Follow the clicks and see the flow. Where do your users really go? Tune in right now and get in the zone. The Mouseflow podcast is on. Yeah. Let's go. Welcome to Mouseflow Podcast. Today's guest is Eddie Casado and Eddie is definitely not a stranger for me. We might as well say it. You have worked at Mouseflow before, in a in a good role, and we still, miss you in in many levels, not just from a work, experience, but also from a people person. So it's so great that you wanna participate, here. But today, you are a performance marketing lead at the Lucen Software. That's correct. So instead of me presenting Lucen Software and yourself, I was thinking that you could tell me a little bit about you and the company. Absolutely. Well, I'm honored. First of all, when I got your message to participate on the podcast, it was without question that I wanted to be here. It's been a while since we've seen each other. It's been fun to catch up with it before the podcast. So now we're just going to get into it. I joined Lucen Software almost four months ago as a performance marketing lead, managing all budget for paid, mainly Google Ads and Microsoft Ads. We are a Microsoft native suite of tools that allows you to create three different things inside the the Microsoft, environment in Office. One of them is our timeline product, which allows you to create timelines within, PowerPoint and connect them to several project management tools to keep your timelines updated, which is super helpful, especially when you're doing board meetings and decks and whatnot. It's just... It it saves a lot of time and effort to keep those things updated across. We have another product called Track, which is a time tracking tool, also Microsoft native that connects to Outlook and, several other, tracking integrations, to track time and bill time and invoice time. And we have a third product that just recently launched called Flowchart, which is at par with Timeline Works within PowerPoint to help you create flowcharts with our, newly released Lucent AI product by prompting away and just making easy, easily editable and also easily to update across the organization. So that's the spiel. That's what we do. Beautiful organization, a lot of clever people, a lot of driven people. And yeah, and I'm happy to have joined the company four months ago. And I'm ever since then I've been just hacking away at performance. It's been a while since I managed large budgets and campaigns that are this robust. Most of the revenue is still hinged on purely PLG motion. However, we do have SLG motions as well for one of the products which is Track and currently picking up as well on that. Cool. That sounds good. Great. Well, thanks again for joining. And I think that the interesting part also with you is that, I mean, you're not just pure performance marketeer. You come from growth, experimentation. Basically, you think about the whole full funnel, not just what's driving clicks. You go in and see what is happening when performance and product market trend aren't aligned. And, you know, know, you look at it, build system processes to basically fix it. And then, of course, from my end, you know, Mouslo super well, if not better than I do sometimes, still probably, because you've been so much in that also behavior analytic, part of it. But yeah, because many performance marketers has, you know, a reputation of being very much obsessed over the clicks and click through rate and so on. But what I'm kind of interested to hear from your side is with all that experience that you have as a from growth to performance marketeer, how do you actually tend back to revenue and, you know, do you what does a typical process look like in practice for you? Good question. And yes, I'm not the typical performance marketer to my, to the joy of my imposter syndrome. I I don't consider myself a performance marketer to the dot, especially because I know that there are people that are, like, very much driven by PPC frameworks and are very like, if you go into LinkedIn, think you can find 10,000 people that are doing things differently or or clever than you. But I think that diversity is part of a virtue. Right? And and I do think beyond the funnel of just getting the clicks in, especially today. I don't think that any any performance marketer today should be running pure PPC clicks campaigns. I I don't think that's where the where the value of the platforms lie today. They are KPIs. They are metrics to look at. I still look at them, but they're not what govern my decision making. CTR to me is just whether or not something is being relevant or getting people's attention. CTR benchmarks, especially in Google has skyrocketed for the past couple of years. I remember that when I was back at Mouselo, a 3% CTR was something that was good. I am seeing six, seven, 8% CTRs today. And it just begs the question about whether or not that's the right metric to look at. And when you look across the funnel, sometimes I've always been very much obsessed with the fact that I needed full visibility over the funnel. Like, we cannot stop at looking