Thread by @micahfk

GA4 is an on-coming disaster.

@googleanalytics I say this with a heavy heart from a place of love.

Let's dive into:

Part I
Who the heck am I and my history

Part II
Why #GA4 has a problem

Part III
If not rectified, what I expect is going to happen

Before I begin, for clarity, I'm generally reticent to critique on Twitter, mostly because I find it unproductive (for myself) to complain about small stuff.

So, long threads when warranted (and time away from gaming) broken into parts so one can quickly jump to interesting bits

PART I: WHO & HISTORY

Most folks that follow me are aware that I'm an SEO, what I don't highlight is a finely detailed path of how my work in web analytics followed along.

It's not my thing to push myself, but for this thread, I will to establish credentials and seriousness.

So, history.

Back in 2007, I started my career as an #SEO specifically, an Associate Search Strategist (or an A.S.S. for short) in an interactive agency called Red Bricks Media (R.I.P.).

There was no real tool for us poor SEOs. Paid had a tool, Email had a tool, etc.

I looked around and saw the power that Analytics could have for SEOs and dived right in with it too.

And yep, Google Analytics was the easy choice. Simple, powerful, and darn well inviting!

Guess what? I pushed that tool to every company I could get it on.

These weren't small companies either. Red Bricks had a knack for getting big companies.

I built the Web Analytics practice at RBM. I went in and trained marketers at SanDisk. Heck, I learned regex because of GA!

When GA launched 3.0, I wrote a blog post about being a game changer and was told that article went far and wide within the team there.

web.archive.org/web/2008122105…

They deserved every praise they got. I'll get into how that differs today further down.

I got a signed Web Analytics 2.0 book from @avinash which I cannot find for the life of me, probably lent it to someone and forgot.

Really annoyed I can't find it; am a book hoarder. I still have my HotMetaLPRO 4.0 book on the shelf (bottom shelf mind you) to advance my HTML.

I established my hatred? - let's say strong dislike - of SiteCatalyst. But I still had passion to do an audit of AMEX's Omniture web analytics system.

That passion to this day has me repeatedly poking the @googlesearchc folks that they should just merge into GA. #sorrynotsorry

I say all this because you know what GA did? It lowered the bar for entry into analytics and gave me the ability to ramp up and become more advanced.

When people ask me what kind of SEO I am today, I say Analytical SEO all because of Google Analytics.

That's the power it had.

Now, I didn't go into the field of Web Analytics; SEO is truly my passion.

I am not on any betas, no inside knowledge of what's going on today, etc.

But I did run a team of Web Analysts at Balsam Brands and managed Zendesk's GA when we didn't have actual web analysts. Counts?

And again, I say all this to make sure it's clear I'm not here to complain for the sake of complaining. This is me giving a stark warning that GA4 has been going down the wrong path.

Cool? Alright, let's talk issues.

PART II: GA4 ISSUES

Going to break this apart between:
-Analyzing from an SEO mindset
-Using the tool (UI/UX basically)

I'll try to focus on the big stuff I've stumbled across, but all the minor things do add up. Also, won't be fully comprehensive, there's just too much.

What's generally the first thing I want to know for how SEO is doing?

Besides which keywords converted!
(Look, I know GA still has the tab for us, but Google isn't giving that back.)

A: What landing pages are doing well.

So, how do we get that? Let's compare current vs new.

Current: Couple of annoying clicks, annoying drop-down pick, done

Now: traffic acquisition, find pencil meaning edit because you can't add a second dimension, add the dimension, remember to set as default, add second dimension of original source/medium, set organic/google, done

Yet, there are better ways to do this, but it's not there by default for a starting SEO analysis.

You know what is default?

GA4. I'll get back to that.

Okay, so you got a list of landing pages, what do you do next?

Search for your set!

And if you're a bit above basic, you'll throw in some regex.

Uh... well... there's always Explore

And I know that data is available. I'll get back to that too.

Okay, so let's say I want to keep that search subset, what's next?

Current: Just Add Segment on that report, Advanced Conditions, set landing page, do regex

GA4: Custom Report? Need to do that to keep the report, but then not Add Comparison because added dimension isn't there.

Sure, you can go build in Explore, but, now you're out of the report with a blank screen.

Think someone will remember what you need to set up to get the right card easily?

And you're staring at a smorgasbord of variables and tab settings. I'll get back to that.

It's at this point you test out Engagement and find that Pages and screens has the pages and source/medium you wanted in the first place.

The terminology from the current GA will lead you to the wrong sections or create other struggles.

I'll get back to that as well.

This is just one analysis path example btw.

I could go further with quickly viewing the line graph by day, week, month (you can't), jumping down dimensions (you can't), etc.

However, it's time to get to the /crux/ of the matter: using with GA4.

And that compounds things.

It's frustrating enough not finding what you need, but having it be difficult or break while not finding what you want?

It's the reason I continue to recommend against GA4 today.

To play off my old article: GA4 will be game ending.

Onward to UX.

Twitter thread breaks/limits, so please continue here:

https://twitter.com/micahfk/status/1461413788226256897