In this episode of In-Ear Insights, Katie and Chris talk about the most overlooked part of any migration, but especially a migration for Google Analytics 4: your people. What are the skills gaps that exist? How bad are they? What should you be doing to mitigate those skills gaps and how strongly will they impact the success of your migration? Tune in to find out!
Subscribe To This Show!
If you’re not already subscribed to In-Ear Insights, get set up now!
Advertisement: Data Science 101 for Marketers
Do you want to understand data science better as a marketer? Would you like to learn whether it’s the right choice for your career? Do you need to know how to manage data science employees and vendors? Take the Data Science 101 workshop from Trust Insights.
In this 90-minute on-demand workshop, learn what data science is, why it matters to marketers, and how to embark on your marketing data science journey. You’ll learn:
- How to build a KPI map
- How to analyze and explore Google Analytics data
- How to construct a valid hypothesis
- Basics of centrality, distribution, regression, and clustering
- Essential soft skills
- How to hire data science professionals or agencies
The course comes with the video, audio recording, PDF of the slides, automated transcript, example KPI map, and sample workbook with data.
Get your copy by clicking here or visiting TrustInsights.ai/datascience101
Sponsor This Show!
Are you struggling to reach the right audiences? Trust Insights offers sponsorships in our newsletters, podcasts, and media properties to help your brand be seen and heard by the right people. Our media properties reach almost 100,000 people every week, from the In Ear Insights podcast to the Almost Timely and In the Headlights newsletters. Reach out to us today to learn more.
Watch the video here:
Can’t see anything? Watch it on YouTube here.
Listen to the audio here:
Machine-Generated Transcript
What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for listening to the episode.
Christopher Penn 0:17
In this week’s In-Ear Insights, July 1 2023, is going to be a big day for a lot of marketers, because it is the day that Google Analytics three, also known as Universal analytics, the web analytics software that 90 plus percent of the web uses will stop working, Google has said that they will be sunsetting it and that Universal analytics accounts will no longer collect data as of that date.
And then we will have access to legacy data for another six months.
And then after that, it’s a big question mark as to whether we’ll even have access to our old data.
So in this today’s episode, we want to talk to the people side of things because as we’ve talked about, in previous episodes of the podcast, the technology and the processes, you know, there’s a lot of focus on that in this week’s live stream, we’re gonna do a live Google Analytics for migration.
So we’ve got the platform of people process platform pretty well covered.
But the challenge that I’m seeing and hearing is the people in our Slack group analytics for marketers, as well as other slack groups, we’re hearing people saying, I don’t know how I’m going to train myself, my team get speed with everything else that’s going on right now.
And do it in time, Google didn’t give us enough time, etc, even though the products been out in the market for 18 months.
Okay, when you hear that Google has essentially opposes arbitrary date some 15 months in the future? And whether we like it or not, we are going to do it.
What do you what’s your answer? If people were saying, Hey, we, we just don’t have the people to do this?
Katie Robbert 1:53
Well, um, I would say, first and foremost, if you want to go ahead, go ahead and have like a mini panic attack.
Go ahead, have it feel your feelings? You know, don’t follow it up like I you know, I was writing the newsletter.
And one of the things I’ve mentioned was, I am not the most technical person.
And so I’m grateful, Chris, that I have someone like you on my team.
So that No, in all seriousness, when announcements like this happen, I personally don’t need to panic as much as someone who doesn’t have a Chris Penn level skill set on their team, because I would otherwise in previous roles, I would be panicking because I know my way around Google Analytics.
I’m familiar with Google Analytics, three, very familiar.
I’m modestly familiar with Tag Manager.
And then I feel pretty confident with Data Studio.
So I have a decent enough skill set to get somebody set up and up and running with Google Analytics three, with Google Analytics for I know, having worked with you on audits and setups and migrations, like it’s a whole different world, and it really is a developer’s software, it’s no longer just sort of the drag and drop super easy, intuitive.
It’s none of that exists.
And so the answer that I have is go ahead and panic.
