Ep #159: The Next UI Revolution: How AI Is Rewriting Human Machine Interaction ft Dennis Mortensen (CEO, LaunchBrightly)

Dennis Mortensen 00:02 For the longest period of time, we had hoped that the conversational UI would be a distinct UI paradigm in its own right, but given language was not a solved science, the whole thing was a little bit handicapped. Now it looks like language is a solved science, and we can do many things that would just be really a complete pipe dream 5, 6, 7, 8, 9 years ago. Taylor Kenerson: 00:27 Welcome to the Hyperengage Podcast. We are so happy to have you along our journey. Here, we uncover bits of knowledge from some of the greatest minds in tech. We unearthed the hows, whys, and whats that drive the tech of today. Welcome to the movement. Adil Saleh 00:44 Hey, greetings, everybody, this is Adil from Hyperengage Podcast. I know that this has been quite a special time for us, as partners with a lot of CS GTM community here in New York City and and meeting some exciting founders and veterans that have, that we have always looked up to in all these years. And today is one of those folks that we are getting on. We're talking about the ex founder of X.ai. It's an AI scheduling tool that made a huge impact back in the years and now, with the LaunchBrightly that is helping customer support and technical support teams with managing their knowledge bases and product docs and product imagery and all of those experiences versus UI, like all of this. That's a big spectrum that we spoke out a lot early this week as well. So thank you very much, Dennis, for taking the time. Dennis Mortensen 01:40 Thanks much for having me. Appreciate it. Adil Saleh 01:42 Love it. So now starting off from your inception, I know that this has, it has to come a long way to where you are today and a lot of these exits and all, so what is your viewpoint on how things are changing with AI compared to you founding some technologies like in the past? Like I know that there is a component of doing it really fast and, doing it big. It's like from a commercial standpoint, how do you see it selling more? Across, across the tech ecosystem this AI thing because a lot of these noise that I had like on Mondays and Tuesdays earlier this week, like a lot of people are building a lot of things and there's not a lot of demand, like they're creating a lot of demand. But what do you think, like what's your viewpoint as a leader, this AI team? Dennis Mortensen 02:33 I do think, like many times before, so this is not, unique in, to me, at least in such a sense that we need to speak about it distinctly at some particular moment in time. I think there's always a step for where things of yesterday doesn't look like things of tomorrow. This time I do find it very exciting to see that it looks like we'll be finally getting a new UI paradigm. So for the longest period of time, we had hoped that the conversational UI would be a distinct UI paradigm in its own right, but given language was not a solved science, the whole thing was a little bit handicapped. Now it looks like language is a solve science, and we can do many things that would just be really a complete pipe dream 5, 6, 7, 8, 9 years ago for where today, you could have a freeform text input box being a very valid way of figuring out what is it that you want me to do for you, dear user. So that I find very exciting. It is so rare, right? As an I grew up on the command line, that's why I took my CS degree. My mom probably grew up on the graphical use interface. That's how she wrote her documents in Word. Perfect. My kids probably grew up, if that's a distinct UI, the touched UI that we got with our mobile phones. And I do think, and it's not like we got a hundred of these paradigms, we only have really a small handful that this new one will certainly start not to replace, but be a real distinct input mechanism. I do think it'll still be a mixed form, right? For where today, you'll touch, no pun intended, multiple kind of paradigms. You'll do some sort of command in your terminal. You'll write some sort of document in your Google Docs. You'll check your calendar on the go, to your next meeting on your phone, and you might even shout some command to Siri or whatever. Else, you might be used as your agent of today. Now, that's one thing. The other thing I also find very interesting is that there's been some sort of curve for where it's become ever cheaper. To go create that MVP or first version, which I'm a big fan of given, I'm just so pro entrepreneur that it hurts. So anything which allows two smart people who might not be, arriving with a CS degree can go create something to me, it's just fantastic. They might be in fashion, they might be in some particular odd vertical of retail, but they can now do things, that was just completely unrealistic a decade or two decades ago, where if you and me teamed up, I don't know, in 2003, we probably needed, short of $5 million to do something that looks like a version one. Just with the whole kind of thing as a backdrop. Now you and me, a couple of diet Cokes, a couple of weekends can dream up a pretty half decent MVP. So that kind of democratization off Access to