#H2HChat Getting the Data Right with Justin Cutroni, Analytics Evangelist, Google

We're live.

Hey, how are you guys doing?

Hello.

We are so excited to be here on another HTH Chat on a Monday, which means that it's Monday at noon and very— on Pacific Standard Time, I should say, for those of you who are in California, which our guest is not. Our guest is actually in Vermont today. where he resides but also works at Google headquarters in Mountain View, I believe. And so very excited. I'm going to give our guest— and stop talking about him like he's not here— in a second. I will, I will do a proper introduction, but first I wanted to say hello to my co-host, Suzy. You can't miss her, she's the one with the red hair. How are you doing, Suzy?

Good. I have my new toy thanks to Jeremiah. one of our tweet chat guests. It's like awesome 3D printed thing. How

cool is that?

Right? Bryan has one too. I'm jealous.

I think I should have gone for red.

Telling you, he asked me what color.

Surprise!

And I was like, there's only one color I can do, right?

Well, I welcome you back. I know you were on vacation last week, and welcome back to another HA Chat and to have you back on HA Chat since we had some partnered HDA chats. And then to now introduce my good friend Justin, Justin Cutroni, sorry Justin, I should know. That's okay. I should say it like I'm Italian and then I'll get it right. But anyway, I met Justin in Italy. We were speaking together at a conference and Justin and I and a bunch of other people hit it off and we had such a great time talking about past, present, and future of analytics, and there is no better person than Justin to talk to, in my eyes, that I've ever talked to about this sort of thing. I'm really excited about today because at the essence of it all, we all want to know what the heck is going on with everything that we're doing. You know, it's kind of like if your tweet is in the forest, can it really be heard, and how are we going to measure it? And that goes for everything that we do in email marketing and social and digital and so on and so forth. And so we are very excited. Justin is the Analytics Evangelist for Google Analytics, and so I will encourage you all to pose some questions to Justin because the last third of this, Suzy is going to be taking your questions and posing them live. With Justin. As you know, for those of you who are not new and for those of you who are new, you can ask those questions on Twitter at #H2Hchat. That's #H2Hchat, and make sure that you pose them at any time during the course of the whole chat, and we will be collecting them. Suzy is working hard at doing that, and so there we go. Now, first things first, Justin, how are you today? I'm doing great.

I'm doing great, getting over a cold, so I sound kind of funky, or I might just sound funky to me, but I'm feeling good.

Well, that's the good news about Google Hangout, your sister software makes you sound normal. Perfect.

And look better.

And look better, there you go. So moving on, so how long have you been with Google Analytics, and tell us just kind of a little bit of your background.

Yep, so I've been at Google for just over 3 and a half years. It'll actually be 4 years in January, and I've spent 20 years working in IT and in software. For the last 10 or 11 years, I've been specifically working in digital analytics. It used to be called web analytics, but then we grew up, and now we call it digital analytics. So I've worked a lot with on-site measurement, advertising metrics, recently app analytics. And prior to joining Google, I spent most of my time at startups that were all analytics-focused, right? Just kind of boutique agencies that would only do measurement. We wouldn't traffic any ads. We wouldn't create any marketing. We just were taking objective looks at the data and evaluating what was happening. I've been blogging for about a decade on analytics. I've written a few books on it as well. So I think I finally figured out what I wanted to be when I have grown up now, and

it's

a digital analytics person.

So let's jump right in. I know, by the way, for everybody who's watching, hang tight. because Justin is going to share some slides with you on the topic for today, which is measuring what matters. So we are going to talk about that in some— in his PowerPoint. But before then, I wanted to ask a couple direct questions that I think is on everyone's mind, which is analytics as a whole. What matters? Like, how do you— you know, there's so many people out there that are barely using analytics the way that they should, and I'm calling a spade a spade now, including myself. You know, I think I know what I'm measuring, but do I really? I don't know what's important. How do you know what's important to measure?

Yeah, so that's a really great question, and it's probably one of the biggest issues that a lot of people have. And it's not just SMBs and beginners. It's everyone, all through, up until massive enterprise operations. And when it comes to analytics, we actually don't want to start with the data and the numbers. We want to start with the business plan and what is the business trying to achieve. And whenever we sit down and you work with anybody, right, any business owner, you're really trying to get at, well, what is their objective? What are they trying to do, right? And so easy examples are like e-commerce or service providers that are looking for leads online. You know, you want to start by saying, well, what does success look like for the business? Is it selling products? Is it getting leads? Is it generating revenue from advertising? And once you kind of get at the heart of that, right, once you get at what the business is trying to achieve and what success looks like, then you can start to put a measurement plan around that, right? And you can start to say, like, okay, well, success is getting 40 leads a month, right? So obviously, you know, you need to measure leads. But then you need to kind of surround that with supporting metrics. And we use the term KPIs, key performance indicators, a lot. Mm-hmm. And what you'll notice is that you kind of have this this hierarchy of metrics, right. So Bryan, the boss, probably only has 1 or 2 metrics that he's really comparative, right. Like, so how many leads am I actually getting every month? Whereas, you know, Justin, who's kind of the digital marketing manager or the analyst, he might have like 20 metrics that he's really, you know, focusing on, right. And Maya, she might have maybe like 30 metrics that she's looking at because she's really focused on social. You know, the metrics kind of evolve based on the role of a person within an organization and what they're trying to do. So the best advice that I can give anybody who's kind of hung up on analytics is don't even look at analytics. Like, whether it's your own site analytics or it's social data, sit down and say, like, what the hell am I trying to do here, right? Mm-hmm. And then let that be the driver of the numbers that you're going to look at, because At the end of the day, you know, you're going to go to marketing and you're going to be like, okay, I have this number and I want to make it go up, so I'm going to do these 10 things.

Yeah.

Now you can attach numbers to those 10 things, and you can start to adjust in real time, like, ooh, I did this thing, but the number didn't change. In fact, it went down, so maybe I shouldn't do that anymore, right?

Yeah.

And that's really the most critical thing, and that's kind of all there is to it. Now, we can layer on some more advanced things that the software lets us do, but really start right there. Start with the business, what you're trying to achieve, what the numbers your boss wants to see, and then you can start to think about how do I get these numbers.

That's great, that's great. So let me ask you this, what do you do to measure people that are Because the entire system is really this unknown. I know you and I talked about this in Italy a little bit. And, you know, so what you're measuring, you're not measuring actual names, you're not measuring people, you're measuring trends. Is that fair to say?

Yeah, I think that's pretty fair to say. It kind of depends a little bit on the dataset that you're looking at, right? There's some datasets that are less noisy and some datasets that are more noisy. So for example, like CRM data, like your customer relationship management software data, like Salesforce data, that data should be pretty clean, not a lot of noise in it. Other data like your website data might be a little bit more noisy because of potentially some spiders and bots and things like that. So it kind of depends on the dataset, but yeah, you're pretty much spot on.

Okay. Now let's talk about conversion. Um, hot thing for any business, right? Like, aren't we all in business to convert? I don't even care if you're a nonprofit.

That's it.

Everyone wants to convert. Um, and, and, and, and the only way that you know if you've converted is, um, is if you've gathered something, you've gotten something in resp— in, in, and you've received that, whether it's an email, it's money, it's, it's a conversion of something. But analytically speaking, it also probably has other ways that you can track that, especially in Google Analytics. Is that fair to say?

Yeah, glad you brought up conversions. We definitely think of conversions as 2 types of conversions. We have macro conversions and we have micro conversions, right? And you actually hit on probably 2 of the most popular that we talk about, right? The macro conversion is Someone gives you money, or someone submits a contact form. Those are the really valuable things. Then you have lots of micro-conversions, right? Someone might sign up for the email newsletter, or someone might click on the follow me on Twitter button or something like that. But yeah, I mean, most of the software like Google Analytics or a lot of the other analytics tools will have specific features. They're usually called goals or conversion tracking. that you can configure in order to track those actions. And the reason why you want to set those things up, the reason why you want to configure your tool to track those things, is because you'll get a whole bunch of reports that will tie your marketing and your advertising activities directly to those conversions, right? So you can have a column, it'll say like email signups, and then you can maybe have a list of, oh, here are my marketing campaigns, and here's how well they did at that specific conversion, right? That's ultimately what we're doing. This high-level number, that's the number you want, Bryan, right. But Suzy, she might be like, well, I did these 5 marketing things, how well did each of them perform at driving those conversions, right. That's a very simplistic example of what we call segmentation, which is critical, right, because the bulk number, good for you to know. But Suzy, who's pulling all those levers and actually doing the work, she needs all those segments so she can be like, ah, I'm not going to do this anymore. I'm not going to do this. Oh, but this thing, I'm going to double down on it next quarter. That's why getting your tool configured to track conversions is a really good thing.

Do you need a degree to do all this? How hard is it for— I mean, I mean this, I love my mom. Could my mom do this?

Yes, exactly.

How hard is this for businesses? How far could we take this? How sophisticated is it?

Oh gosh, it can get really sophisticated really fast. I got to tell you, like, so my job at Google, and I have a whole team that is wonderful and helps me do this, is we're there to make our users smarter. Like, that's our job, is to, like, once you sign up for Analytics, our job is to actually help you use the product more effectively. And we have tens of millions of businesses that use Analytics, and they range from, you know, Procter& Gamble and Johnson Johnson all the way down to Cousin Bob who makes cat clocks in his garage out back. So we work really hard to try to make it approachable. We actually built this thing called Analytics Academy. Anyone can access it. It's analyticsacademy.withgoogle.com. And we have a series of courses. They start out with digital analytics fundamentals, and then it kind of branches off depending on what you're doing. We have one on apps. We have one on e-commerce. It can get hard, and I don't want to sugarcoat it for your mom or for my mom, but we're trying to be there to support people and to help them out. That's why we give away our tools for free, most of them. So I think the important thing is going back to your previous point, right, focusing on measuring what matters, right, and letting that be your guide. And then going, okay, Mr. Tool, how are you going to support these numbers that I need to make decisions? And then kind of figuring it out. Don't let the tool dictate your measurement world. Let your business do that dictation. And it becomes a little bit easier to figure out. You've got to take some time and do it, but it's totally worth the investment.

That's a great resource. So Susie just tweeted that link out. So let's jump into your presentation, your PowerPoint, because I think it's going to help set things up and clarify a lot. I got to sit and see a lot of this, I think if it's the same thing that you're showing today in Italy, and it's just fantastic. So if you wouldn't mind, we can just jump over to that and we'll go from there.

Cool, yep. So what we— let's see, looks pretty good to everybody? Great.

Yeah, we can see it just fine.

Great. I'm just going to kind of pick a few of these slides for you guys. But really what I wanted to talk about is, you know, data is really important. And Bryan, I think you really touched on something critical when you said, is this hard? Can I do it? It's not a matter— it's not really a matter of can you do it? you kind of have to do it. Like, the data is becoming a critical piece of marketing, and the title of this presentation was Data and the Marketing Revolution. And the reason that digital has become such a powerful tool is that we have a tremendous amount of data that helps us understand what people are doing. And really, all of this information It's behavioral info. Bryan, you touched on it when you said conversions. Conversions are really just people doing a specific action. It's a really important action, right? It's people giving you something really valuable to your business, but it's really important to collect that information. Once you have that information, to really understand what's happening around it. What's driving those conversions? Are the same people converting more than once? Where are those people located? What marketing activities are critical to those conversions? And so this data is becoming really important to marketers because the decisions you make are all based on this information. It's all based on what people are actually doing. Now, the really cool thing about digital analytics is that we're seeing some pretty— an evolution, right. We're really seeing a transformation of what's happening, and our digital measurement is becoming less focused on kind of sessions and what's happening on the website and more focused on users and people and what they're doing. And I think this is a really important thing to remember, especially for marketers. Right.

So Justin, would you say humans?

Ah, humans, yes. Now remember—

Sorry, had to do it.

I just have to switch back so I can see your face after doing that. That's awesome. Now remember, a lot of— so that's a great point, and you're really— I think people might be like, wow, they planned this. We didn't plan this, but you make a great point, right, because traditionally in web analytics, right, what we did for a long time, it was anonymous, right? We didn't know who the users were. We didn't know who the humans were. We were just like, it's a cookie, which is totally—

Yeah, that happens every time you say human. I should have warned you.

Not anymore. No, anyways, but what's really cool is like it's evolving, right? Analytics is evolving to become much more personalized, right? We really want to understand users, and we don't really need to know them by name. We don't need to know it's John, Kyle, Justin, Bob, or Paul, but we really want to know, have they been here before, have they purchased before, have we worked with them before, things like that, because that additional information is critical. That data is critical because we can use that data, right, to make better decisions about our marketing and our advertising. And really what I like to say is that we're having a massive convergence moment in the world of analytics. We're actually starting to do things that have been done by database marketers for decades, right. And so if you've been around for a while and you've done database marketing or direct marketing, right, it's the idea of using your CRM data or using a database of customers to create personalized communications, because personalization really will lead to more conversions, more retention, driving down the cost of your marketing, increasing the retention of your customers, and overall just making you become a much more efficient marketing organization. And we're having that moment now in digital, right. I remember growing up, I grew up in the northeast part of the US, and I remember L.L.Bean. L.L.Bean was a huge brand in our house. Mom would buy L.L.Bean backpacks, and we had the duck boots, and we would buy L.L.Bean jackets and all this stuff. It was amazing because their catalogs would come, and they would be personalized. They would have just the right amount of kids' stuff. Now digital marketers are kind of able to do that, customize what's happening. It's really this great renaissance moment where we're having these 2 data sources, the data about people that I know, people that I don't know, those datasets kind of converging and giving us as analysts and marketers the opportunity to use the information to retain customers and attract new ones. Now critical to this entire process is the idea of segmentation of the data. So if we start to look at bigger groups of data brought together, it's important for us to kind of carve off those groups, right. So who are my really valuable customers, who are my kind of valuable customers, who are my people that are at the upper part of the funnel, right, that are just learning about me and my brand, how do I make them into customers, right. We don't want to treat all of these people the same, we want to be able to divide them into groups and then do different things for them. For example, one of the things that's really cool that we can do today is when people visit a website, let's say they visit the Shareology website, Bryan, and they're reading all about your book, and they're reading the excerpts, and they're reading testimonials and reviews, that behavioral information is really important. Right. Because now we could kind of create a segment. We could say, I want to segment all the people that came to my website and read about Shareology, but maybe didn't buy the book. Then what you can actually do using some analytics platforms like our analytics platform is you can actually create remarketing, which is a type of digital advertising where you change the advertising that that specific user sees as they browse the internet. You can actually say, oh, people who visited my website and they read the Shareology excerpt, we can actually have them see different ads, maybe something like, buy Shirology, 10% off, or, did our excerpt catch your interest, blah, blah, blah. Now, obviously, there's a super fine line between being a little bit creepy, but these types of data, this behavioral data, can drive way more conversions and make your advertising much more customized and effective. Having this information about users and what they're doing and then using it to kind of customize their advertising experience is really important. And the data that you can use to customize experiences isn't just kind of restricted to your website and to display advertising. There's data about all sorts of behavior. And so I kind of like to think of this kind of quadrant system here, right? 2 different types of users usually that we're dealing with. We have known users and we have unknown users, right? Known users are people who have maybe purchased from me in the past, or they've come to the website and they're kind of really engaged. Unknown users are people that I've never met before, I haven't connected with, they haven't visited my website, they're just complete unknowns. And then we have a third type of user. We have kind of the other vertical parts of this quadrant system are kind of the properties that I own, right? We have lots of different properties that we manage these days if you're a marketer. You have your own site, which is where you post content, your blog, you might have a store there, you might have white papers, whatever. But then we have all these other platforms that we're on— Facebook and Twitter and Pinterest and— LinkedIn. Who knows what else, right? There's a new one every day. We also have all the different places that we do display advertising. We have search engines like Bing and Google, right, where we don't really control content, but we have the ability to reach out to users. And so when we kind of look at this matrix, we can actually use these channels in order to reach out to people, right? So like on my site, for known users, I can customize the content that they see on the site. Because I know maybe how many times they've been to the site. I know if they've purchased. I can customize email to them because I have their email address, and maybe I know some of their interests. But known users outside of my site, on other sites, I can anonymously show them certain content. So if they've been to my site and I know who they are, then I could do things with display advertising and customize it based on their behavior on my site. Or search advertising. I can change the search ads that they see, pay-per-click search ads, based on the behavior on the site. Same with social, right? There's connections, and I can start to say, well, you've done this thing on my site, I'm gonna change what the ads are I'm showing you on social. Now what's really cool is that even if we don't know who you are, we can do a lot of the same things. And this is where the concept in advertising of audiences come into play, where I can say, well, I'm trying to reach a certain audience like men who are 40 to 45 who still think they're 23 and are way too adventurous. I can create segments of those users using third-party data, and I can kind of target the stuff that they do using display advertising or social or search. And then when they come to my site, even though I don't know who they are, I might know which geographic area they're coming from or which marketing activity they're responding to. and I can customize the content. So the point I'm trying to make with all of this is that we have tons and tons of data, first-party data that you have about your users, third-party data that lives in social and advertising platforms, and it's this data that is becoming critical to us as marketers because we're able to put a better message in front of a better customer or potential customer at the exact moment we want to, and hopefully get them to do what we want to do. So I know that's kind of pie in the sky, and sometimes it can take a lot of work to get there, but like I said, it's really this watershed moment for analytics where we can do some pretty amazing stuff that we've never been able to do before.