at how many clicks we're driving. We need to look at what those clicks are doing, which is where Mouseload came in in my career. And I was like, okay, this is great because now I know what... What's happening inside the landing page. Now I know what I need to optimize for. Now I know what do I need to bring over from the ads into the landing page in order to make it relevant. So that's the... Those are things that I still look at. But, know, if you ask me what are the top metrics that I look at today, One of them will be search impression share, the suite of search impression share metrics, for budget and rank. Especially rank has given me a hard time recently along with quality score, which has been a subject because it's been very much, changed since I since the time I was at Mouse. Yeah. So those are things that I swear by like beyond just the clicks or even the price of the click. I am not too obsessed over those metrics, which is when somebody that it's a pure PPC marketer would say this guy is absolutely bananas because you need to look at your CPCs. You need to look at your clicks. I'm not saying you don't need to. I'm saying like, that's not what drives me. A $2 click is not the same value as a $4 click. It's not the same audience, especially if you're doing smart bidding. So you need to also make sure that you need to know what you're allowed to pay per click, but you also need to know what effect that's gonna have on the business. And that's why ultimately outside of the platforms overstating conversions, which they do, they all fight for the same conversion unless you have a good marketing mix or a modeling mix for the marketing conversions, which are expensive today, although with cloud and AI, you can also build your own. All campaigns, all platforms are gonna fight for the conversion of the credit. So ultimately, end up looking at ARR Sheets and Fincore and Chartmogul and and and then Growpanel and all these tools that just tell you what is the bottom line effect. Now the caveat to that is that you cannot react as fast as you would before. That is the problem that I see today is that unfortunately, and I hate giving the excuse that that's how the platform works. That's just, I feel like I'm gaslighting stakeholders when I say that, but it's the truth. Sometime yesterday you could just launch a campaign and have data in three, four or five days a week. Now you have to wait two weeks, three weeks for the learning phase to start over. And God forbid you make a change because that starts it all over again and the data will continue to optimize and learn. So what I found is that you also have to give yourself plenty of time and be patient, which is uncommon for performance. That is another big challenge that I found. You can't just go back to your CEO and say like, I need a month to tell you whether or not this is the right campaign to to be running at this price. Because you still have a thirty day conversion window and you can't assess the numbers until that window closes. I will argue that that's always been a difficult thing to say to a CEO that you need to But definitely, it's now even more waiting and more changes. Yeah. So as you also mentioned, like just the slide, we can't just do 10 different changes at once because then we won't know what's working and what's not. But it is it's a difficult run, right, because it's a it's a time or race against time, I would say, sometimes. And, especially in the world that we ironically live in today, with the market and especially with AI, I mean, we are for now not seeing necessarily, you know, I've heard people like saying, oh, maybe we will see that we can work only four days per week instead of five. But I mean, that's well... Yeah. Human beings are just racing to be Sweden. Right? Wants to have a not going to happen. We just make more things happening in five days and let's see what that means for the world. That's a whole other story. But yeah, the same goes a bit for for these kind of things. But, you know, it's you also mentioned like how I've like how I approach things and as growth and experimentation. I mean, me, everything is growth. Like every every the rest of the performance is channel for me. Like I approach everything as growth. Experimentation is another channel, another vertical. I obviously don't try to be everything at once because I've learned that, you know, it doesn't matter how much you know, you really don't know until you meet people that are in the trenches, like especially with CRO people, for example. I I thought I knew them when we did that campaign back in the day for CRO until I said, ran one interview and I was like, I'm definitely out of my depth here. This this is not about AB testing. But I think that one of the things that helps a lot is to have a strong framework for growth, especially when approaching performance today. Because of AI, you can have now very comprehensive logs. If you run like I, for example, run all my campaigns to Claude Code, I am connected to supermetrics, everything's pulled from supermetrics and I have my whole system there. And