Are you using Google Analytics in the most effective way possible?
Don’t worry! If you’re newer to Google Analytics or aren’t sure if you’re getting the most out of it, this episode is for you.
Google Analytics consultant Kyle Akerman explains how to utilize the tool in order to better understand and optimize your content.
In this episode, you’ll learn:
- The Google Analytics alterations you should set up from the get-go
- The importance of segmenting data (and what type of segments to look at)
- How to set up appropriate Goals
- Mistakes people make when determining the ROI of their marketing tactics
- Google Analytics
- What is Google Tag Manager and why use it? The truth about Google Tag Manager. [blog post]
Amanda: This week I am pleased to welcome to the show Google Analytics consultant, Kyle Akerman. Welcome, Kyle.
Kyle: Hi, Amanda, how are you?
Amanda: I'm good. How are you doing?
Kyle: I'm doing alright, as alright as one can be these days.
Amanda: Exactly. Yeah. It is definitely a kind of a bleak time but hopefully, whether you're listening to this now or later, it will be a helpful episode for you if you work in content marketing, because it is all about Google Analytics. And if you're like a Google Analytics expert, this might not be the episode for you. If you're somebody who has dabbled in it, and maybe you're not getting everything you can out of it, I think this episode will be really valuable. So, Kyle, also, I want to shout out to Kyle's newsletter, or his email list because I love it. He emails, tips about Google Analytics that are really helpful and we're going to cover some of the things he's talked about in it. But if you want more information about this, like, if you're listening to this, and you're like, "I want to know more.", definitely check that out and I'll include a link in the show notes.
Kyle: Yeah, that's awesome. Thank you for the shout out. Yeah, I've only been doing the newsletter since January. So, it's been something I wanted to do for a couple years. So, I'm finally glad I started doing it and that people are finding it valuable.
Amanda: Yeah. I really liked the format that you did it in, and I highly recommend it. So, check it out. So, before we dive in anything at all, for Google Analytics, when you access it, hypothetically, for the first time, I'm sure a lot of people listening have already accessed it, but what do you have to modify? Like, what do you need to do in Google Analytics to make sure that everything you do afterwards is as accurate and clean as possible?
Kyle: Yeah, I mean, there's a handful of things, but just a couple ones that I recommend are, you definitely want to set up some filters, and especially one for your internal traffic. So, if you have a so, like, say Fractl, if you have a bunch of people in an office, on a shared IP address, they might be visiting your site pretty regularly. So, you're going to want to filter that traffic out of Google Analytics, because that's not your desired audience. So, you would set up a filter that removes traffic from a specific IP address. The biggest amount of traffic I've seen is, I've seen some companies that had upwards of 30% of their traffic was from employees. So, that's on the high side. A lot of companies, it's not that high and you don't necessarily have to worry about it, but just trying to keep your data clean, you want to remove your internal employees. A second thing that you need to set up is conversion goals, because Google doesn't know what success looks like for your website so, you have to tell it. And the way we do that is by setting up conversion goals, a very simple example would be, maybe someone signs up for your newsletter and after they fill out the form, they go to a thank you page. So, you would measure as a goal, people that get to that thank you page. Because otherwise, you just-- you don't know, you know how your website's performing.
Another similar example is maybe somebody requests a quote, download a PDF, or an E book. So, these are all things that you could set up to measure as goals. And these goals should be aligned to your business and they can be macro or micro goals. So, that's just something to think about. Oh, and then another big one. It's not necessarily within Google Analytics, but if you send a lot of emails that link back to your website, Google doesn't know that that traffic came from an email click, they only know-- they can only identify traffic that comes from another website, or a search engine, or maybe Google paid ad. So, if you're sending an email with links to your site, you need to add some tracking code to it, it's called UTM codes. And it's basically a code that gets added to the end of the link that you're sharing and that, when someone clicks that link, it tells Google, "Hey, this was from an email.", because otherwise, all your email clicks will show up in the direct traffic bucket, which is kind of a black hole in Google Analytics, because it's kind of like a miscellaneous bucket. So, it contains things like email clicks that weren't tagged, clicks in an instant message, maybe I clicked from a PDF, you know, things that just google can't identify or a bookmark, or somebody typed, you know, your company name into the URL browser bar. So, I would say those are the three things, filters, goals, and then set up your email clicks.
Amanda: Yeah, I'm really glad you brought that up because I remember when I first set up Google Analytics, and I was playing around with it for the first time, I'm like, what is the deal with all this direct traffic? Like, they're clearly not typing in these URLs, like, what are all of these different sources that can be communicated as direct traffic, right? It's like, super intuitive.
