Google Analytics Blog from Jeffalytics

This weekend I had the pleasure of speaking at the DMSS conference in Bali. This was the first year of DMSS, and the conference was an incredible success.

To me, the main success metric for a conference is the quality of ideas that you walk away with from the speeches you attend, plus how much knowledge you gain from the informal conversations you have in between sessions. The more excited you are to get back to work, the better the conference.

I have never been more excited to get back to work.

While I would love to spend an entire post talking about what I learned (from other speakers and various meet-ups), I wanted to put an analytics angle on the whole experience.

At the conference I gave two speeches; one focused on Google Analytics, and the other focused on PPC. You can probably guess where those topics originated. 

My first talk covered many of the more intermediate/advanced things you can do in Google Analytics, and the total talk lasted about 45 minutes, followed by some robust Q&A. Because of the time-frame, I couldn’t go as in-depth as my online courses. I could only hit the highlights.

“How many of you use Google Analytics?”

Early-on, I asked the audience to raise their hands if they used Google Analytics.

Only 25% of the room put a hand up.

Thinking on my feet, I said “the rest of you must just be too cool to raise your hands, because I assume you are using GA.”

It’s 2017, and these are Internet marketers. Surely they use Google Analytics. Right?

I don’t have an answer. But it has had my jet-lagged brain thinking about when and how to approach web analytics. At every stage of the online marketing spectrum.

When do analytics matter?

When you’re up on stage, you have to appeal to the majority of the audience to deliver a successful presentation. You can tailor your presentation based on demographics, industries, skill levels, or any other insights that the conference organizers can provide.

My presentation focused on intermediate and advanced topics within Google Analytics.

Now I find myself asking a simple question.

“Advanced for who?”

For the past seven years, I’ve shown a variation of this slide in my presentations.

Hundreds of times, on stages worldwide, I declared my LOVE of Google Analytics. No changes to that slide, because there has never been a reason; I still shout my love to the mountains.

Hundreds of presentations, receiving thousands of points of feedback.

The ratings and comments of these speeches are revealing. 

For the same exact speech, I may get feedback along the lines of:

“This speech was a firehose of information… I don’t understand any of it, I’m just getting started and I feel overwhelmed.”

Followed by another piece of feedback saying.

“This is so basic. I already knew everything. Jeff is a hack, I’m smarter than him. He sucks. And no, I will not leave my email address or identify myself. Baba-booey.”

Same exact speech. Two different audiences. Completely different interpretations. Fortunately, most other speech raters find it somewhere in the middle, which means I keep on getting invited to do more of these.

Why the discrepancy? Because attendees are at different phases of their journey to mastering online marketing, and their needs for analytics are very different.

Seeking a third dimension

Beginner. Advanced. It doesn’t matter.

What’s missing is context. Context around who is in the audience, what they are trying to achieve, and how long they have been trying to solve that particular problem.

While the beginner-intermediate-advanced paradigm serves as a nice anchor, I am considering a new way of approaching my presentations and teachings.

Start with a visual representation of the phases of an online business, and share what type of analytics and metrics are important at each level.

For example, bloggers trying to build an online business.

Many dreamers will make their first foray into making money online by creating a blog. They will be drawn in by dreams of riches, only to learn that beginner blogs are a dime a dozen.

The survivors decide to differentiate. Maybe they add an email newsletter sign-up button and experiment with emailing their audience. Or a “follow us on social” buttons. It brings traffic back to the site, and all is good in their world.

Once there are eyeballs on the blog, it’s time to try and make some money. They experiment with easy to implement/low quality ads. It works until it doesn’t. So they graduate to harder to implement/higher quality ads. It works better, but they want more.

Now that there’s a “profit” on the blog, the blogger wants to scale. So they start advertising, inexpensively at first. They find that there is no return on that investment, so they want to learn how to advertise more effectively.

It turns out that advertising works most effectively when you have a great product to sell. And when you have an automated sales-process for turning your ad-clickers into product-purchasers.

So they study funnels, marketing automation, direct-response marketers, and everything they can get their hands on. They study and implement. Study and implement. Soon revenues are rising, and the business is profitable. So profitable they don’t know what to do next.

