Use Data Or Be Used By Data!

The October 20 issue of Seotistics is here for you!

The new year is close and many of you want to upskill.

Whether you are a marketer or a data professional, it's always good to have more skills, especially during a recession.

I'll show you what you can learn and how to avoid spending time on topics you will never touch.

P.S. I've just written an article on GA4 metrics (+ BigQuery calcs), your upskilling starts there!

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 your browser

📣📣 Important Announcements 📣📣

The big events are done, now it's time to focus again on my products, services and content production ❤️

For those who asked, I won't be at brightonSEO this week, I'll be most likely attending in April, though.

Expect more content until the end of 2025 👀

And prepare your wallet for my next product updates and Black Friday sales (starting November 27).

What To Focus On In 2026

The current trend in Web Analytics is to get closer to Data Engineering.

This is great advice but not really, at the same time.

The reasons are many:

overcrowding the already crowded Engineering space
forgetting about our job: making money
doing tasks that companies won't give you lol

Marketers who become too technical suffer from the same issue...

they try to do what they shouldn't.

I advocate for a hybrid approach where you can do the analysis yourself and still retain your domain knowledge.

Still, you may want to have a look at these topics because they are useful:

BigQuery

I always talk about it because it has changed the way Analytics is done!

Before BigQuery, it was a tragedy to have data in one place...

let's take Search Console:

different results based on your queries
heavy sampling

Now, you don't need to worry about this anymore.

GA4 is the cherry on top with its complex (yet useful) structure.

I know, GA4 is hard and y'all hate it...

but the BQ export makes a big difference.

Check my articles below, it's all there...

I can recall my past days when I worried about having the data for tomorrow in the right format.

Now, it's all in BQ and I just need to query the data, simple.

As detailed in my 2 articles on the topic:

Really, having all the web data in one place makes a BIG difference.

Absolutely prioritize this!

You can't do serious Analytics without a data warehouse/lake, forget about it.

DBT/Dataform

While Engineering is a separate discipline, this is a nice one to have.

I use both occasionally if there are specific tasks that require processing.

To keep it simple, these tools process your data in BigQuery and make them more usable.

DBT was recently bought by Fivetran... let's see what happens next.

Pick either and find some practical use cases, like:

transforming GSC data
unnesting GA4 data/breaking it down into multiple tables

Remember, you can simply use packages or solutions built by other people.

Learn what you need to carry out the job and you are done.

Recently, I've been learning and playing with Dataform.

It's not strictly required BUT there are too many situations where some knowledge makes your life easier.

Data Modeling

Yep, if we talk about DBT/Dataform, then we must mention the art and science of data modeling.

The tables we get are often in inusable or hostile formats for consumption.

So you need to understand how these tables should look like:

I often mention great tools such as GA4Dataform and PipedOut... but for custom use cases, you need practice.

Web Analysts have always been hybrid so some Engineering skills won't hurt you.

Expect a detailed article soon enough 👀

Coding (To A Certain Extent)

A lot of people ask me about coding in general...

because it's considered the hardest part.

Short answer: yes but not too much.

Analytics is NOT coding.

Analytics is NOT coding.

Analytics is NOT coding.

SQL is the 1st natural choice due to BigQuery (and because data is often stored in databases).

Claude Code helps MASSIVELY with coding and prototyping

Then, you can choose one between Python and Javascript.

I went for the 1st due to my background, but if you are obsessed with GTM and Dataform, I recommend the latter.

AI made it easier, you aren't even required to be as proficient as before.

Back then, you had to be more involved with coding and I was every day on Stack Overflow.

Now? I hardly check it, Claude is the best 💯

Our job is about analyzing data and communication, simple syntax can get you quite far.

SQL and Python are well-documented... AI can write entire scripts without many issues.

Some "veterans" consider this taboo because they are afraid of change.

It literally made me save hours per month and immensely boosted my productivity!

What You Don't Need

It's easier to tell you what NOT to study because I was in this position years ago.

The shiny topics are the worst ones:

AI (complex models)
the Machine Learning you study in universities

I was lucky because I spent my college years reading good advice on LinkedIn.

Deep Learning and the "complex" models are topics I completely skipped because many companies don't need them.

At least, in the Web industry!

Analytics is simple per se. But simple doesn't means easy!

The Web industry is a little bit quirkier but it's still boring and repetitive (great for making money).

Most of my work consists of:

framing business problems
deciding which data to use
running SQL to get insights (occasionally Python/R)
communicate them
prepare actions

See? No mention of strange models whatsoever.

You don't even need fancy degrees because it's all practice (and industry expertise).

Some of my connections went back to college to study topics that are outdated by now...

and no one really cares about the Web world in the academic world.

P.S. A good degree makes the difference (Engineering) but not the "cool ones" about Data Science lol

Lucky you, the basics will get you far, I saved you 1000s of $.

My Personal Experience

I started with a background in Business Administration and Marketing and then moved to Data/Computer Science.

So I'd say this is a great example of upskilling.

Back then, there was no reliable resource to learn from, it was pretty much all generic stuff and hyped content.

The most impactful skill is without a doubt understanding problems and marketing.

You can save so much money with the right mindset.

Don't start that project if the motivations aren't clear.

Avoid pruning half of your website because the "data suggests so".

This is what makes the difference with a mediocre data professional.

If you feel in doubt, remember this picture here:

You convert data (raw material) by giving it a meaning put into the right context that eventually leads to action.

It all depends on your understanding!

A Ready Roadmap

Like in RPGs, you can choose different paths.

While reality is more complex than binary choices, this is often true:

Seotistics is clearly on the left end of the spectrum as most of my content is about business.

You can follow the mainstream and become another half-engineer...

so the market will be saturated again and you will make even less.

Business acumen and having a clear mindset are underrated...

because many people won't even grasp them.

Yet, this is how you can make a concrete difference.

For SEOs, this is what's been working great: