Latest / Talk Commerce / Marketing Intelligence Insights with Joshua Lauer
Transcript
- Brent Peterson (00:02.299) Welcome to this episode of Talk Commerce. Today I have Josh Lauer, the marketing data guy. Josh, go ahead. Joshua, I'm sorry, I got your name wrong already. Right off the bat, I screwed it up. Tell us your day-to-day role, one of your passions life, and what you're excited about for business this year. Joshua (00:22.7) Yeah, so day to day, I have a variety of different accounts that I work with on basically marketing intelligence kind of topics, which could include, you know, tracking things that are happening on your website. also also what comes after that is the reporting. And quite often there's a question that comes up after campaign or after last month, you know, regarding how did we do on this metric or that metric and why did it go this way or that way? So lots of like kind of combing through the data and trying to see, you know, is anything weird and in any of this data and can we make more sense out of it? All of that kind of goes together into making systems that then allow for better personalization, better targeting, and hopefully more revenue for my customers. Brent Peterson (01:22.929) That's awesome. And how about a passion? Do you have a passion outside of out of tracking things? Joshua (01:28.694) Yeah, actually, thing I was into before I got into this was songwriting. so I spent most of my 20s kind living in my parents' basement trying to write songs and... eventually went back to school for a bachelor's in internet marketing and kind of finished that I guess 13 years ago and now I've been doing this thing ever since but still kind of write songs and record them on regular basis. Brent Peterson (02:01.115) That's awesome. My son is in a band in the Twin Cities called the Michael Caine and Co. or Caine and Co. I think is his new name. But anyways, he plays bass. He tried the song writing gig and it was really hard to to break in. And now he's just a session artist, whatever. So that's great. Very cool. So Joshua, before we get started, you have graciously volunteered to be part of the Free Joke Project. I'm going to tell you a joke. You just give me a rating. 8-13. So here we go. My wife accused me of stealing her thesaurus. Not only was I shocked, but I was appalled, aghast, and dismayed. Joshua (02:44.206) 13 on that one. Brent Peterson (02:45.849) Wow. That's great. Yeah. Thank you. Appreciate it. Right. So Joshua, tell us, I just want to start it here because I think a lot of people look at marketing and they just do things in marketing and they don't even bother to track it. Like I think what you're doing, you're a step ahead of probably half the people out there. don't know if you have a, do you have a metric on how many people do marketing and actually track what they're doing? Joshua (03:11.662) That's funny. mean, if they're talking to me, the metrics are 100%, right? But no, I mean, I have a story that jumped on my head, right, when you said that going back, like, I don't know, 2016 or so, I was working at a company that will remain nameless for the sake of this. a colleague of mine approaches me he says, hey, we spent 50K last month on these ads and I need to know how they're performing. It looks like there's says they're not doing anything over here and that just doesn't sound right. And so when I kind of crack open the hood and dig in, I find out that those ads that were run had no UTM parameters on them, meaning in Google Analytics, they showed up as direct traffic. And that was kind of a big moment there for him, learning, OK, we got to make sure we got the tracking in place. But also, yeah, it's kind of an interesting reminder that marketing teams tend to be very creative. And when it comes to the data, that's sort of oftentimes an afterthought, and it probably shouldn't be an afterthought. Certainly, we should have a tracking strategy in place. if we're using that tracking strategy, we're probably looking at that data before we do the next thing to kind of say, how can we do this next one bigger and better than the last time? but as far as how many people are, are not tracking, I would say if, if you've got any meaningful amount of business, there's a good chance you're tracking something, whether or not it's the metrics. would recommend tracking is another question. but certainly in the beginning, there's, there's kind of a lot of people that will get going and their main focus is actually what they're doing. And so they're not spending time trying to track. They're trying to spend time on the deliverables and Joshua (05:06.752) all the other things that you find yourself needing to take care of when you're running a business. Brent Peterson (05:13.585) I know there's probably hundreds of different marketing data points that you should track. Do you think, well, do you think AI is helping people with that data tracking part of it? Or do you think people are still primarily using AI just simply for generative tasks? Joshua (05:34.734) That's a good question. In my experience, have not personally used AI to figure out the tracking part of it. I have come across scenarios where somebody tried to take care of something and they thought, well, I'll just ask ChatGPT how to do this. And then it doesn't work out the way they hoped. I get brought in and I'm looking at this and I'm going, what actually happened here guys? And it's like, well, I asked ChatGPT how to install this and do these things. And then we ended up kind of, know, sort of doing it the way ChatGPT said. We installed Tag Manager. I think this example, they wanted to get a split test up and running. Brent Peterson (05:59.622) Ha Joshua (06:18.568) And of course, Google Optimize was sunset a few years ago, the probably favorite free solution for split testing is gone. So there's a need to probably use like a VWL or something. In that example, I'm sure chat.tbt could have told them a variety of tools that they might want to use for split testing. But I think the interesting thing in my experience with chat GBT is that, yeah, it seems like it's really great at writing content. You can ask it to, I asked it to write a bio for myself recently and I was like this. I don't think anyone's written a bio about me before that sounds as good as that does. So there's certainly some great things. feel even the prompts you can get creative with, can ask for, I've done some stuff where I'm asking it to do some research for me, but then I like to see things in tables, go figure. And so I'm like, can you present that in a table with all the key points laid out? And it'll give me a table. which is pretty great. Brent Peterson (07:30.787) Yeah, good. So what do you think is, when a customer comes to you, do you have a standard list or is it by vertical or tell us a little bit about the approach you take to coming up with the metrics for people that are doing it in marketing. Joshua (07:50.766) Yeah, I think a lot of times the reason I'm being contacted is that either something's not tracking yet they needed to track or something's broken or people are spending ad dollars on things and they're not really sure if it's incremental in that respect. know, like are these ads actually driving more revenue for us? And so we're talking about an e-commerce site, like the typical things you'd want to know about that Google Analytics provides our session data. We've got now with GA4 instead of Bounce Rate, we've got engagement. So we decided to be optimistic instead of pessimistic with our metrics. Key conversions, which are gonna include purchases and revenue. And then from there you can get at things like average order value. And you have the purchase data and the session data, you can get at your e-commerce conversion rate. And so those things are really kind of the starting point. Once we have those things in place, it unlocks reports around like what channels, whether it's Google Ads, Facebook Ads, TikTok Ads. Organic search, organic social, what channels are driving traffic to your site? How are they converting? With GA4, things are like this data-driven attribution algorithm now. kind of think like they took a machine learning sort of solution and applied it to all the various kind of attribution methods that are out there, whether it's first touch, last touch, time decay model, all those. And then it kind of looks at those results and says, think this is how these channels deserve credit based off of that. the data-driven thing is kind of newer with GA4. I don't know if it's. Joshua (09:37.108) as understood as for those people that are used to like the older Google Analytics that had, was really last non-direct click. So even how we attribute things within GA4, you know, has an aspect of machine learning into it. Another thing we can unlock with just those basic metrics is the landing pages report, which is a pretty good one as far as like, did they come to your homepage? Did they come to a specific product page or like a category page where you had a subset of products? And what's the conversion rate when they land on one of those specific pages? So that's kind of a probably baseline starting point. as far as what I would approach just out of the gates if we're talking about an e-commerce site. Brent Peterson (10:22.363) When you mentioned earlier creative people aren't always the best at tracking numbers. Do you encounter teams that are super creative and they're just like clueless about how the numbers work or do you, maybe it's probably not a whole team because hopefully they'd have somebody that's sort of number oriented. do you encounter that a lot where the creative person just has no clue about like that maybe doesn't even care about the numbers part of it? Joshua (10:50.658) You know, it's interesting. I think in my experience it's been that maybe someone internally has recommended one of more creative folks has a chat with me about something because I might be able to tell them how this maybe landing page content is doing compared to a different page they worked on. And if one's doing better, then they're probably expecting me to say, like, why is that one doing better? And it's sort of when you have that content that's seen as, let's see how to say this. So yeah, when you're in that creative mode, you're trying to think of like, what are my users or readers or whatever people come to the site wanna see? Probably, is it emotional? How do you make it more emotional? If it's more emotional, we're gonna have more of a reaction to it. And a variety of other things that are, know, mainly probably due to the way it flows and all the things you might think about if you're making it creative. So I think the thing there is to really ask them what is the purpose of this content? Is it just... You know, is it just informational? You know, if we're talking about a privacy policy, we're not expecting numbers to jump off the page at us. On the other hand, if we're revamping the