Finally, they make a decision. They want more! More sales. More customers. More profits! They discover the only way to get there is through testing. Testing blog headlines. Email subscription boxes. Product pricing. Ad copy. Landing Pages. Funnel steps.

Everything they learned to this point comes full circle. They decide that a few well-executed tests can double their business overnight.

Not bad for a dreamer with a blog.

Building out analytics for every step of the journey. For every journey.

There are analytics for every single step of the story I just told. There are metrics that support blog post visits, email subscriber growth, revenue, advertising results, marketing funnels and testing.

Metrics to support each of these initiatives can be found in myriad tools. From Google Analytics to advertising platforms to blog software to email vendors.

Google Analytics is the closest thing that we have to a database of record. It is usually not the best tool at any phase of growth. But it is a solid backbone for analyzing any growth strategy.

Beginner. Intermediate. Advanced. That’s not be the best way to look at analytics.

Analytics exists to support all of the phases we go through in growing a business. It’s with us every step of the way.

Structure analytics around business phases, not complexity. 

The more I think through the problem, the more I believe in this solution.

What do you think? I would love to hear your thoughts in the comments.

P.S. The idea for this newsletter post came to me in the shower today. Sorry in advance for the visual, but if you’re reading this far you probably didn’t even notice. Oops, I just made it awkward. We’re over 1,000 words now, so I just wanted to make you laugh as a reward for reading this far.

This might be a new record for me. Last week’s newsletter, I wrote about Google’s new tracking code and expressed my skepticism.

The post was written on a Tuesday afternoon and scheduled to be sent out to you on Wednesday. As soon as I hit publish, my opinion on gtag.js had already started to change.

For this week’s newsletter I would like to explain what happened.

Start writing a blog post, get frustrated, tweet your frustration

While I was researching last week’s post, I couldn’t find answers to any of my pressing concerns. So I tweeted my frustration and thought nothing of it.

I went back to my hangry writing and scheduled everything for the next day.

When I woke up the next morning, a modern day hero from Google, Brian Kuhn, had offered to answer my questions.

He cleared up a lot of my fear, uncertainty and doubt (FUD), and made me more comfortable with the change. All of my comments from the blog post still stand, but I’m more optimistic about the new tracking method now than I was a week ago. Here’s why.

Gtag.js rollout was in response to changes in Safari going live in iOS 11

I am reading between the lines here, but it doesn’t seem like Google was ready to release gtag.js two weeks ago. However, they felt obligated to push it live to correspond with millions of people getting their new iPhones and upgrading their operating systems (iOS 11 was released on September 19th).

It seems like Google would have waited longer to release gtag.js if they had their way. But they were up against the clock, and documentation hasn’t quite caught up.

Their hand was forced (even with Apple’s initial announcement in June/July giving 3 months warning), and so a bunch of changes were pushed before being ready for prime-time. Unfortunately, Google only announced the changes with a pair of blog posts and a stealth update to the tracking code section of AdWords and Analytics.

My theory is that this change is also why Google finally rolled out the new AdWords interface to everyone on October 11th. This was a change they announced in May of 2016, but kept delaying the roll-out.

Think of Gtag.js as a lightweight Google Tag Manager

I have learned a lot about the new gtag.js over the past week.

1) For example, gtag.js does not replace analytics.js, it sits on top of it. So Universal Analytics isn’t going away or changing. It’s just getting a new coat of paint. Painfully, this means I need to put a new coat of paint on Analytics Course now. But I digress.

2) The datalayer in gtag.js does not conflict with GTM, and can coexist no problem.

3) gtag.js is considered by Google to be a lighter version of GTM. This means they are similar, but gtag.js is easier to set up. It’s a less technical version of GTM and makes core tracking super easy. One set of code on your site, declarations for AdWords and Analytics, and you are good to go. [IN THEORY.]

4) The Jury is still out! I am skeptical about gtag.js, but it is the future. It will get better as Google releases more support documentation and the analytics community embraces the change.

5) I also learned I should eat something before writing blog posts.

So that’s what has changed since I last wrote you. We’ll try not to let this newsletter turn into weekly updates on Google drama and product roll-outs, but hopefully you are finding these useful since it impacts everyone with an analytics account.