That’s fine.
But then once you get the panic and out of the way, we need to take a step back and go, Okay, we have to figure out what we’re going to do about getting our people trained.
We talked about this a few weeks ago, Chris with general life cycles, and why life cycles don’t include people.
That’s the part that’s always skipped over.
And then so you’re left with this thing that nobody knows how to execute, because it wasn’t factored into the plan.
So here’s the good news, you have time to factor people into the plan.
So I would say as of you know, the second you hear this podcast, the second you see the news, both Google start making the plan for getting somebody on your team trained up, whether it’s you or somebody else, or you bring in a contractor, nominate your champion, basically, this person who is going to learn Google Analytics for inside out because that needs to happen in order to set yourself up for success.
So that’s where I would start is figuring out what skills you have on the team currently.
And what you need.
So one of the things we know about Google Analytics three to four is that both systems rely on Tag Manager.
Both systems want you to be using Data Studio as the reporting interface.
So if there’s a comfort level with Tag Manager and Data Studio, you’re already a good chunk of the way there.
Christopher Penn 4:41
Okay.
One of the things that’s happened here, that’s going to be difficult even for people who are familiar with Data Studio is understanding that the architecture of GA four is so different, that even like your existing data studio dashboards will not work, as is you have to read You build them.
And the metrics have changed, which is, you know, something of an interesting challenge.
I’ll give an example.
In GA three, we’re used to seeing source medium, right? What’s the source of this session? Or this user, etc.
And there’s three of them now, in GA four, there’s first user source, the session source and those conversion source.
And it’s not really well documented anywhere.
which one to use? And when? How do you, for people who are saying, okay, cool, we’re going to skill up our team on this? What’s the approach because people are going to jump in, say, Okay, I think I know Data Studio, I’m going to go in there.
And then you’re presented with your 210 dimensions and metrics, which by the way, is missing 300 of them, because there’s 520 of them in GA three.
They’re gonna go and say, Okay, I’m ready to start rebuilding reports and just run face first into this, what do we do here?
Katie Robbert 6:03
Well, like any good project, I would recommend you start with your user stories, and, excuse me business requirements.
And so, you know, for some people, they’re going to be building brand new Data Studio dashboards, because they’re like, Oh, I’ve never had to do this before, I’ve always just looked at the data inside Google Analytics three, which is fine, but it’s not available to you in Google Analytics for so you have to be using Data Studio.
So I would start with your user story.
So you know, the basic structure of the user story is, as a person, I want to take an action so that I can get an outcome.
And so, you know, a simple one is, as a marketing manager, I want to know how many people are visiting my website, so that I can see that my campaigns are successful.
Or as a marketing manager, I want to know where people are coming from when they come to my website, so that I can resource my campaigns appropriately.
So those are some pretty common user stories.
So that’s how you would start to build out your dashboard.
So now that you know, the purpose of your dashboard with your users, so you can start to go into the more details of like, what metrics do I need, I probably need source and medium, I probably need number of sessions to my website, I probably need date controls.
And so you can go down the list of the actual requirements that you need in order to build the dashboard.
And so that’s where I would start to keep everything organized to see Do I even have my system set up correctly to be able to answer this question?
Christopher Penn 7:37
That makes a lot of sense.
The one thing that I think is going to be problematic for a few folks is the fact that if we think of this, like a kitchen, right, you know, what the menu is, the user story is essentially, in some ways, the recipe, right? It’s, it’s how you’re going to get to the outcome that you want.
Once you’ve decided what the outcomes are.
The appliances are different, because GA four is a different beast.
The people are the same.
But the ingredients now are different.
Right? Again, the difference between source medium and GA three, and then you first use a source session source and conversion source and GA for when the ingredients change.
How much time do people need to invest in getting to, to understand, oh, yeah, we don’t use wheat anymore.
Now we’re using tapioca for everything.
And having noticed here like what tapioca does, for example.