homepage, arguably the most important page on any website, then it's quite important to be able to provide information to them about how many people click on the hero image. If you have a lot of homepages have different like treatments as you go down like with various calls to action, how often are those getting clicked? What kind of navigation? are people clicking on from the home page? Do they go into a category? Do they perform a search first? And how does that flow into the rest of the site? So I try to guide people into thinking about if they're working on something or something new. Joshua (12:49.102) how do you start changing that in a way where you're kind of keeping these other things in mind? And sometimes it's like as easy as kind of saying, okay, let's just go on a whiteboard session and talk about what you'd like to do instead. And let's just throw everything out the window and just pretend that the site didn't exist and you get to do whatever you want. And so then we'll have the designer or the creative person on the call throw out some ideas. Then I'll start getting in the analytics and saying, okay, Well, if we look at historically, if we took this out, we're going to lose potentially this many people that used to do this. Is that okay for the business? Is that something we, know, is that whatever thing is going away and maybe to click on a certain treatment? was that previously driving a lot of revenue? And if it was, maybe we say, okay, no, we gotta keep that. Is there some other way we could treat it? But I try to make it an iterative sort of conversation instead of just being like, hey, like I put an Excel sheet together with historical data and you need to make sure that you review this before you do anything because, you know, I gave you the data so we can't say that you didn't have anything to look at when you're working on it. Brent Peterson (14:02.811) Do you think that the data driven person always wins even if they're not creative simply because they can do that iterative approach to figuring out what works and what doesn't work and they might get there by accident where a creative person is going to spend a lot of time trying to figure something out but not doing the analytics behind it to see how they got there? Joshua (14:24.824) That's quite interesting. I would say data a lot of times tells you what happened, but it doesn't necessarily tell you why, right? So the person who's being really creative and like digging in deep to like the why and trying to answer that question, they're doing a totally different task, right? The data is gonna tell us what happened afterwards. We might be able to predict if we think something's gonna be better, how much better do we think it's gonna be? and then kind of say if we got the same volume flowing through and this was actually this much better, that would amount to XYZ potential number in the future. But I think with the data, one thing that's maybe a surprise here is that it's not uncommon that I will actually be looking at the data, kind of scratching my head and going, what happened here? And then digging in more and finding, I've got a recommendation that we should do. this and when I present that recommendation, not kidding, some leaders have actually said, yeah, well, let us know when you've got some data to back that up. I'm like, well, this came from me looking at the data. So it's one of those things where, yes, I'm looking at the data. I have a hypothesis. We could make a split test about it. Ultimately, though, if you're bringing it up to someone and it's not something that they agree with or they don't want to do, most of the time they won't. So it's kind an interesting thing where... to an extent, you know, that what gets measured gets moved is real. And if we just make you aware of that as you're working, it's gonna kind of be there in the back of your head, even if you're not a data person. So it's really kind of about how do you communicate this stuff in a way where like, you know, especially as a consultant, like there's, don't necessarily own your decisions. I can only pull data and share if it was my place, what I would probably do. Joshua (16:27.288) But as far as someone who's creative, it may make sense, it may not make sense for them. And ultimately they're gonna have to come up with a piece of content or whatever their deliverable is. And so I find it's sort of a personal thing how much you integrate that or not. And I've seen some... some questionable things in the past that were creative maybe went a little too far. One example was a site that had a trial, a free trial, but you had to give your credit card, right? So as you're signed up for the free trial, the idea is that, you know, the first, I think it was actually free as long as you're under a certain amount of people in your account, because it was an email tool. And once you got past a certain number, it's time to pay, right? But if we have a paywall that suddenly prevents you from sending any messages, that wouldn't be so great. So that's why we needed the credit card. So as soon as you sort of grew to that level, we could just keep the lights on and then charge you accordingly. And the phrase on the page, the main headline for, you know, basically why we need to have you give us your credit card info was 00 colon 30 so you don't get a kink in your hose. And at first, I remember seeing this before it went live, thankfully, and you're looking at, I'm looking at this from a perspective