Use Data Or Be Used By Data!

The February 10 issue of Seotistics is here for you!

After a lot of issues about business and data logic, it's time to talk about some good ol' analysis.

The topics included here are what you should be familiar with, as those are common use cases.

Once you combine this with the previous issues of Seotistics, you will become unstoppable.

P.S. Now with AI you can code stuff yourself but if you want a deeper breakdown, check out my paid resources.

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

📣📣 Weekly Announcements 📣📣

Cool stuff next week, stay tuned!

I can't say more for the time being...

but you will be delighted!

The progress for the new course (Thinking like a Data Analyst, temp. name) is steady.

The preorders will be open as soon as possible, with a nice 30% launch discount.

Before We Start

Some of these simple analyses can be done in GA4... but you know I can't propose you this inaccurate and basic stuff.

Be sure to extract your data from BigQuery (or the APIs in the worst case) to create meaningful analyses.

If not, you will defeat the purpose of this issue which is to get the most out of your web data 👀

Some of these topics below I will also add to my existing course.

Cohort Analysis

This is a classic and one of the topics I've been studying the most recently.

It's so important because it allows you to work on retention and figure out what to do next...

as well as understand when people churn the most (or drop off).

Depending on your business model, how you structure it can change.

A cohort is a group of users with some common characteristics monitored over time.

The picture below is an example of a cohort:

Basic GA4 example from Explorations where Active Users are displayed inside the cells.

There are too many variants of this analysis based on your business model:

Content websites: Active Users to see how many people come back over time.

B2B SaaS: churn over time (you can use % values instead, start from 100%). In this case, the next time periods won't go above the original 100%.

Ecommerce: repeated purchases/revenue. Here's the catch, customers can spend MORE money in the future.

In terms of actions, you could target specific cohorts and offer them discounts or offers after a certain point in time.

Remember, past customers should be nurtured! This means thinking about special deals or specific plans to make them spend again.

I am going to add this topic to my "Analytics for SEO" course and this will be also featured in my upcoming product!

Good Ol' Content Auditing

This is a big one...

because the opportunities are endless.

The issue is that it often gets relegated to 3rd-party tools or checking 2 metrics in GA4 and calling it a day!

I still follow my good old template that is an evergreen:

As you know, I am a big on this topic, as I have shared a lot of content around it:

I recommend monitoring your content consistently and pay more attention every 3-4 months, if you manage a content-heavy website.

This is to prevent decay or "surprises" and avoid accumulating damage.

If you go beyond a simple audit and tie your recommendations to actions, you will notice that content distribution is the real GOAT.

For example, an article that receives a lot of Organic traffic from Google can be split down into 3-4 posts on social media.

Remember that not every piece of content can be distributed as is!

SEO content doesn't actually exist but... some content is purely created to rank.

E.g. How tall are penguins?

Classification & Query Analysis

Listing down all the pages and writing a recommendation for each is impossible unless you work with tiny websites.

I've been there in the past... don't follow this approach!

The best you can do is:

prioritizing the right pages
understanding which pages contribute the most to the revenue/traffic of the website
Same but with search queries

Traffic drops are harmless if vanity traffic is lost BUT remember that a minority of websites monetize with Display Ads.

An example of page classification with this nice schema.

In that case, more traffic is always good unless it's from lower-tier countries (extremely low RPMs).

The Case For Product/Category Pages

Products and categories aren't exactly "content" but they still deserve an audit!

Much like content, you shouldn't simply ignore or prune pages that don't have high traffic.

That's because:

a niche product can be high-margin
it's needed for catalog variety
you need it for price discrimination

Not what you usually read online, uh?

Yes, you may intentionally keep "mid" products so that people feel forced to buy a more expensive and better version.

If you simply look at the data, you'd remove the page and call it a day.

This is why I encourage critical thought.

Anyway, let's move to the workflow:

Nothing too fancy... be conservative in PLS (Product Line Simplification) and don't just propose drastic changes.

OK... But Over Time???

Some of you may have noticed that it's also important to monitor changes over time and not just at the end.

Plus, seasonality plays an important role in understanding how traffic is affected and for estimates.

This is why we often talk about Content Decay, an implicit part of every (decent) content audit:

The same goes for evergreen pages...

To summarize, these are some traffic-related metrics/KPIs you need to consider:

Risk is another important piece because having a lot of pages with 0 traffic is a risk.