So in the GA, for example, you’ve got 60% of the dimensions and metrics are probably new, even if they have they share similar names, Watson, GMP, they function differently?
Katie Robbert 8:36
Well, this goes into the, you know, nominating, or, you know, finding someone who can train up and really understand the new Google Analytics for in depth.
And so one of the biggest differences that I’ve learned from listening to you speak about it, Chris, is that in Google Analytics, four, you cannot edit the channel groupings.
So the reason why this is important is because in Google Analytics, three out of the box, the channel groupings, the emails, social, direct, those things are not set up correctly, such that Google cannot recognize its own Gmail system as email, it comes in as direct traffic or unknown.
And so in Google Analytics three, we can take some steps to correct this in the settings in Google Analytics for that setting does not exist.
So whether or not they’re going to roll it out.
They haven’t said, this means that there needs to be a whole new training around governance around UTM codes.
UTM codes are going to be the way in which you can determine source and medium in Google Analytics for so it’s something you should have been doing for Google Analytics three, but if you haven’t, and you’ve been relying on the system settings, now it becomes even more important as you’re trying to fight for your budget to say my social media campaign is actually working.
Well guess what Google analytics for unless it’s tagged correctly, nobody’s gonna know.
And then when you bring that those metrics into Data Studio, that’s where you’re really going to see nothing’s tagged correctly.
Christopher Penn 10:12
Yep, yeah, the governance issues, I think are gonna bite a lot of people very hard because it is less flexible, is less accommodating.
Google said, here’s the way to do it.
And if you want the product to work as designed, you got to do it this way.
Which is, again, not entirely unreasonable, because the vast majority of us are not paying for this.
So you do get what you pay for.
But it is challenging for people who want to even be able to do apples to apples comparisons with GA three, like if you look at the two systems, you’re gonna get different numbers, because the way things are measured, delineated, are different.
Ga four is based entirely on events, and Google aggregates and rolls up events together to form things like sessions, for example, as a result, you could look at GA three and J, four for the exact same website, and you’re gonna see different numbers, even basins like how many visitors you have today, they will actually be different numbers sometimes.
Katie Robbert 11:10
And so one of the things that we’re recommending for teams to be doing right now, if they haven’t already done it, is to at the very least stand up a Google Analytics, for instance, even if it’s not 100% configured, with goals and all of those things, you can start to be collecting your Google Analytics three, and your Google Analytics four data in parallel, so that you’re seeing the delta between the two systems and working now instead of 15 months from now, to correct those when it’s too late.
Christopher Penn 11:45
Especially if you need historical data, because when you move to GA for data doesn’t go with you.
Katie Robbert 11:52
So Chris, uh, this is something I probably could have looked it up myself, but I haven’t yet.
No, in all seriousness, it’s something I’ve been curious about how long has Google Analytics universal Google Analytics three been around? So how, like, how many years could some companies have in historical data that they’re about to lose?
Christopher Penn 12:11
Since 2006?
Katie Robbert 12:16
So that is what 2010? I don’t know, math.
16 years? It’s a lot of years.
Christopher Penn 12:25
Yeah, like my personal websites, data goes back to 2007.
Katie Robbert 12:28
Okay, so using your personal website as the example, you have 15 years of historical data, you can do really deep dives, in your data of the analysis, and whatever you want to do, as of July 1 2023, unless you have exported all of that data somewhere into a sequel, big query table, God forbid, it’s in, you know, 15 years of data is in an Excel spreadsheet, but maybe you don’t get a lot of traffic.
So it’s not a big deal.
We don’t know.
Unless you have actively taken that data out of the system.
You’re basically you’re going to be up a creek without a paddle because Google’s gonna say, Okay, bye, bye historical data.
Now is so that’s a correct statement, right?
Christopher Penn 13:16
That’s a correct statement, Google has said they will want for you will have access to GA three data, your old today for at least six months after the shut off date.
They said they will announce, you know what the file cutoff period will be for the historical data.
But in the docket and the technical documentation, they do say we strongly recommend exporting all of your raw data.
