potential

✅ Learn Python/SQL for your needs

✅ Get a complete blueprint for auditing websites

✅ Learn how to 10x your productivity

✅ Learn BigQuery to work on large websites

I teach you what's needed to go from 0 to a professional Data Analyst.

Even if you leave SEO, the foundations are the same for other jobs!

Learn Analytics for SEO Now!

Also in ebook:

Unlock Powerful Insights Now!

📚 Recommended Reads - Peak Content 🗻

Read these peaks:

• ​Content Management & Knowledge Base With Obsidian & LLMs​
• ​Insights from My First Google Search Live in Zurich by Simone de Palma
• ​How to Audit E-E-A-T at Scale with Screaming Frog and OpenAI API by Ian Sorin
• ​The sin, the oil, the cracks, and the crisis by Juliana Jackson
• ​The Most Useful, Timeless Skill to Learn as a Data Professional by Ergest Xheblati
• ​Analytical Skills for AI and Data Science : Building Skills for an AI-Driven enterprise​
• ​Data Literacy In Practice​
• ​Becoming a Data Head​
• ​Storytelling with Data​

As usual, my most recent LinkedIn content is here.

❗️ Feedback and Recommendations

If you have ideas/recommendations for the next issues of Seotistics, you can simply reply to this email.

Marco Giordano
​Data/Web Analyst

Follow me on 🔽🔽🔽:

Bernerstrasse Süd 169, Zurich, Switzerland
​Unsubscribe · Preferences​