Kyle: And a couple other ways direct trafficking show up is, if there's a third party platform that your site interacts with, so maybe say like PayPal. So, if you send someone over to PayPal, and then they come back, well, that part where they come back is going to show up as direct unless you do some additional kind of configurations. So, it can be things like third party platforms, and you know, things like that. So, yeah, if you see a high percentage of direct traffic, you may want to dig into what's really happening.
Amanda: And I'm glad you brought up the concept that Google Analytics doesn't inherently know what success looks like for your site, and it sounds like, before you even get started with this, you have to have a clear understanding of what exactly are you trying to measure with this tool? And how does it apply to your overall marketing goals, right?
Kyle: Right. And yeah, you definitely need a plan and that plan is going to contain like, the major questions that you want to answer and the answers to those questions will help you, you know, grow your revenue or leads or you know, whatever your, you know, kind of main metrics are. You know, so, a lot of times you might ask questions about like, the specific traffic source that you're using, you know, so if you're putting a lot of money into paid ads, you want to know which ads are performing the best, or if you do invest a lot in email, you know, you want to know how your email is doing, you know, that kind of thing. So, those are some very basic questions, but you need to know what you want to measure and then you can start going from there.
Amanda: And I remember you mentioned something about the importance of not just looking at data in the aggregate. Can you talk about why that's important?
Kyle: Yeah. Because so, most platforms, and I mean, we're talking here specifically about Google Analytics, but they tend to aggregate all of your data. So, it's all together, it's, and a lot of the data they give you is like averages. So, the real insights are kind of hidden in those numbers. You know, so, for example, you know, if you're looking-- if you're just looking at an, a number for the total number of visits you had to your website, you know, like this week or something, that's great. But you probably want to know, just, you know, the breakdown of that. So, how many paid ad visits did I get? How many Facebook visits did I get? How many organic searches? You know, if you're creating a lot of content, so digging down into the subsets of that data is what you eventually want to do and that can kind of come out of your question, you know, like, a starting question might be, how much traffic did we get to our site this month? I mean, that's a decent question. But a better question might be, how much organic traffic did we get to our site? And then an even more specific question might be, how much organic traffic did we get to our site with this specific blog post? So, you know, you kind of narrow down your focus to get better insights as to what is happening on your site, because a lot of those insights can be hidden, you know, kind of in the noise of the data, if you will.
Amanda: And can the same be said about segmentation? So, taking all this information about the type of traffic you're getting, but then narrowing it down by like, who exactly it is who's looking at your content?
Kyle: Yeah, that's kind of what we're, yeah, that's kind of what we're even talking about, you know, what I was just talking about, so you know, organic traffic back, that can be your segment. Segments can be very simple. So, like that one, it's just one thing; it's organic traffic. Or I could make a segment, you know, organic traffic from mobile devices that landed on a specific landing point. So, like, you can make your segments very specific or very broad and again, a lot of it is just, you know, what are questions you're trying to answer, and then finding that balance between granularity and, you know, kind of just the general numbers.
Amanda: Yeah, it sounds like those first questions are really important to figure out how you're going to set everything up in Google Analytics to see what you want to see, and to get more specific sometimes.
Kyle: Exactly, and a lot of times too, the questions and then the answers to the questions, they need to be actionable. So, you know, if you get the results and the answers of a specific question, but it doesn't drive any kind of new action, then you might have to get a little more specific with your question. You know, so like one that I have discussions with people all the time over, time spent on a page, and I'm not a huge fan of that metric, because I'm like, what do you do with that? Like, how do you change your marketing based off of how much time someone spends on a page? You know, so like, I really like to focus on outcomes whenever possible, and those can be like the goals that we talked about. Now, one instance where time might be useful is if you're just, if you're selling ads on sites, because then the more time people are on a page, then maybe the more likely they are to see that ad and then that brings you money, and I guess just an overarching thing with any of this analytic stuff is, if you look at your results, if all you do is like smile or frown, and then you move on to the next task, then you really don't need to measure anything. Like, the whole point of this stuff is that you'll ask a question, we'll find the answer, based on that answer, you'll change what you're doing or double down on what you're doing, if it's successful,
Amanda: Right, yeah, that makes total sense. And it's not just for confirmation, it's for action, and you should have an idea of what you'll do based on the data that you collect and analyze.
Kyle: And I mean, it's very common that people don't really take any action.
Amanda: Yep, I believe it. I think, I mean, it's tricky, this is the way to set everything up and start looking at things, but I think one of the things that people struggle with the most or is like multi-channel attribution, seeing how everything fits together. Do you have any tips on how to approach that in the most efficient way?