Wow. Yeah, so you have some tweets coming in that are basically— Ursula Ringham just friend of ours just said, mind blown. I mean, I know a lot of this is built into the software, but it's funny how powerful the software really is. I think the biggest thing that we're all facing is shiny penny syndrome. Of, you know, as marketers, we all see all these different things going all over the place. We're tweeting like there's no tomorrow. We're, you know, we're sending out email campaigns like there's no tomorrow. But at the end of the day, are we moving the needle as much as we can in the right places?

And that's it, right? And that's why, like, we go back to step 1, right? If you have a measurement plan, You can be like, I don't give an F about that number, because it doesn't support what I'm trying to do. And I think people get so intimidated, but if you have a measurement plan, you can immediately say, oh, okay, Twitter just came out with a new metric, and it kind of describes how far my tweet reaches, or maybe the influence of the audience that I'm reaching. Now, for me, I can be like, don't care, didn't bring me any money, right? That's what I care about. But for you, Bryan, you might be like, well, my objective this quarter is to reach a bigger, better audience on Twitter. Snap! That number's totally awesome. Like, I'm going to try to make that number go up and to the right. And it's an important thing for you. So, you know, if you have that plan in place, it becomes a really easy process to Get out of that shiny penny syndrome. I'm super guilty of that as well. I'm just a product nerd. I love all the technology products that we have today, whether it's a social thing or an analytics thing or even a new function in Excel. But yeah, having that plan kind of brings you back to reality in a hurry.

So is there in your— and I'm just about to pass the baton to Suzy here, just giving her a heads up. And you're not going to be shy of questions. They are pouring in. And they're good ones. Don't worry. I don't mean to scare you, but the question that I was going to ask is, whether it's in your resources area, the Academy, or where would you go to really kind of learn, like what you just said, to build the right plan?

Yeah, right.

Does the Academy help you build that plan you're talking about? Are there templates there? Yep. How do you do that? And what does a measurement plan even look like?

Yeah, well, right in our very first course, Digital Analytics Fundamentals, on the Academy, that's exactly what we start with— building a measurement plan. And, you know, it doesn't have to be fancy. It can be a napkin and some squares, right? Usually, The biggest thing about having a measurement plan is talking to the right people, because Suzy can sit down and she can build out a measurement plan, and she's like, this looks pretty damn good to me. I'm all good. But then she might go talk to you, Bryan, and you might be like, well, you know what? That number is kind of a waste of time for me. So again, within an organization— right now, if the organization is 4 of you sitting in a cube, you can bang this out over lunch. But the bigger the organization, the more complex these things can get, right? Yeah. Different people, different stakeholders, different Objectives, so different data. But, you know, I could sit here and look at a whiteboard with you guys, and that's a great place to start.