kind of new ideas is just awesome. I think that, I like as well. And then perhaps not lastly, but in my list, there's a whole new set of predictions we can make with high accuracy that one couldn't really make in the past or they will be extremely expensive to make in the past. And I'm saying this as somebody who, you know, with the team of plus 100 labelers, try to make predictions in the language space of yesterday year. And it was a staggering amount, money, which we needed to put in place to make those predictions at some level of accuracy. But now we can make a new type of predictions that will allow you to create a new type of product again, that you couldn't really create before. So that I also find exciting. Now, do I think this changes everything? Nah. I just think that new technology arrives all the time and things tomorrow looks a little different than yesterday. Now is there some sort of acceleration? I think so, but that was always the case. I said it's not like my memory in my PC was linear. That was also exponential. Or my compute in my laptop that was also semi exponential and all sorts of other things. From, I just had to explain to my daughter yesterday, who's at grad school who understand it, that, hey, my first piece of equipment had eight kilobyte of memory. She understands, you know the numbers, kind of the metric, but can't compute like, but for everything. Yeah. That's kinda for everything. Both gonna handle my display to handle kinda everything. Just she can't compute, like that's like a small corner of an image on my iPhone. Like I, it just doesn't make any sense. I do think it's a very exciting time, but I'm not in that category for where I'm so awestruck that I'm confused about what the future kind of looks like. Nah. I just expect things to be more powerful, was always the case. Adil Saleh 08:19 Yeah. And this is, this has been the biggest talk of the town especially like how it's the AI impact. Is it more like optimizing the bandwidth for people? Cutting out jobs? Or is it more like creating more demand across the industries that are not well tech enabled, like manufacturing, oil and gas, all of these healthcare, all of this. So how do you see this, like both side of this spectrum? Dennis Mortensen 08:43 But was that not... Adil Saleh 08:46 Like how do you see like both side of these? Dennis Mortensen 08:49 So I'm not naive to the idea that something so dramatic could arrive. That we don't immediately find solutions to hundreds of millions of people being forced to do something else tomorrow. Tomorrow can be any day in the near future. So I do see the scenario, but when I look back, it looks like that was always the case, that we always wanted to put some piece of software in place to automate a particular process. That haven't changed. Sure. It is more powerful today, but is it so powerful that it will outrun our kind of ingenuity to have those very people do new things, which again, was always the case. So we've had plenty of jobs. Not just way back in the past, every year that disappeared or got fully automated away, and it was really super sexy jobs. It was jobs that, to be honest, we want it to disappear, and it looks like many of the jobs that we are likely to see disappear are also jobs that'll be wonderful if they would just not exist. Of course, we need to have some sort of alignment on tempo for where, once they disappear, what do those very people then do? Now, on the optimistic side, my hope is just that, some of that kind of creativity that we see strangled in every organization, even in our own small startups, right? So I'm not just talking about IBM or Citibank. I'm talking about once you get to 20 people, you have enough processes for where even yourself on some Thursday afternoon, we'll think, oh fuck this. Like what? I'm the one who put this in place and I don't like it. And we are only 20 people. Imagine when we are like a hundred or 500 or a thousand. Like they're not just strangling myself. So I do hope a lot of that possible kind of process noise that we put in place could disappear. So the very kind of idea of a job becomes, even more full of joy. And I'm saying that as a guy, who's excited to get up in the morning and don't have any kind of set number hours that I want to work. I just wanna work until I get tired, and then I guess I stop working and I'll do the whole thing again tomorrow. So I'm generally just optimistic about the whole thing. Famous last words. I understand. Adil Saleh 11:40 Love it. This is this is going so, so big, like how it is enabling people to upskill them. Not just think about, hey, our jobs will be, replaced by AI or boss and all of those, but, to upskill them with adjacent skills and, upgrade them. And this is a big opportunity for them. And of course a lot of these leadership jobs being replaced if you just only think about that too. So, perfect. So now talking about your industry, your category, I know, marketing tooling and product tooling that has gained huge volumes. In the past, I would say six to eight months, and when GPT-3 came out, it was more like conversational AI toolings, with email marketing and content and, creativity and all. Now I see