of do all the things track so when it goes live we know that we're tracking out everything, but seeing that headline I was like, what I know is that when somebody is on a checkout page and they're about to give you credit card information, the last thing you want them to do is scratch their head and go, what, like what? So that was a. Brent Peterson (18:26.821) Right. Joshua (18:30.552) kind of a scenario where I flagged that one pretty quickly, like we need to redo this content. I'm not really sure what it means. I think it was supposed to read like another 30 seconds of questions, you know? And it was playful and it was meant to be playful. I think it came from a nice place, but it just, it was one of those things where how you say it out loud maybe is a lot different than how someone who's not familiar with the concept would just read that in that moment. Brent Peterson (18:59.985) How important is it to start with an outcome in your marketing before you even start? Do you start with the outcome and then work backwards? Joshua (19:10.24) I think if you do, you're more likely to have success. There's plenty of scenarios where I've seen time and time again a marketing leader says, we're going to accomplish this goal. And it's a number that everyone in the room looks around and says, I don't know how we're going to do that. And then they do it. And I don't know how that works exactly. It seems like when you challenge people, they rise up and presumably these numbers that have been tossed out would probably arrive that, you know, by some methodical method, hopefully. But there's also a chance that it's just like, if we shoot for the stars and we miss, like, well, how far did we get? And is that way farther than if we were just like, well, I guess we can just try it and see what happens. So I think it definitely makes a lot of sense to say, here's what I think is reasonable. Here's the total stretch goal. And let's just, let's just double that and say that's the goal and see what happens. Brent Peterson (20:08.549) Yeah, and see what happens, right? That's where the testing comes in. If somebody comes to you with an e-commerce site, what's like right now, what's the number one thing, the first thing you tell them besides hire me. Joshua (20:22.658) Well, probably the first thing is it's gonna be a conversation probably around what's going on with your numbers, unless I happen to have been a customer on the site for some reason and I found a glaring user experience issue that I felt like they should know about. But typically there's a question and it's like, We don't know how this got set up because there's five different people that have owned this over the course of the past five to 10 years and each one had a different idea about how to do it. We've got, if we're talking about Shopify, there's a lot of stuff in Shopify that's sort of done for you, which sounds awesome, but you can put custom code in. So you could have somebody put something custom in. and it's buried in a theme somewhere. also have, excuse me, the GA4 app is disconnected and that's tracking stuff. But because there's custom code somewhere, maybe these things are double tracking. So a lot of times it's a question of like, can you figure out what's going on here and help us have better confidence in our data? if you've been in Shopify and you've been in Google Analytics, There are scenarios where those numbers don't look similar. So how do you know if they're right or not? Some examples would be like Shopify has the data as it's coming in. it's like if you're to use landing page data from Shopify to try to get at like is it organic search, Facebook ads, what have you. that data is very much last click. And it's even more last click than the old version of Google Analytics because it's not last, not direct click. And so that can kind of... Joshua (22:21.94) inflate the amount of direct traffic that you're getting, which then has the impact of like deflating the ad performance, which anyone that's spending money on ads that has a boss they need to report to is very much like, no, no, no, the ads are important. We can't be devaluing these things. that's where kind of talking about the nuance of why you might see something different between the two platforms is important to understand. I think, along those same lines, you know, that gets us down a path of sort of, you know, multi-channel attribution kind of discussions and those can be quite complex and that's where maybe machine learning can help you kind of figure some of that stuff out as well. Brent Peterson (23:07.023) That's cool. Joshua, we have a few minutes left at the end of the podcast. I give everybody a chance to do a shameless plug about anything they'd like. What would you like to plug today? Joshua (23:18.094) Well, I guess I'll plug myself here first off. want to say thanks. Thanks Brent for having me It's been a pleasure to be on the show You can find me on LinkedIn if you just search for Joshua Lauer. There's a photo of me from about 15 years ago And and yeah, just send me a message with the words deep dive audit And I'll get back to you with a calendar link and we can schedule some time to figure out how I can help you clean up your analytics data. Brent Peterson (23:50.747) That's awesome. I will make sure I get those on the show notes. Joshua Lauer for Power. Thanks so much for being here. Joshua (23:56.024) Yep. Thank you. Brent Peterson (24:02.095) All right, well great, how'd it go for you?