Yeah. Okay, final question before I pass the baton. How do you figure out the value of a potential customer?

Oh my God.

I mean, yeah, right? It's— sorry, but isn't that like what this is all about? Like when you're planning out your goals, and your goals are to convert, and the conversion is around a number, and that number is X, how do you even get there? If there's no simple way, is there a big picture thinking around this?

Susie, what was that we were going to do? Can we jump to Susie? No.

Normally I ask the tough questions. This is— he's rolling. He's so excited right now.

She can tell.

All right, so I think what you're asking is the lifetime value, right? How do I calculate the value of a customer? Yeah. This is also another really, really interesting thing that's happening, right? And over the last— some people will argue that since we've had mobile apps really kind of explode, lifetime value has become a much more important metric. I might argue that since we've had SaaS-based business models, software as a service-based business models, lifetime value has become much more important. Lifetime value is kind of a hard number to get at, right, because it's a forward-looking number. And, you know, really simply you could say it's like, well, you know, the value of a customer, you know, since they first became a customer, everything they spent, that's the value of a customer. That's kind of overly simplistic. I think I'm going to kind of stay away from that one for now, because we're still evolving to figure out what's the best way to measure lifetime value. If you're an app, it's a lot easier, right, because you have clearly defined users, you know their activity patterns, you know churn and things like that. So it's a tough number to get at. What I will say is that The important thing is when you have a measurement plan, a critical piece of it is adding the targets that your business is setting. So again, if you're commerce-driven, you know, we want to generate $10,000 in revenue in March. Boom, right? There's the best context for your revenue metric that you should be looking at, right? How far to $10,000 am I? Am I getting closer? Am I getting further away? That's really what you want to focus on, is like putting some targets along with all of those things to provide context into your, your reporting of the data.

Great. Well, I have more questions, but I'm not gonna pose them because I think that was a nice one to close, close out with, at least for now. I may jump back in since we've got a little bit of time. So I'm gonna pass the baton over to Susie, who has some questions from the audience as well as I'm sure some herself, as she is a data— and I mean this in a good way— a data freak. And so I will pass the baton over to her.

Awesome. So totally self-proclaimed. I like data freak. I would call myself a data geek, but you know, you know, it works. So please keep your questions coming. We have some great stuff. I know there's tons of information coming through, so Keep them coming though. I am working to curate them. First of all, thank you, Justin. This is absolutely great. I am learning absolutely so much. Just kind of to get started, to get kind of started on— I've had a comment that I'm going to turn into a question from a few people, which is commenting on— you've been talking a lot about personalization. And everything that we can do with data, right, and how we can really drill down. And I mean, you're at Google, masters of data, right? So what is that line though between personalization and creepy? We were actually chatting about this right before we went on air. How, in your perspective, right, where— how do we deal with data to make sure that we're really engaging? And creating positive things as opposed to the other kind of, ah, where'd you get that info?

Yeah, totally. And there's no algorithm that says, hold on a second, what you just did was creepy. It doesn't work that way. We haven't figured it out. I think I always actually— the way I try to stay grounded in all this digital stuff is just by talking to my wife and trying to be just a neutral observer sometimes. Let me give you an example. I've got a little bit of woodworker in me. I love working and building stuff.

Nice.

Excuse me. I frequent this website called rockler.com. They sell great gadgets for woodworkers. One day I was on the site and I had a few things that I liked, put them in my cart. Just didn't want to buy, took off, and got a nice friendly email saying, hey, don't forget you've got this awesome stuff that's waiting here. If I separate myself out of the whole process, I was like, that's not too creepy, because those are things that I just— I like them and I want them, but I just wasn't ready to make that call yet. I think we all have that experience, right? For me, that's totally fine. I think we all have to do our part to take a step back in our work every day and say, is that lame? Am I awful?

Yeah.

It's up to us. Susan Wojcicki, who leads YouTube at Google, and she was the former Senior Vice President of Advertising at Google, my boss's boss, And she was on a crusade to kill bad advertising. And the only way we're going to kill it is through all of us stepping back and saying, you know what, that would make my mom feel creepy if I did that. And just don't do it. So honestly, I think that it's human to human. Right, Bryan?

Applause.

So think about your fellow humans.

I love that. That's, that's exactly what, what I always tell my clients when I'm doing community building, right? Come back and even think about, you know, you mentioned your mom. I love that. My thing is, how would I want to be approached? But I like that. That's even a step further, right? Yeah, I get irritated sometimes when I like that, that initial email, but like, I just had someone who was like, they were pinging me like 5 different times, shopping cart, shopping cart, and it becomes, you know, You're kind of trying to figure that out. So, I think that's great. So, I have a great question here from one of our pals, JS Gilbert, a regular on H2H Chat from the very beginning. Shout out to him. He says, what numbers do you think people fail to pay attention to, but really should?