They also don’t tell you how to do that.
Katie Robbert 13:41
Yeah, the question I would then have, and I don’t know if this is something that they consider it or that you considered is, can you import the data back into Google Analytics for So you then have that complete data set? Or is that basically saying like, Okay, if you’ve set up a BigQuery table, which is what we at Google Analytics for want you to do, you can take your Google Analytics, three data imported into BigQuery, and then start pushing all of your Google Analytics for data into BigQuery.
And then maybe if you’re lucky, if all the metrics match up, you could have a complete dataset.
Christopher Penn 14:19
You can’t unfortunately, because the schema for the GA for data is totally different than for GA three, like, you know, again, GA four is based on the event model.
So every single row and entry in there is event based, whereas GA three is sort of hit based.
They’re very, very different.
Structurally, the data itself is very different.
One of the things we were discussing this morning on analytics for marketers was that there will probably be a small cottage industry of people building an exporter that can take Google Analytics, the data and push to an open source package like matomo, like because matomo just has a MySQL database on the backend.
So if you’re able to get all your data out GA three format correctly for matomo You could then essentially loaded into the MySQL table there.
And then the matomo interface would allow you to essentially view your historical data as though it was current.
Well, it’s one of the other reasons why, you know, we’ve been saying never probably almost two years, you should have a backup analytic system running alongside it’s like, again, a Trust Insights, we’ve been using matomo on our site for two years now, I think.
And even though we hardly ever check it, other than, you know, I check it once a month to make sure it’s still there.
It’s there, and it will provide that continuity, you know, in case I don’t feel like exporting data.
Katie Robbert 15:34
But this goes back to the original question of the people.
So you and I have both been talking about all of these different processes that you can follow to get your data to do the thing.
That’s all well and good.
If you know how to do that, if you know how to export your data into a sequel system and re import it into a matomo.
If you even know how to set up a matomo.
If you even were aware that that was a thing that existed.
So this still comes back to what do you do about the people on your team who are panicking right now, who are super reliant on this data to make decisions to demonstrate, you know, effectiveness of their jobs, their campaigns, you know, to stakeholders, board members, customers, whoever? And so
Christopher Penn 16:23
I guess I would ask, I would ask you pretend I didn’t work here, what would you do?
Katie Robbert 16:30
Well, what? A quarter.
Yeah, after I’m done panic, getting a crying in the corner.
Um, you know, the first thing I would do is basically an audit, not so much an audit of the systems but an audit of the business in terms of who’s relying on Google Analytics data right now? What is it being used for, so that I have a good understanding of all the people and the teams who are going to be impacted by this change? So I would start there.
Now, the nice thing is that for Trust Insights, that’s you and me.
So it’s a pretty short audit, you know, I could probably do it on the back of a post it or something.
But for larger enterprise, companies, like some of our clients, that’s a bigger ask, because there’s so many different moving pieces that once a report gets created, they don’t necessarily have a clean trail of how many people have seen it, how many people are reliant on it? How many different decisions are being made by it.
So that’s where you need to start is number one, you need to understand who is using the data and for what purposes and what specific data is being used.
So that’s number one is sort of the ecosystem of the data itself, then you can start to go through, you know, if you need to go through report by report team by team and say, Does this still hold? Do you still need all of this data? Do you need different data? So you start to go through that requirements gathering, with each team for each report to understand what’s changed? Is this new is this different? Or has nobody looked at this report in six years, but for some reason, we’re still generating it.
So that’s where that’s the advice that I would give to myself in terms of, here’s where I would start.
One of the reasons I like to start there, not only from just an understanding everything is that it’s something I have control over.
And so for those of us who were, you know, higher anxiety, you need to find those places that you have a little bit of control over situations that are really out of your control.
And this is one way you can say, Okay, if I can collect all the information to say I understand all the pieces of this puzzle, I might not know how they all fit together, but I at least understand all the pieces of it.
Christopher Penn 18:50
Gotcha.