For the past few weeks I have been fielding a lot of questions about gtag.js on the Analytics Course forums. Students are asking why my recommended method of tracking sites is analytics.js, yet their account only allows them to install gtag.js on their site.

The same thing goes for Google AdWords conversion tracking. The only option is to install gtag.js.

What gives?

You may be wondering the same thing since gtag.js is the only script-based tracking option we see inside of our Google Analytics accounts.

Why, Jeff, are you recommending one thing, and Google is saying something different!

This tag was rolled out with little fanfare, so it’s natural to be confused about what is happening.

This post aims to clear up some of the most common questions.

I’m sure that gtag.js will be awesome, someday

To tell the truth, I don’t know much about gtag.js, because I don’t like trusting my advertising and marketing data to incomplete Google products. I have not used gtag.js, because it would be insane to use it at this point in time. This post will show you why.

Maybe I’m cautious after being burned by Google Wave, Google Buzz and the first time Google released Universal Analytics into the wild.

Or I’m damaged after moving too fast to Google’s “biggest and baddest” systems too soon in the past, and I don’t want to make the same mistake again.

Or maybe I’m just annoyed that Google is forcing this change upon us, yet not even telling their product team at GTM about the change.

One tag to rule them all

We’ve all heard this one before. Google wants to simplify user experience and make things easier for us. So they introduce a piece of tracking that is supposed to make things easier.

AND EVERYTHING GETS HARDER

They release an incomplete set of protocols and expect users to make the switch. They subject us to their half-baked ideas; standards that aren’t ready for prime-time.

Something that doesn’t even work with other Google products!

No offense to you, Gtag.js, but WHAT ABOUT GTM?

If you read any of Google’s marketing materials, it’s clear they think their users are not very smart. They undermine their intelligence with nearly every piece of marketing collateral, over-simplifying every step of the way. Dumbing things down, as if marketers and analysts don’t want the details.

And yet tools like Google Tag Manager gain mass adoption from users because we are dying for details. We want the benefits of a technically sound analytics implementation, without the slow cycles that come from putting everything in the hands of developers. Same goes for Google Analytics. It’s successful because it’s accessible to everyone.

There is no reason for gtag.js in a world where everyone uses Google Tag Manager. It is solving the same fundamental problem, only in a more limited way.

What about Google Tag Manager?

This post started when I tried to install conversion tracking for a new AdWords account. I created a conversion event and was given the following code to place on every page of my site. Plus another piece of code for the conversion event.

First of all, this is way more work than the old method of conversion tracking. But I’ll humor you, Google.

So I went into Google Tag Manager to get this code on the site, only to find nothing of consequence.

No mention of gtag.js anywhere. Just support for the “old” way of doing things. When I say “old,” I mean the ways that have worked for years, only to suffer a violent death in the past few weeks. Methods that are now impossible to utilize, since Google no longer displays the parameters you need to make AdWords Conversion Tracking work.

That makes this change not just annoying, but impossible. Google is requiring their advertisers to strip GTM from their site in order to track a Google AdWords conversion. I’m sure that wasn’t their intent, but it’s my reality.

They have changed their entire tracking process with no way to make it work, forcing an incomplete product on all of us without any support or documentation. C’mon Google, you’re better than this!

Fortunately, my account needs are simple. So I just imported my goals from Google Analytics and moved on. But there’s no way major advertisers can move this fast. Think about it this way: the internal Google Tag Manager team hasn’t even implemented this yet! And they likely knew about the change for years before it went live.

One swift Google product update made another Google product obsolete. At least until they implement a solution.

I still don’t know where gtag.js fits into the world of tracking

Is it replacing Google Tag Manager?

Is it already Google Tag Manager?

When will it be necessary to switch?

When will it support all of the great functionality found in analytics.js?

What will an installation look like if you want to track both AdWords and analytics in the same tag?

What benefit will it provide users over time?

Does it make us more reliant on Google? Or less?

If you have any thoughts, please leave a comment. I am as much of a student as a teacher on this one.

Update: After I wrote this article, I got some clarification from Google to my pressing questions. I’ve included them in the blog post and will keep on updating there as developments come in.