one of the things I like about working with Claude Code is that the context window and the memory is far better than mine, especially for remembering the stuff that I should be checking on. Right? Because now, yes, you have a lot more capacity. You can do a lot more, but that also comes at a cost. And the beauty of AI, at least for me, I was telling you, what a time to be alive, is that now I can have a better log of and recollection of everything that I've done and better pacing for the experiments. So for me, performance as experimentation and performance as growth has allowed me to actually scale campaigns better than if I were just running welly nilly updates on the campaigns and just not caring about the pace of the changes or not allowing the campaigns to mature enough. So that's one thing. Like, if I could leave anybody with with some advice is run Claude Code. Make sure we run our campaigns from it. Yeah. It's amazing, right? Yeah. It's I mean, sometimes I'm saying, what was I doing nine months ago without cloud? Well, I mean I do this. Two years ago, were just, like, amazed at ChatGPT for making images. Right? Yeah. Exactly. Yeah. Full suite of a full suite of Everybody that I talk to that has asked me, How do you think things have changed since you last managed Mouseflow ? It's like, I've become Bionic. Mouseflow was all manual. Competitive analysis, landing page, heuristic analysis. It was like me spending time looking at things and faking it. Like that anymore, I would say either. We also have no idea. Know. Not throwing shade. But back in the day without AI, it was just like, we were using AI to maybe validate some copy and get some ideas. Now it's Proofreading. Proofreading, right? Or translations, right? So those things, those were the things that we were doing before. But now, you know, we have skills and we have all of these things that it just makes you go so fast. But everybody's going through the same struggle. This is what I wanted to get at. Everybody's struggling with the platforms. Everybody's struggling with what is the edge that you need. And for me, the edge is understanding how AI works. It's pattern matching at best. We haven't reached generative AI. So you need to take it as as for what it is, what is it good for, but also have a good framework behind you because everybody is going to have access to essentially the same information. But what's going to change is the human mind. If you have a good framework and understand what you're looking for, that's when AI enhances your work, not the other way around. So that's at least what I've found until today. Look at revenue, bottom line, you should be driving revenue. That's the purpose, unless you're driving some other type of campaigns. But if we're talking PPC or the old the good old PPC, that's consideration at best. Definitely decision stage of the funnel. You should be driving revenue bottom of the funnel. So that's what you should be looking at. Now, it just takes a month in order to understand whether the changes you made worked. Whereas before it was, you know, like the good old days of last click or first click and just like modeling attribution modeling was just so easy. Right now it's not that easy because you lose, I don't know, 30, 40% of the clicks and click IDs along the way. And for example, most of my time today is spent reconciliating different data sources and figuring out what is the next level of configuration for us to actually weaponize the data that we have and make better decisions. But even if you get that data matching to a certain degree of confidence, you'll still have to wait a month to see if the campaigns actually work, which is the dreadful part, at least for me, because when you're talking performance to a CMO like you or you to a CEO, it's hard to understand how things have shifted. It's hard to understand that it's not $1 in, $1 out. It's not that the revenue influence... I remember when we were working together that slashing campaign budgets and it dropped organic and direct traffic altogether. You can now just attribute one channel revenue and neglecting the influence it has over the rest of the channels. And that's the tricky part, which unfortunately you learn making some mistakes and actually experimenting. Yeah, exactly. So what do you what have you have you experienced over the last year? I mean, I've seen it, our traffic basically increasing, but I also have experienced that I get more and more unassigned traffic. So, basically, I don't get any page views. I just get a lot of sessions. Have you experienced anything like that where it it's difficult to see, especially within AI search? Is this people? Is it bugs? Is it trolls? Is it an LM that's kind of reading something because they actually don't do any page views? So it's a little bit ironic in that sense because, you know, the traffic has increased, but it doesn't mean that my conversion has increased. So how... Have you have you seen some of that... Those, yeah, changes? And how do you kind of work with it to basically figure out, is this a real person behind the