Kyle: I mean, it's tricky. I think everybody is still trying to crack that nut. I mean, I think at a very high level, just know, start with everywhere you're promoting your content or where people could find it, and then, are you, first of all, are you able to measure all those different, you know, pathways to your content? I think that's the first thing. And then every company might have a different way to weight each of those steps. So, you know, one that's pretty common is, you know, people use paid ads. So, is the paid ad, is it, you know, you want to know if it's more useful for that top of funnel awareness. Like, that's one case where the ad might be contributing, or maybe it's getting people that are very close to buying to push them over the edge to buy, or to call you or request a demo, or whatever it is. So, I think just knowing if something is maybe at the top of the funnel or the bottom of the funnel is useful, understanding that, you know, we have multiple steps to a lot of conversions, and especially if you're-- I think it differs between B and B2C. But if you're B2B, you're probably going to be a lot more touches for conversion. You know, because B2B is typically a higher value sale, it might even be multiple people have to agree, you know, on to spend, you know, thousands of dollars on whatever the thing is that you're offering.
There are some attribution reports in Google Analytics, and they allow you-- you can actually, you know, they give you some default models. And like, that's a good place to start, you can also go in and kind of create your own, that gets a little more advanced. One thing to realize is, when you're looking at a report, so say like a traffic report, and then if you've got goals set up, when you look across from the traffic sources, you know, like organic search, you might look all the way across to the conversions and it'll say, you know, it'll give you a conversion rate and how many completions that it's attributing to organic search. But the way that report works is, all it's doing, it's basically giving 100% of the credit to the last non direct click. Which basically means if you had multiple steps, so, let's see if I can think of an example. Maybe I came to your site from Facebook, and then I came back later from an email and then I came back a third time from organic, and then I filled out the request for a quote, then that request for a quote is going to be attributed to the last step in the organic search. Like, what about those other two steps, like they were important too, right? So, the main, in the end acquisition section, the main traffic report, it can be a little misleading but in the conversions report, and I know we don't have, you know, this is an audio but if somebody's looking at their conversions report in Google Analytics, you'll see a couple reports in there for, there's one called top path and then there's one called like attribution. And if you start clicking around in those, you'll see the actual number of visits that contributed to conversion and the traffic source. So, it starts to give you a better picture of, "Oh, okay, it takes somebody three to five steps before they sign up for my live demo of my software.", or whatever it is. So, attribution is not easy.
Amanda: Yeah. Well, I think, you know, people listening are all in content. I think that's a really tricky part of it, because you make the great point that knowing what the goal is to the content if it's top of the funnel, bottom of the funnel, you can kind of get at, especially if it's bottom of the funnel, are people converting from it? But when it's top, it's often kind of left behind in terms of attribution if somebody gets there from organic and then leaves and comes back another time.
Kyle: A simple one that might help people know how their content is performing is just, you can look at so, you could create a segment for just organic traffic. So, you're just looking at that. And then you could look at a specific content piece, and over time, say over like six months or something, and see it, you know, is the organic visits to that piece, is it going up? Is it going down? Is it staying steady? That'll give you a sense of just how it's performing.
Amanda: Right. Is there a certain segments or goal that you think people tend to overlook?
Kyle: One that I think people might not always look at is just the breakdown between mobile and desktop. Because there's a lot of-- the behavior can be very different on those two segments. So, at a very high level, usually people that are visiting on a desktop, they're either at home or at work. So, that's a different environment. People on mobile could be commuting, they could be waiting in line somewhere. So, usually those two cases, they usually have different intent so, maybe somebody in mobile is just going to kind of quickly browse your blog content, you know, kind of scroll through some of that stuff, and maybe never look at any of your service pages, product pages, what we call the money pages. Whereas desktop people, they're probably more likely to look at those money pages. Perhaps you know, fill out a, if your form is long to sign up for, you know, I don't know, to register for something or sign up for a demo or. So, I think looking at those two segments and just looking at the pages that people look at, in each of those, to determine if their intent and behavior is different, and then look at the conversions between those two. You know, you may have very low conversions on mobile devices, it could be one of two things, it could be that they're just not in the mindset to convert when they're, you know, commuting or whatever. Or it might mean you have something wrong with your site because think about the screen sizes you're dealing with. If you have a form that you want somebody to fill out, maybe it's not rendering great in the mobile device. And so, if somebody wanted to fill out the form, they looked at it, they started to fill it out, and then they're just like, "Yeah, forget it.".