One thing that as you’re talking that occurred to me is that there’s also a interesting market opportunity here for marketing automation systems, because they also collect web analytics data, right? When you deploy Hubspot, or Marketo, or Pardot, or any of these things that you do have a widget that goes on your website.
So from a continuity perspective, if you’re doing apples to apples, those might become the systems of record.
Now, if if you know that you can’t carry over your historical data, that might be the easiest way for companies to, you know, sort of get once you know, a single source of truth, even if they lack things like attribution modeling, and stuff, you know, just for Hey, how many visitors do we have in 2016? If the data is in your marketing automation system, it might be a way to preserve that.
That creates the interesting challenge of then you’re locked into your marketing automation vendor.
Katie Robbert 19:43
That’s true.
And so that then again, it sort of goes back to the business requirements have what is it we’re what is the question we’re trying to answer? So if the only thing you care about is website visits, then relying on a different system other than Google Analytics might be fine, I would first make sure that they’re sort of a one to one in terms of the data you have now versus the data you’re going to get instead, and make sure that it’s a set a data set that you’re comfortable with.
But if it’s more in depth, if you need those attribution models, if you need, you know, to know, source medium, and the marketing automation system you have doesn’t provide that information for some reason, then that might not be the right choice for you.
So it always goes back to what are the questions that you’re trying to answer with the data versus, I just need to have a system in place because everybody has a system in place? You know, when they have a website, make sure you have the right system in place, Google Analytics might not be the system that you need anymore.
And that’s okay.
Christopher Penn 20:41
Yeah, I think that’s a really important point is that you, it’s not a foregone conclusion that everyone in the marketing world uses Google Analytics, there are advantages to using a well known system, you know, it’s easy to find talent.
You know, the talent may be cheaper, because there’s there’s more candidates, but it is not the only game in town.
And so part of the requirements can be, is this the time to make a change?
Katie Robbert 21:08
And if it is, so let’s say, Chris, you decide for your website after 15 years, Google Analytics isn’t the thing anymore, and you want to move to a system like Adobe analytics.
That’s great.
However, it’s not enough to just make the decision, you need a whole migration plan to say, these are the metrics that if you still have to do those requirements gathering, these are the metrics that I care about, these are the questions that I want to answer.
Here’s how I answer them currently, in Google Analytics, here’s how I’m going to answer them moving forward in this different system in Adobe analytics, or matomo, or something else.
And so you still need to do that requirements.
And then, you know, if you are staying with Google Analytics, you had mentioned, Chris, that all of the Data Studio dashboards are going to break, well have a good, good understanding of what that looks like, and how to test to make sure that things are working correctly when you reset them up.
So you need to be putting plans in place migration plans, test plans, training plans, it sounds like a lot.
And it is, but if you start now, it’s very doable.
Christopher Penn 22:20
What do you do about stakeholders and things? Who are, shall we say prone to procrastination? Oh, we don’t have anything in the budget for this year.
So you have to wait till 2022 adjoined to 23, I should say.
And so you know, it is March 18 and 2022.
If somebody says, Yeah, as soon as we can get your budget, do anything is January of next year.
And you have till July.
What do you do about those folks? Do you just say, I’m going to go up and LinkedIn profile because this companies do.
Katie Robbert 22:52
That’s one option.
However, the more likely option is to, you know, do that risk assessment, play out the scenarios.
And so helping the decision maker understand which decision you need them to make today and why.
So Chris, let’s say I came to you and said, Hey, Google just made this announcement, we need to upgrade to Google Analytics four, and you say, Sorry, kid, I don’t have time to do that until March of 2023.
It’s now incumbent upon me to say, alright, Chris, I hear what you’re saying.
I need to, I need this to be a higher priority.
Here’s why.
And then you start to list out sort of the different scenarios.
If we don’t do this until March of 2023.
Here is all the data we could potentially lose, here is all the money, it’s gonna cost us to expedite this for three months ahead of the deadline here.
If we do this now, here’s the cost savings, here are the resources that we could get, here’s the data integrity that we could achieve.