computer? Yeah. Well, to to be honest with you, I have experienced a surge in bot traffic because Lucen actually went through a rebrand in May. We were called Office Timeline before and we move into Lucen Software. And ever since that migration, a few blocked IPs for some reason because of the migration started scraping some of the content for one of our products. And I only found out because we had a search in traffic. And my VP was like, can you find out what happened? Because obviously the conversion rate dropped. So that was one instance. Now, when it comes to LLM traffic and the unassigned, I know that the unassigned has been an issue for Google Analytics for a while now, especially with the with the increase of of LLM traffic. Those definitions I am not that close to right now. Have a webmaster that manages and also a couple of ops person in the team that manage that traffic split. So I can't speak to those. What I can tell you is that my brand and content team has a tool for measuring mentions in LLMs, which can actually help you kind of close the gap between are we getting mentioned, are we getting recommended by LLMs or not, and get you... Give you ideas on how to actually appear for the searches that people are making in LLMs. So like, think of it as, you know, questions that, you know, AnswerThePublic, was it the name of it? Like the tool, I think Neil Patel bought it a few a few years ago. It's the same for LLMs. And the tool is called, Cognizo, not affiliated by the way, but that's the one that we're using. And it just gives us that information about how many people are actually searching for the relevant keywords in LLMs. And that information comes back to me so that can do ads on Tagipity. And then the content team actually creates the content to answer to those queries on the LLMs. And that kind of can tell you exactly maybe there is a proportion or a relationship between how many unassigned sessions you have versus how many times you're coming up in LLM searches. But I I I really can't speak about that whole area. I've been very much focused over the past four months and just getting a grip on the campaigns and making sure that the campaigns are firing well and conversion tracking and so on and so forth. Not not my... I don't think I can speak well. But I don't I'm not even sure Google Analytics know exactly either. It's still a question mark for many people. Yeah, you know what? You know, I don't like Google Analytics at all. So... But but, hey, you know, it is it is what it is. But I think there are some... Obviously, you know, I I found Reddit to be super helpful and raw. So it it may be worth a question to a larger group about how they're dealing with on sign traffic coming into Google Analytics. And if there's a way of classifying those in in a way that can help you your reporting because the the the bigger that chunk, the less the conversion. Because those are like how to's queries. It's not essentially a conversion query. And and that that brings me a little bit to... So some of the the the part here of, actually understanding, these ads and how do they actually become customers, that gap between I clicked the ad and, now, hey. Cool. It's a customer. Sometimes it's super difficult to see what's actually been the travel to that, and you can't totally necessarily see the value. So I'm kinda curious on how you think about in general optimizing that part of the funnel. How do you look at it? Yeah. Good question. It's part... It's it's been part of my existential crisis ever since I joined Lucent because I think that the answer is not... I wish I could give you an answer that was straight on as to, like, how to actually measure it to the click. I think that the mindset has to shift a little bit along with the user journey. And the reason I say this is that for example, both Mousel and Lucent have low ticket, considerably low ticket cost. But some of the companies that are buying this have still an enterprise motion to go through. Even if it's a small plan, even if it's one license, they still have to go through procurement. And this is when things start to get messy, because the same person that is trying the product is not necessarily the same person that is buying the product. It's broadly the case, actually. Because it has to go through, in your case, a CMO or accounting or the tool owner. Very few companies that I found are that the user actually buys a product. Very few companies, especially when you go into this, let's call it in this environment in which you have a low ticket item with enterprise mid market procurement, which is far more strict, where you have to have everything tied down. Thankfully, that's one of the things that AI has allowed me to do. Before, I couldn't actually get to this data without having to go through a developer or knowing actually Python or code. Right now I can just speak to the data. So a lot of my time have been spent understanding the efficiency of the campaigns and understanding what's actually running revenue, driving revenue, at least directionally. And what I found is that most of my revenue