Kyle: No, those are-- again, those are, that might be a question you ask, you might say, you know, "How is my-- how are my conversion rates on mobile?", and then go take a look, because if you're doing like a responsive design, maybe it's not rendering correctly in mobile. So, I think that's a good segment. Oh, another one that people might be missing is just looking at visits by location, and this is especially important if you're doing regional campaigns, or if you only service a specific area. Now you might want to know, you know, which, so, maybe you're doing all Midwestern states. Well, you might want to know which one of those states performs better, or where are you, you know, where you could make some improvement. So, location doesn't always tell you anything interesting, but it can. So, I think that's another one.
Amanda: Yeah, I think those are great segmentation examples because like you said, in the aggregate, neither of those might really give you anything interesting, it's only when you dig down, like if you looked at mobile and desktop together, but you saw that on desktop, you're getting almost all of your conversions compared to mobile. I mean, that's a huge difference compared to looking at like, "It performs okay.", you know? Maybe you're killing it on one and can use some help in the other.
Kyle: Right, and you may not, and like I said too, it might just be the intent of the person. So, you're may not be able to, you know, if they're looking at a specific page on mobile, they may never be ready to take that, you know, action conversion, because they're, you know, they're just commuting on the train, they're just kind of looking at your site, you know, they're just, you know, so, it doesn't necessarily mean something is broken. It just might give you more information about kind of your user persona, you know, if you want to get as granular as, "Okay, what is my mobile user-- what are they thinking about when they look at, you know, these specific content page?".
Amanda: Absolutely. So, similar question is, what kind of-- what do you think are like the biggest mistakes people make when they're looking in Google Analytics, and they're trying to determine their ROI?
Kyle: I think sometimes people make incorrect decision or they make decisions based off of limited data. So, you want to make sure that you have as much data as you can get, you know, the larger-- this is getting a little into statistics. But if you can have a larger data set, a larger sample size, that'll usually be more indicative of the true data versus a small sample size. The statistics part of it is, if you have a small sample size, the variability of the data is a lot higher; that's about as simply as I can express it in this. So, I think collecting as much data as you can, let's see what else. Just, you know, maybe not filtering out specific things or segmenting out specific things, you know, which we kind of talked about, you know, so make sure that whatever you're looking at doesn't contain your employee traffic, that kind of thing. Yeah, I'm kind of blanking out right now on other ones.
Amanda: No, I think those are good. I just like asking those types of questions just because if they're really common, you know.
Kyle: Oh, yeah, here's another one, keeping in mind seasonality. So, you know, don't assume that your business doesn't have any kind of seasonality. Like, that should just be a basic question. You know, do we suffer from seasonality? Because it might be okay, that your traffic and conversions dropped in June, you know, because maybe you're selling winter coats like, well, yeah, nobody's going to buy that in June. You know, so, I think just knowing your business, but there could be things that you don't realize so. And okay, so, yeah, here's one, not only just getting a bigger sample size, but looking over at longer time frame. and they kind of, you know, help you do both. You need to see where there's peaks and valleys in your data, and it can also help you find out if maybe something is now broken in your setup. Because if you look at a longer time frame, you might see where your numbers were really high, and then all of a sudden, there was a sharp decline, that could be an indication of a setup issue. Like, it might not necessarily be a problem with, you know, your marketing or people not showing up to your site, you might have an implementation issue. So, I think always looking at a wider data set, and then a narrower one, you know, when it makes sense to, just keep it in perspective, the whole time frame, like wide and narrow.
Amanda: Yeah, it seems like all of this kind of reflects that the numbers and the data themselves are not going to tell the whole story, without their context, like you have to bring that perspective too, using this tool in the first place.
Kyle: Yeah. Another one, I don't know if this is necessarily a mistake, but sometimes I come across people that obsess about bounce rate; that's another one. Kind of like, time on page but, it helps you a little bit. Like, you could look at like maybe the-- well, first of all, bounce rate, like at a site level, that's a useless metric. Because we need that-- because that's what we were talking about before; that's an aggregated metric. You know, so, if I say that, you know, fractl.com, I don't know if that's the URL. But if the bounce rate for the entire site is 75, what does that tell me? Like, that doesn't tell me anything. But if you tell me that, you know, the bounce rate to, I don't know, the service, the content marketing service page is 90%. Well, that starts to give me a little more information because that's one specific page. So, then in that case, bounce rate can be a little more useful and then not getting hung up on, especially bounce rates from blog posts. Blog posts usually have high bounce rates because people come to your site, they read the post, and then they leave. Like, that's not necessarily a bad thing.