Yeah, so that's a great question. For me, it's micro-conversions. And I'm not going to— well, let me— there's 2 types, right? First is micro-conversions. I think people fixate on just macro-conversions too much. They don't think about all the little steps that happen before the macro-conversion, right? And maybe let's just focus on a website and the data that comes from the website. They don't focus on how much is read. They don't focus on what is read. They don't focus on— If people are using tools, things like that. So if you're a travel website, how do people interact with your travel planning tools and things like that? I think those are really important metrics. A lot of people settle for kind of your out-of-the-box metrics like time on site and page views per session and things like that. I think people should kind of, instead of taking those canned metrics, they should be like, I want to grab some of these actions. Because if someone clicks that button, damn it, they are way better to me than 4.5 page views per visit. Yeah. So I think that's really important. And obviously, those types of metrics require a little bit more instrumentation, so some people might freak out a little because it's not out of the box. The other metrics that I think people are really not taking advantage of yet are attribution metrics. And this is something that I didn't rant about Earlier, Bryan, but attribution is something that's taken off over the last few years. It's really taken off because the software now has been able to do so much more than it had in the past. Here's a really, really simple metric that can get anybody into attribution. It's called assisted conversions. We talk about a conversion, a macro conversion, and traditionally we look at Here are my macro conversions. What did people do to generate those conversions? Email, search, paid search, whatever. But reality is, it doesn't usually happen off of one click. It takes time. Yeah. It could be a week, it could be a month, it could even just be a day, but it might take multiple interactions with the website in order to drive that conversion. And so understanding how multiple marketing activities work together, right, to generate that conversion, that's where attribution comes in. And so an assist is basically, you know, it basically says, okay, this marketing activity, it didn't generate the final conversion, but it appeared— this marketing channel appeared in the person's path to conversion. So for example, I wake up in the morning, pop open my inbox, I get an email from Patagonia, and it's like, check out the awesome new fleece jackets we're making this fall. Click, visit, wow, great fleece, got to go to work. Right. Go to work, but then like 5 hours later, I'm like, all right, well, Patagonia's I'm ready to get some new ski gear, just bang in a quick search, and I'm back on the website and I buy something. So those 2 marketing activities definitely interacted, right. And so understanding that, we're to this point now in the world where we don't have to ignore the email when it comes to attributing value, right. We can actually say, sure, the search generated the conversion. But you know what, 4 hours ago email drove Justin to the website. Don't you think email should get a little bit of the love here? Right. And so we can actually start to attribute some of the value, whether it's monetary or just show you how many times certain marketing activities influenced conversions. So I think that's another area, folks, that you should totally start to think about is attribution and assisting conversions.

I love that. I absolutely love that. So then, then here's my, my— so this is, this is actually, this ties in really well to a question that Alan, Alan Berkson, who is part of Freshdesk, who actually is one of our Shareology sponsors. So shout out, Alan. Thanks for joining us. Right. Whoo. Okay, I have another cat incoming. Those of you who are regulars know that my cats attack me. Mid-H2H chat, it's human. Anyway, going to Alan's question, he says, I get a fly. Love it. He says ROI on social is a sore point. How can Google Analytics be applied/help? And kind of to add to mine, you know, this assist concept, you know, as a social strategist, I love it. It reminds me, I always talk about kind of the, you know, traditional rule of 7, right, where you want to see things. It usually takes on average seeing things 7 different times before you make that conversion. And when we have social, once again, it could be the email headline. I might not even click on the email, right. It's just if the headline is well, I may see it someplace, right. You get all these different touchpoints you're trying to aggregate. On some of these, they can really be tracked. Yeah. On social, it seems like it can get a bit trickier. Talk about this.

Yeah, so that's a great call. And really, excuse me, what I would say is social has— well, first, all of these platforms, right, excuse me, it could be Twitter, it could be display advertising, it could be Facebook, right, it's how you use them. That is a critical thing to take into account, right. Really, a marketer can really use them in 2 different functions, right. You can use them for brand, or you can use them for performance. That's the terms that we use, brand or performance, meaning like a direct response. And it's the same with email, right. Email can be used to build your brand, or email can be used to generate a direct response. That's the first thing you need when you're going to measure any of these things. What are you trying to do, right? Another clear example, right. Bryan is a Twitterholic, right? And he uses it for a couple of things, right? He obviously uses it for brand. That's why we see tweets about quadruple rainbows and his kids hiking. And that's also— we also see him using it to send people directly to Amazon and other places to buy his book, because he likes money at the end of the day, right? So no, like, cash register, Bryan, for

that one?

Thank you. There you go. That's the first thing you need to do, figure out what is my objective in using this platform. Then you start to measure it. If you're going to measure one of these things based on a direct response, sure as hell, I want to see how many clicks did those links get, how many people landed on the website, how many people added the book to the cart, how many people checked out, and how much revenue did it generate. It is a direct response campaign. My metrics should support direct response objectives. Mm-hmm. Now, the other way to do it is if you're doing brand stuff. Brand is super fluffy, like all of those tweets about rainbows and unicorns and whatever. What are those for? Those are there to get Bryan more followers and to generate some engagement and to get I'm not going to use the H word— people talking to each other. Those are a little bit harder, right? So you might want to focus on metrics like, all right, well, did I get more followers? Did my reach expand?

Right.