cannot be attributed to a single campaign. And then upon digging, I started looking at the domains that were buying and then I could see a 10, 20% lift of looking. Okay, so the assumption or the hypothesis is Eduardo at office timeline dot com started a trial, but Lauren at officedimeland dot com ended up purchasing, there must be a connection. That's at least the hypothesis. And when you make those connections via AI and just reconciling the data, that's when you see, oh, okay, so last click or first click, depending on we use first click at Lucent Software, we can see that X amount of money was driven by ads. But when you actually look at the domain purchases per domain net new purchases, you can add 10, 20% more revenue because that could be one of the journeys that people are following. Not to say that it cannot be. Thankfully, I have a great PMM team that is working on researching what the user journey looks like so that we can understand the data flowing in. So beyond just having the systems in place, you also have to understand the journey. That's to me, that's the biggest thought block because then the conversation changes. You also have to be comfortable with the unknown. That's something that I've found because back in the day when we were running PPC at Mouseload, it was just click data that we had. Then I remember server side came up and ConsentV2 came up and everything went sideways. We didn't know what was heaven or hell. And we had to figure out, but sitting in that inconsolable place of like, I don't even know if that's a word, but sitting in that place of, man, now my data doesn't speak well about what I'm doing. It's it's it's very uncomfortable. It's very uncomfortable. That's something that I've come to accept. Like, I cannot attach a dollar amount to a to a to a purchase because if I can even if I could, I'm neglecting the halo effect that the campaigns have. And that's that's something that is very true. If you shut down one of your channels, you'll see the other channels go down because they do have an effect. So what I'm doing today is reconciling all my data, making sure that we're running server side conversions and offline conversion. That has been one of the biggest unlocks for me is to use offline conversions, however you want to do them, because it's what's going to enable the platforms to actually learn more from the actual conversions. That's interesting. Because if you if like that's that's what's been the unlock for me and also taking advantage like as much as I wanted to fight Brock match and as much as I wanted to fight smart bidding, it's here to stay. And I can tell you now from my own experience, I'm still not doing Brock match, by the way. But I can tell you from my experience that moving to conversion value bidding has been a godsend. Like I just moved one of my campaigns to conversion value bidding a month ago. And I could also I could see a 5% bump in activation inside the product by just feeding the right information to Google and to Microsoft Bing. It also allows you to recover a lot of conversions that happen to have dropped the click IDs because it matches the email as well. That lets you also teach the system what's happening beyond your UI and feed that back to the algorithm. Now, the mindful step here, as I said earlier, is that they will overstate conversions. So if you have a like a robust marketing mix of LinkedIn, Microsoft, Bing, Google Ads, Meta, TikTok, etcetera, etcetera. They will all fight for the same conversion. And that's when you have to do the modeling yourself to adjust where you're going to bet and understand your funnel. So this is a long way to say that it's not easy. Optimizing, I cannot optimize anymore for ads that drive revenue. The ads remain on the CTR, search impression share, and in page on page conversion side of things, but not revenue. Because I it's it then the numbers look too tiny and then your campaigns are all underperforming. You need to look at the broader scheme of things and understand if you increase your budget, how much does that increase your ARR? And then under understanding that dynamic, you just increase your budget by 30%, but you only get an 8% bump in ARR, that tells you there is some efficiency to be gained of 22% somewhere to actually get that those numbers to be even. And that process for anybody that's listening is not a month's worth of work. No. Or two months or three months. So I want to advocate for performance marketing marketers and growth people in the world in front of their bosses that this does need time because it's one of the issues that I see today. There's not enough patience to understand that we are learning as we go. And now there's a whole other discussion about whether or not you're clever or smart or like whether you're like in it to win it kind of mentality. But we're we're all learning. And it's rarely that somebody is going to have a straight answer. I've worked with agencies before and I've worked with freelancers before. They all do wonderful audits, but it isn't until you get into the account that you start figuring out