And so really helping them understand the full picture, versus just the urgent we need to do this now.
Well, there’s no budget, okay, by, like, that’s a terrible conversation, but usually how it goes.
Christopher Penn 24:06
So if you need supporting evidence, for this tip for your, for your stakeholders, you can tell them that the analytics from Trust Insights says that in 2023, prices for Google Analytics migrations will be double what they are this year, so
Katie Robbert 24:23
well, and that, you know, that’s the old you know, supply and demand.
And so people who procrastinate and wait until the last minute for anything, ended up paying more money than if they had planned ahead and done it before they became long lines of people who got on board to say, Oh, I also need to do this.
And so what’ll end up happening is if you start to wait until January, February, March, April of 23, firms like Trust Insights who do this kind of work will be completely booked up.
And then you’ll say but but but I need it now.
Great.
You have to add an extra zero to expedite it.
Sorry, but that’s exactly it, Chris, it’s gonna cost a heck of a lot more money to find the talent to find the resources to find the support to find the agencies.
And waiting that long doesn’t necessarily mean that Google is going to roll out any free resources either.
Christopher Penn 25:18
Exactly, I think probably on our to do list is we should be looking at writing code for exporting to a system like matomo.
So that as part of my migration, we can, we can help people out that way.
But it sounds like we need to get the people sorted out as quickly as possible, because there’s the part that always takes the longest.
And if we can do that, then the rest of the pieces should fall into place.
In a straightforward, if not easy way.
Katie Robbert 25:45
That’s exactly it.
So you know, as we’re recording this episode, it is mid March of 2022.
So we have a good 15 months before this, you know, gloom and doom deadline, ticker clock, you know, runs out, this is a great time to start finding those contractors to start finding those agencies to start finding those people on your team who have the skills and getting them ready for Google Analytics for now.
Of that deadline, the sooner the better.
Because again, the longer you wait, the more expensive resources are going to become if
Christopher Penn 26:21
you can get them at all, yes, anybody who wants to draw to make sourdough bread in April 2020.
how well that worked out well.
Katie Robbert 26:28
You know, even now, ahead of the announcement, finding reasonably priced resources, contractors who were skilled up in this arena, was still pretty difficult, because they’re very much in demand.
And now that Google has made the official announcement, it’s going to become even more difficult, you know, prices are going to skyrocket.
People will be demanding, you know, more money.
And the law.
Again, I don’t know if we can emphasize, emphasize this enough.
The longer you wait, the more expensive it’s going to be.
And the longer you then will have to wait in line for someone to become available to do this work for you.
Christopher Penn 27:10
Exactly.
So as you have for all things in the past for GDPR.
For CCPA.
For the for getting sourdough for getting wheat flour during the pandemic, whatever.
Whatever the deadline, is it you know, it’s coming.
Google has made it clear has drawn a line in the sand, you know, it’s going to happen.
Please do not wait.
Whether or not you engage us whether you’re not engaged an outside agency at all.
Please do not wait and get ahead of it.
You will save yourself so much time so much stress.
You will your stomach lining.
Well, thank you.
If you’ve got comments or questions about a Google Analytics for or any analytics pop on over to our free slack, go to trust insights.ai/analytics For markers, where we’re answering these questions at a very high volume, but you and your 2200 other marketers are asking and answering questions for each other all day long.
And wherever it is you watch or listen to this episode of the show.
If there’s a place you’d rather get it go to trust insights.ai/t AI podcast, and you can get the show on most other platforms.
Thanks for tuning in, and we’ll talk to you next time.
Need help with your marketing data and analytics? |
You might also enjoy: |
Get unique data, analysis, and perspectives on analytics, insights, machine learning, marketing, and AI in the weekly Trust Insights newsletter, Data in the Headlights. Subscribe now for free; new issues every Wednesday! |
Want to learn more about data, analytics, and insights? Subscribe to In-Ear Insights, the Trust Insights podcast, with new 10-minute or less episodes every week. |