Now if you wanted to then tie those into assisted metrics, that becomes hard, because you're dealing with data on 2 separate platforms. You're dealing with data like in Twitter, And you're dealing with data like on your website, right? If the person didn't go to the website, it's really hard to understand how it assisted. Now, there are some really advanced attribution tools that are out there. We make a product called Adometry. Those are the types of products that will actually consume data from multiple systems and start to do some of the estimation and the joining of the datasets, but we're getting into kind of expensive solutions in order to understand them. So I think The most important thing that people can do is really tie the measurement to the objective and the use of the platform, and that will get you started.

Yep. Yeah, I'm always— I mean, this is the elusive thing, right? I get pitched by a lot of different analytics companies, and anytime they're talking about ROI, I go, if you've cracked this, you are going to I

will say, platforms like Facebook, platforms like Twitter, they do have conversion tags. You can go into Twitter, you can get a conversion tag, put it on your website. They'll start to tell you, here's the lookback window. You can get that. For anyone that's maybe been doing analytics for a while, just rewind 5, 7 years. And we were talking about display advertising, and we were using metrics like view-throughs.

Yep.

Well, that's basically the same effing thing. We're still doing view-throughs, but we're doing it— it's like, well, what was the reach of this tweet, and in 28 days did that person ever buy anything? So we are getting there. It's just not like, you know, kind of out-of-the-box analytics solutions are not able to tie all the pieces together unless you want to drop a ton of money.

Yeah, well, I mean, in some ways too, not to wax eloquent, but I mean, the thing that I always talk about is it's kind of trying to get the ROI out of a conversation, right? There's a point at which— or your relationship, right? There's a point at which I just can't measure that, right?

Oh, Suzy.

There's a certain point there.

Yeah, but I'm going to push back on you as the data guy who likes numbers. And these little pesky things called facts. The point of everything today is not to generate money. It's literally not to generate money, right? So when we're done with this, and I'm going to write you guys, and I'm going to say, here are the metrics I want. I want to know how many people did we reach. I want to know how engaged they were. And then I want you to tell me how awesome I am compared to all the other guests you've had.

Oh, hey, I can get you that. I can absolutely get you that.

But that's the objective.

Up and down, yes. With your objective, yes.

But on the flip side, if I was sitting here with a new book to peddle, and I'd be like, Suzy, 7 clicks, 2 sales, really? That's it? I did all this for that? Come on, my kids got to go to college. While we do say that, and I'm the first one to admit that social and engagement is really important, it's how this works, how we're using it today, and what we're trying to get out of it.

Yep, yep, no, I completely agree. You have to begin with a goal. This is something we talk about a lot. Totally agree. If you are measuring the entire network, then you end up with those little pesky little things. Remember I told you I'm a social scientist, you know, background and all that stuff. But no, I completely agree. Love it.

But then your reporting should be like, okay, in Q4 for social, we are going to dedicate this much effort to grow

Yep.

And when we activate them, we're going to expect this much revenue out of them. Yep. And now it's like, boom, there's a revenue number, right? And then I can go back and I can say like, okay, it took me 2.5 Suzies to generate this much revenue. Is that above or below zero?

There you go.

Right, so—

Hey, Justin, give me one example or even two of how you've seen analytics change behavior which evolved maybe a business. You know, it evolved maybe the cultural thinking or the way it thinks about its business or its product or its solution path that it takes. What kind of decisions out of the analytics, and I'm kind of jumping now beyond just the software, it's what What kind of things did it course correct or say, hey, did you pay attention to this? And then they did and it made a big difference.

Yeah, so gosh, it happens on so many different levels, right. I think at the most fundamental level, analytics can be transformative to how a business communicates internally, how decisions are made internally, and how different groups work together, right. A lot of times in larger organizations, what they suffer from are silos, right, because you'll have, for example, one group working on customer retention and one group working on customer acquisition, and never the twain shall meet. And that's a huge issue, right, because in reality acquiring customers and retaining customers, those folks should be talking about how they are communicating to said customers and to those people before they become customers, so there's consistent messaging and there's better measurement. We see siloing across different marketing channels, right. The search people might be completely separated from the email marketing teams. And so I think at a very fundamental level, Bryan, when we start talking about having a unified measurement plan, we start talking about people sharing metrics. And I think especially breaking down those data silos, attribution really forces that, right, because when you say Hey, Mr. Email Guy, Miss Search Person, she's going to get a little bit more credit because you guys are actually working together for a common goal. So I think you can have these really transformative moments of how an organization is structured, the language that they talk in, reporting. I think at a day-to-day level, right, the thing about the data is people make changes all the time. And so Susie, you brought up an example about email with the subject line.

Mm-hmm.

That's analytics, right? One email, 5 subject lines, I'm going to test to see which one works best, and that's basic analytics. So tweaking marketing campaigns, changing advertising campaigns to be more productive, right? Because the basic gist behind analysis is, did this thing work? Yes, then do more of it. Did this thing work? No, stop doing that. It's not really overly complicated, right. And so those things happen all the time. So it's unbelievable, like from simple tactical decisions that can change to organizational transformation that can change just from doing analytics.

Love that, I love that. So I think we have time for a little bit more, maybe one or two questions. So kind of getting to the practical aspect, right? So I have a question. I'm going to combine 2 questions. One is from Haley. She's a first-time H2H chatter. Hey, Haley. She chose an awesome one to start off with.