why things aren't working. And you need to have, like I said, a good growth framework to experiment, like experiment well, put your hypothesis forth, understand the changes that you're making and what impact you're expecting to get out of and making sure you can defend those. Whether you win it or lose it, it's just a matter of learning. It's always about learning. Learning, like learn as much as you can. So if you had to basically well, you almost already said it, wrapping up here, with your best, like, advice or top three, I'm not saying that you have to say three exact, but you already almost said some of it. Like, you know, you need to let the the algorithm and the the campaigns run before you basically take some of those decisions. If you had to give one last advice, what would that be? Oh, always tough questions. Well, know that one I can never say just one thing. Yeah, you are allowed to say more than one. I always pivot. I've been asked this question before. And we are bombarded with so much information today. And there are so many people doing great things. And unfortunately, the window we have at the world is something it's LinkedIn and it's never, it's never the healthiest window into what really happens on the day to day work of everybody. So my first advice is be kind to yourself. Understand where you are. Understand that you're learning. Be humble. Be humble to just understand where you are, where we all are today. I don't I don't think there is a single person that can say that they have mastered this. I think there are very good people out there that don't rest until they have achieved results. At least that's how I see it. People call I mean, AI today, I mean, everything is changing also ad targeting the way that creative optimization needs to go faster. Is a difficult place to be. You have one foot in one place and you have the other foot where you need to be patient, right? Correct. It all comes at a cost. If you're really career driven like myself, it does come at a cost. So I always try to leave people with that, you know, as much as I can fake it here. And you could think of me as an expert. I don't consider myself an expert at all because I know that there is a world of things that I could be learning. Like, I'm not a traditional PPC guy. I'm not a traditional growth guy. I'm not a traditional anything because of, you know, that's how I am. And that's who I am. I like to I'm super curious and I'd like to learn new things and, know, be comfortable enough to move things forward. So that would be the first part. That's like, let's call it the soft skill advice, you know, and then when it comes to hard skills. It it it is a cliche, to say, hey, you know, leverage AI. But to me, it's it's hard to give a hard skill today that doesn't rely on leveraging AI. It's very hard. Mean, I can sit here and come up with something, but, you know, I'll give you example. And this is something I've publicly said a few times. I hate an Excel sheet with a passion. Like, I can't like my brain doesn't work that way. My brain can't think of rows and columns. Some people do like them. I don't have to like them anymore. I don't have to be forced to like them anymore because it's all about the data. Now I just export CSV, boom, that's it. And then I can have AI do the visualization that I want that my brain can comprehend. So beyond just saying like, hey, AI, leverage AI, use this tool, use that tool. I think it's just a matter of understanding yourself and how AI can enhance and make you bionic. Don't pay attention to whatever else everybody's doing, the skills and the GitHub repositories and yada yada. Just try to find your own rhythm, try to find your own need and how that need can be met by AI. That is the hardest, softest advice I can give you. And then if you want to have it be about, you know, hard skills, spend time on your copywriting. It goes wonders when you do spend time with actually copywriting. That's the easiest cop out right now is that people just do AI copywriting. Do spend time going through your stuff. Train yourself to spot AI slop in your own work, not other people's. I am an expert in spotting AI slob for everybody but myself. And that's a skill that I have to build up because it's unfortunate, but it is true. And I'm sorry to say that, but it's, you know, it's my skin. So but it is one of the things that I would say and that every organization is struggling with. The output of documents right now is 100x across all departments. And I have seen how one single AI generated report can derail a whole conversation and strategy meeting. So be comfortable with recognizing that AI is a tool, but you're the ultimate responsible for the output. And with that, I think that I'm going to wrap it up with that very wise sentence that you ended up with here. So thanks, Eddie, for joining this podcast, and thanks for sharing all this wisdom with us. It's been a pleasure. Always a pleasure. Always a pleasure. So thank you. And... Yeah. We'll keep in touch. We'll see each other. We'll keep in touch. Yeah. Thank you, Dorothea. Bye bye. Bye bye.