a lot of these design tooling are having AI component that are replacing a lot of these big tooling as well, cutting down on tooling. So as an entrepreneur in tech for almost 15 years now. So what made you jump into this category of, I, I'm talking about a little bit about LaunchBrightly and your customer segment. Dennis Mortensen 12:49 So, I am not, and nobody should, but I'm not a fan of trying to build anything around a distinct technology. I think that's a dangerous path to be on. That doesn't mean that I can't geek out on the weekends like you or anybody else, or have fun just, pushing a few buttons here and there. Great. But if you wanna build a business, I think the only thing which you can and should do is to find a unique, distinct pain, which is true, and then try to kinda solve for that, typically with a piece of software given the industry that we are in. And that means, sure, I'll take advantages of whatever technology might be available at this very moment in time, but it might not even be the right attack angle, but the pain stays intact and the same, and I'll just attack it from the other side so I'm not overly invested in the idea that I must apply a particular type of prediction to this particular problem. I just need to see Tim or Tina be happy not doing the very thing that they didn't like to do yesterday by using my platform. So if I ask anybody who runs a reasonably sized knowledge space, hey, so you got 800 some odd articles, about three or four kind of screenshots in each one of them, so a couple of thousand of screenshots. How do you take them? Dennis. I have a demo account. I log in, I go to the right pace. I put it into the right state command, shift four on my keyboard. I got a MacBook, do my little rectangle. It lands on my desktop. I open it up, I move it into Canva or Figma, depending on how sophisticated I am. I add my annotations a little. I set a rectangle, some arrows. I save it. I log into Zendesk. I have somehow memorized which three or four articles might be in need of an update with this particular image. But I actually don't have any kind of key pair setting of this. So I'll find those articles, drag in the new image, click save. And you know what? It's not hard. 15, 20 minutes, but I do it all the time, every fucking day. And you know what, it's not even like I'm up to date. Half my images are rotten. So if I can date that task? I will love you forever and of course there's nothing better when you do a venture and any venture, you sing two songs and you would know this, right? You sing the song of the particular pain for where you hope they agree on that being a true and honest pain, not just some sort of invented pain that you imagined because you want, wanted to work on some visa technology if they believe that. Great song two, here's the solution. For where I think I can remove that pain. And obviously what can happen sometimes is that you end up having to sing the song on the particular pain for so long that you might have to accept that. I don't think this is a real pain, as in if something I thought was a pain, which can happen if you don't attack a particular vertical where you have intimate knowledge of it. Or even better that you were also in pain where I don't need to interview anybody, I can just close my eyes. Remember what yesterday looked like and say, I didn't like it. There must be other people like me. The space I am in have been very nice for where, Stop anybody in the street who runs some knowledge space and ask them, how did you take those images? Yeah. They'll give you that story I just told you. Did you like it? Nah, but it's my job and I not do it because if I don't do it, what I'm gonna end up having, this is a whole set of obstructive emails coming in, asking, I tried to add two factual authentication or tried to add segmentation on the table. Or I tried or tried and I couldn't figure it out, 'cause the bloody help article was outdated. So we spend very little time talking about that. Pain of outdated product screenshots because, Just as a side note, we put all of this tooling in place for engineers, so they're now at max speed, right? I said I'm old enough to remember to push things to production. I would FTP into the server and copy my code over as a set of files like a crazy person, but that's what you did. Not that many moons ago. Now, of course we have, CICD and all sorts of, tests put in place. And even the most junior engineer will put something to production while you and me sit here and talk on this particular kind of podcast, right? 