I think so.

Yeah, right? You're not at all biased though, I'm sure, right?

Not at all.

So she says, how do we open this conversation to help our client measure what matters and get stakeholder buy-in? I also had kind of an additional question from our friend JS Gilbert, who also said, where— how can a small business embark on this, right?

Yeah, tons of consultants, tons of SMB consultants, SMB owners as well. You know, if you're working with clients, I think The thing to tell them is that— and I just want to quickly just give shoutouts to my buddy Avinash, who's the digital marketing evangelist at Google, Avinash Kaushik. He's been great about this, is really just helping people understand that all of these systems, all of these tools are so great because the data is there. Right. To be honest, if you're a consultant, I might not even talk to them about analytics. But what I would stress to people is that we're going to do some unbelievable digital marketing, and you should be totally psyched to take some chances, because I'm going to give you data about every single thing I do. So when I come to you with a marketing idea, I'm going to tell you, we're going to do this awesome thing with email, and here are the numbers that we're going to use to measure if it works. So I can change really quickly if things aren't working. It's not like I'm going to go and spend a bunch of your money and then 4 months later come back. It's going to be like I'm going to spend some of your money and then tomorrow I'm going to tell you how it worked. And then next week I'm going to follow up again and tell you what happened in the 7 days since that time. Yep. And then the next time we do this thing, I'm going to reference back to this data and say, like, remember we did that thing? And it was really good. We're going to do it again, but we're going to make this little tweak and see if this little tweak makes it even super awesome. Right? So I might not even bring up analytics because it's just kind of overly complicated. Now, if you're a little bit bigger and you want to get into it, I think opening the conversation around creating that measurement plan, understanding what's really important to them, and translating the objectives that are in their head and why they're doing this into numbers is really a good place to go.

I love that. I love that. One final question from Carrie. So she says, I agree that different roles need different KPIs. Do you— have you seen in your experience an ideal number of KPIs to track? Open-ended.

Yeah. So here's my kind of rule of thumb when it comes to KPIs, right? Like, you really shouldn't have more than like 3 or 5, right? Mm-hmm. And like, here's an example. I have tons of KPIs because I— like, my teams— if you ever use the Google Analytics Help Center, my teams write that. There are 5 numbers that I think about every week. That's it, just 5 numbers, right? How many new people came? How many returning people came, how many people looked at one page, what was their satisfaction, and which languages are they really interested in. So some segmentation in there, but those are the only things that I look at. Now, are there more things that I could get? F yeah, but I don't really need that data. I just want to see if those things are going up, going down, or if there's really anything new happening, and that's kind of what I want. Once you get— now you, Susie, you might be in the weeds. Let's say you work for me, and you're like, hey, remember we sent out that email to 2 million Analytics users to help them set up conversion tracking? You might be like, oh, I need to know how those did. So you might need a few more. I think your top-line KPIs, 3, 4, maybe 5. Someone that's in the weeds, a lot more data.

Yep, yep. Makes total sense. All right, that's it for me, Justin. It's been seriously awesome. Great.

Awesome. Justin, where can everybody find you? And just one more time for anybody that popped in for the second half, where can they find more information about Google Analytics?

Yep, so I'm Justin. I'm on Twitter. I try to be pretty active there, but sometimes work gets busy. Just @JustinCutroni, my name. If you want to learn a little bit more about just the fundamentals of digital analytics or Google Analytics, check out analyticsacademy.withgoogle.com, and occasionally I bang out a blog post at cutroni.com.

Awesome. Those links are coming out everyone's way, by the way.

Awesome. Well, good. Hey, Justin, I really can't tell you how much we appreciate this. It was something I've been looking forward to ever since we met and you offered to do this. I know everybody was— has lots of questions, so your resource center will come into play, I'm sure. And even my own team is pinging me behind the scenes saying, oh cool, we got to do more. So obviously, you know, we're going to be jumping on that as well.

Don't worry, Bryan. I think your IBM buddies are doing some pretty awesome stuff as well.

Yeah, they are, and actually ironically that's a great transition to next week where IBM will be on with us. And so anyway, one quick shout out to our sponsors, for if it weren't for them we wouldn't be able to do this. Our sponsors are Newstar, MasterCard, Cisco, IBM, Little Bird, HootSuite, Freshdesk, Speakeasy, Influitive, Mutual Mind, and Cox Communications. And if you would like to join us next week, we've got— we've got— oh, I just lost it. It was good. Here we go. Reimagining What Employee Communication Should Be. And we're going to be talking about the traditional workplace as we know it has vanished. These rapid changes are fueled by new technologies And how do we see those changes coming? We've got an IBM panelist, Jaques Pavlini, and Denise Holt from Grazzez, a CMO and co-founder, established podcaster, speaker, writer, social business strategist, so you can speak medicine. So it's a mouthful. Anyway, I hope you do join us again next week. Again, Justin, can't thank you enough. We will keep in touch, I'm sure. Thank you. And Susie, hat tip to you.

Take care, guys. It's been fun.

Thanks, everybody out there. Appreciate it.