After all, no traffic = no conversions!

Entity Analysis/SERP Scraping

It's no secret that Google gives you audience data and shows the search share of a market.

Sure, I don't really agree with the SERPs now... but it's invaluable information nonetheless.

Once you are at it, you can also extract the main entities in each page you scrape to get an idea of what are the main concepts.

Avoid the Google NLP API like the plague though, I recommend 3rd-party options because they are more accurate 👀

Mind you, I don't think entities per se are mandatory for ranking...

it's purely from a business/research perspective.

Recently, I haven't been doing much entity analysis because it can be expensive...

and scraping recently got harder due to Google requiring Javascript to scrape SERPs.

Still, search data IS audience data and you want to get your hands on that juicy info, like my friend Elias Dabbas shows with his visualization:

Elias Dabbas on X: "1/ Some #SERP analysis "learn <language>" "<language> course" for 20 languages in 4 countries: US, UK, CA, AU 160 queries (1,600 results) Data: https://t.co/wS2dsfWX5W Code: https://t.co/vN9WLodszt #DataScience #DataVisualization #

Market research is a big one for all the channels, it gives you ideas for content as a whole.

Once you go beyond the mainstream view of data, life will get easier.

If you want to learn more about these analyses + get the code + learn the winning frameworks, check out my course:

[Analytics For SEO Course - v4]

If you want to be guided and start from scratch, this is the course for you!

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!

Path Analysis (With Caution)

A classic and also available via GA4 Explorations.

This involves understanding how your users move across the website and where they go in general.

Are people actually going to money pages? If yes, from where?

From a CRO perspective, this is great to understand friction points in the navigation.

Many analysts look at engagement metrics (which I deem as almost useless) and give vague advice like "improve the experience on this page".

This doesn't work and it's based on nothing.

A quick example made in 1 minute... it needs a lot of refining to be useful. I recommend BigQuery for this, once again!

The issue with this analysis is that you need to spend some time on it to make it decent.

The example above can be worked a little bit more with specific breakdowns or even by checking which events happen next.

Instead, find friction points and then use a tool like Microsoft Clarity to understand where users often struggle.

You don't need crazy experiments, if a handful of users can't figure out your website, it's time to fix it!

Anomaly Detection (GA4 doesn't cut it)

GA4 offers you this feature and it's quite handy...

but if you want customization or more specific stuff, it's all on you!

There are different ways to detect anomalies on your website:

typical ML method (literally Anomaly Detection)
look for peaks
outliers

My advice for bot traffic is to remove those malicious agents who inflate specific days that show abnormal behavior.

Don't simply remove ALL the bot traffic you find and don't you ever think bot traffic can be prevented altogether.

Cope with the issue but don't overthink it.

Honorable Mentions

Market basket analysis allows you to group products that are usually purchased together, which is great!

You can create bundles based on this information...

The same goes for pages, ideally you could detect which pages are usually displayed in the same session to improve the navigation.

(This use case is remarkably less robust though, I warned you).

The other honorable mention is international cannibalization which is one of my favorite use cases... and I worked on it for years.

Imagine the USA subdomain/subfolder ranking instead of the correct UK subdomain/subfolder in the UK...

Yep, this is a big issue for conversions!

If done properly, you can find a LOT of opportunities across different local subdomains.

Funnel analysis is another interesting analysis to perform BUT I don't really believe in linear funnels.

People don't simply move from one stage to another in a linear fashion...

Still, it's available in GA4 and can be reproduced with BigQuery too.

P.S. If you ask, I can expand on all these topics but there must be demand for it.

👥 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.

Join our Discord Community!

🔎 Analytics For SEO Ebook (v8)

If you want to learn about Analytics for SEO and prefer self-study, this is for you!

It will teach you or your employees to:

👉 Prepare audits that make sense and are actionable 🔥

👉 Avoid common pitfalls that cost you money 💸

👉 Create analyses that add value and money💯

👉 Move the needle faster with efficient SEO systems ⏳

This comes with periodical updates to keep the content fresh.

Unlock Powerful Insights Now!

📚 Recommended Reads - Peak Content 🗻

Usual but solid advice:

My LinkedIn content:

❗️ 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

Seotistics - Web Analytics + Business + Strategy

🔎 The Analyses You Must Do On Websites