'cause we can, so 10 times a day we'll push to production, but we're not updating that knowledge base 10 times a day. So that stays here, the product just moves further and further away. So it's been very very nice. But it's also one of those things where. I had the pain myself, and there's certainly nothing as founders, right? So here's now, but as a founder, you live in the future. So I'm not even selling now. I'm selling a version of our product that I think will be in place in about four months. My sales cycle is seven months. Anyway. I'll sell about four months into the future. Then I look at my knowledge base or other kind of collateral. They're four months in the past fuck, we are a year apart. As a founder, we are so far apart that I can barely recognize it. I know I'm a little too far into the future, but damn guys don't show the past. If anything, show yesterday, bring me up to speed. So it's just one of those things where I fucking hate it. Seeing outdated imagery and commentary about a product for where we had so much good velocity on the engineering side. And then in that kind of, help article or in that case study or on the website we show just, oh shit, not cool. So yeah, that's how we got started on it and Adil Saleh 18:51 Love it. Yeah, so Dennis, we spoke to a lot of these platforms that are directly like as chat platforms and all on AI, first, AI native, AI powered, all of these. So a lot of these folks, they focus on, like with Tier 1 support, like they're focusing on resolving tickets, first responses, and all of those. So how do you see it has an impact on the customer support, and then of course, success of the customer, because ultimately, all these orgs work together for the success of the customer. Highest KPIs being like retention and expansion, all of that. So how do you see this impact? Not just specifically talking about the product, but the use cases, like keeping those help docs a hundred percent optimized, up-to-date and make sure that they're not outdated. They're well acknowledged and explanatory. And I know that you this is pretty cool that I found like you had some annotation features as well. Like you can treat them for different use cases to make sure those help docs are slightly personalized to different industries and all. So how do you see this having an impact and ultimately customer success org? Dennis Mortensen 20:00 So. There's obviously some sort of utopia one could live in for where the piece of software that you're using is so obvious, so intuitive and so easy to use that I will never need to speak to anybody and there is no ticket coming into our system. That is obviously a utopia and it doesn't exist. It's of it's what we should aspire to at all times. But given we have users with all sorts of kind of, levels of technical ability we'll never kinda arrive in that particular place. Now nobody, when they can't figure anything out, really want to speak to anybody. So at 9:30 PM I don't wanna, even if you had live support, I could call in and speak to somebody right now. I wouldn't want to use it. What I want is to sort this particular table while I apply this particular filter, but I can't for the hell of it, figure it out. Now, I do think that the idea of having human support will become a luxury for only the most exceptional cases. I do think that the agent that we are in the midst of putting into place will solve most of this. And we've certainly seen this and perhaps it's a silly example, but I don't know, go have a look at your own history. There was a time for it. If I needed a little bit of help on some particular library, I'll go on, Stack Overflow, read some articles there to figure it out, perhaps even read the docs or whatever. Now a lot of that have moved to some agent settings where I'll just ask, my kind of agent of choice. Hey, what does that look like? We write everything in node on the backend. Can you kinda gimme a couple of examples? Like the whole thing just had me move away into an agent setting and that works actually so well, but I've changed my own behavior now. I'm not so sure we have yet fully solve it. We are move towards the goal for where I would rather speak to a machine agent versus a human agent for many support cases. So we are still trying to figure out exactly how to do this so well. But I do think there must, I think it's inevitable, that there'll be an inflection point where I'd rather speak to the machine agent, which is just more knowledgeable. I don't think. In some cases we've arrived. Many other cases we have not yet arrived. But moving towards that goal now. Tooting my own horn, a little bit. The difference I think between success and failure, which is why I still sleep well for where somebody who's clever might say, but then it's the picture just painting suggests that there is no need for that particular help article that you're trying to put in place. Pause. Sure. I don't really. My users don't necessarily need people to read the help article. They need their users to understand how to sort that table or add that user, but the only way that your LLM or whatever kind of agent which you put in place can figure that out is by you having stellar documentation. I said, what are they supposed to train on? That might be some morning where they can train on the application itself. We are not there yet, so they'll have to train on Pristine to date help documentation. That's the space I'm in. So I might just be in a space where most of the output we end up generating, which today is being consumed by humans, will be initially consumed by agents who will then spit back subsets of it to those who ask questions. So it's a very exciting space. But I do think, circle back to your question, I think we are moving towards some inflection point for where I would rather speak to a