Kyle: You know?
Amanda: Yeah, I know these debates happen a lot on Twitter, it's like, well, what's the point of the content like you just said, like, do you think that people are going to come to this, especially if it's top of the funnel and then just like go to every part of your site? Probably not, right? So, the bounce rate doesn't matter as much and like same with time on site I would think, if you wrote this, like very comprehensive guide that you're hoping people actually read, you might look at that to see if the content is interesting enough, and maybe you should update it. But if they're coming to get an answer to a question that you provide very quickly, and they spend like two seconds, that's again, like, that's not a bad thing. You just have to know what you're kind of aiming for otherwise.
Kyle: Yeah, if you create, like a long guide that, you know, is on a page, if you really want to know, you wouldn't necessarily have to look at time, you could add in tracking to measure how far down they scroll. That might be more useful, you know, because if you see that a lot of people are scrolling all the way to the bottom, then maybe they're, you know, consuming it.
Amanda: Yeah, that's a great point.
Kyle: So there, yeah, there's a lot of other, I mean, there's a ton of things that you can measure in analytics, but they require additional setup. So, scrolling is one of those things, things like button clicks, any kind of video interaction so, if you have a lot of videos on your site, and if you want to know if people are watching them, that requires additional setup. So, those are usually called, those are called Events in Google Analytics, and it basically involves anything that is not a page view. So, because really, Google only tracks movement between pages, it doesn't really track actions on a specific page. So, if I click a button, or if I click play on the video, or if I click if you have tabs, like tabs that you know, extend content on your site, none of that is measured by default in Google Analytics. But if those are important things for you to know, and to measure your user behavior, then you can set those up. Typically, you would, the easiest way to do that is with Google Tag Manager and that's a whole other tool to discuss; that's a big rabbit hole, but it's a good thing for people to learn even if it's Just at a very high level.
Amanda: Yeah, I think that's a great point. I'm glad you mentioned it. Because even if we can't talk about it now, it's something that people can dive into if they're like, "Oh, yeah, I do a lot of videos, and I'm not really tracking that very well.".
Kyle: Yeah. And if you're interested in learning more about Google Tag Manager, there's Amanda Gant at Orbit Media, she wrote a great article, just kind of explaining, at a very high level, what Tag Manager is and how you can use it.
Amanda: Cool. Yeah. I'll be sure to include that link in the show notes.
Kyle: Yeah, no, it's a good one. It's a good starting point. Yeah.
Amanda: Awesome. So, Kyle, knowing the objective of the show, do you have recommendations on who should be guests in the future?
Kyle: Yeah, I was thinking about, since you're kind of looking at, seems like just all types of content people in general, I got a couple ideas for you. I don't know, have you covered email yet?
Amanda: Not a ton. No.
Kyle: Okay. So, if you want to talk to an email expert, there's a woman named Jessica Best, she goes by Jess. She is a great email expert, she's based out of Kansas City, she's presented a lot at Content Marketing World so, I think she would be awesome. If you want to pursue kind of content research and how to think about doing that and setting that up, Michelle Linn.
Amanda: Yeah, I had Michelle Lynn on like the first iteration of this show years ago. Yeah, Michelle's great.
Kyle: Well, yeah, you may want to call her back up because she's now, she's no longer with CMI but she's doing, I think her company is called Mantis Research. I forget her in the business partner, but they help companies do content related, how to research and then publish original research as content.
Amanda: Yeah, we chatted, I don't remember, I think it was maybe the Content Marketing World conference last year we met in person because we have very similar philosophies on data focused content and such but, yeah, that's a good idea.
Kyle: I got one more for you.
Amanda: All right, Bring it on.
Kyle: You know Aaron Orendorff?
Kyle: I don't know if I have his name spelled right, I'll send it to you. But I think it's O-H-R-E-N-D-O-R-F. Well, he worked at Shopify for a while, now he works at a company called Common Thread Collective. But he is great about talking about like e-commerce type content, and has done a lot of great work, he's also spoken at Content Marketing World. So, he's another one to consider.
Amanda: Awesome. No, I really appreciate it because those are definitely some topics we haven't covered and I'm trying to hit every angle.
Kyle: Yes. Well, when I was looking at what you had covered, I was like, "Ooh, I think these might be new."
Amanda: No, you actually did an excellent job.
Kyle: Alright, I feel good.
Amanda: Thank you. Thank you, Kyle, for coming on the show and taking the time to share your suggestions.
Kyle: Yeah, thank you for the invite. It's always great talking with fellow content marketers.