machine agent. We'll see when we get, Adil Saleh 24:06 Okay. Love it. And I also, I was thinking like workflow integration is a big thing for a lot of these these tooling. So how do you sit well with a lot of these companies, like New Age startups on the other side of the support and all they're hosting support inside Slack. And also what kind of integrations do you guys have? Like how do you guys best fit into the workflow? Dennis Mortensen 24:29 So today, so we are still just a startup. So as a startup you wanna do a hundred things, but you got money, time, energy, skill to do seven. So you need to do, to figure out, okay, what seven things can I do then? And it's just a general entrepreneurial side note. It is very easy to begin a half-assed at a handful of things. It's almost impossible to be world class at one thing, so you should at least try to be world class at one thing. So right now we have this kinda hyperfocused vertical for where we try to be really good at, One, You're creating a set of automation recipes where we can regenerate any kind of product feature, screenshot from your product at any moment and create that kind of bucket of collateral. Hey, you want me to generate 1800 fresh screenshot every Sunday night? Hey, I'll do that for you and I'll keep running that kind of process. Two, Hey, it would be nice if you can, connect to my knowledge base of choice. We connect to a dozen or so of the usual suspects and you connect your API keys and if you can then go read all my articles, extract all of the images, then run a diff against that fresh bucket you have, what I see in this kind of help center and give me a little bit of workflow in the middle where I can approve and decline all of those diffs. That's what we do today, and just do it, over and over again. Now, do we have ambitions to do more? Absolutely. But right now, if I can generate some nice PNGs for Tina, do repeatedly every Sunday night. Say I generated 1800, I found 92 discrepancies. Have a look. Ah, 70 are really just persistent drift. Or some CSS flim flam. Yeah. Yeah. Just inject them directly into the articles. Oh, Tim, damn. Why didn't product tell me that? The modal is a sidebar, or the dropdown is now set of radio button so we can, now there's a new two tabs. You promised me to push this into Slack, but you didn't, but I spotted it. Let me just pause that before I sync it. I'll need to go update the text, go beat the hell out of a product manager, and then I'll sync it in two days from now. So that process is just what we do over and over again. So we tend to play well with most. We also try to add value to this. We’re just very hell-bent on doing some nice PNGs and do that repeatedly. Adil Saleh 27:07 Loved it and have enabling teams to, use all of these annotations, post that and, integrate into their workflow. And do you have SDK or APIs out of integrations for this platform? They want to embed inside LaunchBrightly like inside their platform? Dennis Mortensen 27:26 Yeah. So we do, To kinda to get information in, as in I need to read your articles. Anything from, Salesforce to Intercom, to Zendesk, to what have you. We will ingest that and be able to kinda run the audit against that and once you approve it, sync it directly back into the article so you don't have to kind of do anything yourself. We also do integrations to, the usual suspects, where, Hey, Dennis, it would be nice if you put all these images in some S3 bucket or R2 or Google Drive or Microsoft OneDrive or something else because, I don't want my SDRs with a half-arsed talents to create these shitty decks. They should use my Pristine up-to-date product screenshots. So I'm gonna give them a bucket of 1100 images, which they can just take from. So we do have all these integrations to ingest and run the audits and kinda opportunities to put the collateral elsewhere. Adil Saleh 28:27 Love it. Love this because a lot of these sales cycle in B2B, they go with like decks and all and, these platforms even they don't have models. They like go about demos and you go to demo, they're presenting something else and maybe decks and all. So it's it's pretty big for sales decks and even business reviews as well for all the new features that, teams are pretty excited about. Yeah. Dennis Mortensen 28:54 On success and there's a thousand ways we can die in the not too distant future, but on success we become the defacto holder of your product screenshots. So if you need the core truth of what this feature looks like, go look in our platform. So I would like to kinda be. Almost like a, some will call these digital asset management platforms and there's real kind of platforms that kind of work mostly with marketing to make sure that, hey, if you want to use a logo or something else, Here are all the rules and the latest versions and what have you. I wanna be that for your software. So don't ever try to take a screenshot yourself. We would've done that for you already, and we did it yesterday. We ran a diff and I can give you all sorts of kind of history that comes attached to it, and that's an, by the way, which we've seen ourselves. So, When we run a diff and a spot of difference, I'll inject it into a history. So your login screen changes every, what, 18 months. You don't change it very often. You put it in place on day one, then you almost forget about it. Then there's other screens. Changes. You work on the feature, it might change like every three days for a full quarter. We are running little experiments. We're tuning it and getting it tight and that's just a very large history. So we keep that full history and it's a little bit. It's almost I'm sure you've all done this. So there's a history on Google Maps, right? So you can see what your street looked like 10 years ago or see what Manhattan looks like. So it's just dang. Yeah, that building did come up, or yeah, they did change the facade or whatever it might be. We do the same for those screenshots, and it's a little bit fun to go back and see that modal from two years ago. That's for amateurs. Why did I even put that in place? I thought it was proper, but it isn't. Then I see every single iteration to what we have today, which I think is slick. And I'm sure I'll have a laugh about this version when I look about it in, a year's time. But there's certainly some real value we've seen people kind of use where, why did we see an increase in confusion in this particular corner of the application, you look back and look and say, you know what? We did make a change. And yes, the change is surely technically true, solid and all of that. But people are not getting it. They're not seeing what we see here. I think we need to either revisit or, re-explain or even revert back to kind some of the prior design. So that's an interesting kind of added value. Being that full history. Adil Saleh 31:47 Interesting. And also, like I was thinking like the Go-To-Market side of things you have a freemium plan. I know this is a bigger concern, like how to turn those free to paid customers and then expand them. What kind of funnel do you guys have internally built, like in terms of measuring success across all these plans and, based on user interaction, consumption any kind of success metrics that you have. So what is the processing currently for you guys? If we just only talk about success for your customers? Dennis Mortensen 32:16 So we are still I'll talk about the ultimate success is our customer not being in pain anymore. So them not having to sit and fiddle with manual product, screenshots, memorizing what articles, what images live in and update them, once every fortnight and still have rot images across everything. So once that disappears and they enter a territory where they feel that every single article is full of fresh images that was generated this Sunday, then we are successful. Now, That's just the the salesperson speaking. On the internal set of metrics, which is not a disconnect to that story, we look at the number of reprocessed screenshots. So how many, kind of, if I look at my last venture and just as a entrepreneurial side note very early on, I'm a huge fan of being able to run your whole company on a single metric. It should be a metric that is, so core to what you do, that if you do more of it, everything else will fall into place. Money, NPS, whatever else you might think you need to go look at, those who would be good, In the future, it should be that core. In our prior venture, it was monthly meetings scheduled. In our prior venture where we did predictive analytics for media, it was predictions taken. We did a decision support platform, so any prediction which they took us up on was a success all around. This time, if I reprocessed a new screenshot, I've done a good job, as in this is what you would otherwise have had to do yourself is part of my pricing metric is part of the value metric for where if they spot a diff they didn't see before, they are slightly happier. There's some sort of ROI and they didn't have to spend the 20 minutes times a thousand times over. So that is the one core metric that goes on the on the plasma screen, where we did more than yesterday or more than last month. We are in good shape. Now, do I have the right kind of cutoff between what features goes into what edition do I have the right price per screenshots processed? Probably not. Do I have the right kind of setting for our freemium model? Probably not. They're all kind of experiments that we can run, but I don't think it changes the fact that if I just keep spitting out more nice fresh PNGs for our users, they'll be happy. Yes. I'll run some optimization on figuring out how do I get you from free to pro to business? How do I have you also not do the screenshot once a quarter, but do it once a month, perhaps do it once a week. How do I have you not do the subset, but all of them? How do I have you also, spit out, not just the desktop light English, I should spit out the mobile version. Hey, you're in the German market. I can run your whole in German. We should spit out the German screenshots as well. So that all comes back to that single metric. Adil Saleh 35:39 I love it. And also measuring the success of on the feature level and, feature to, ROI, conversion, that is also important. And also all of that revolves around that. Dennis Mortensen 35:53 As in if I do a feature ah, don't be so cocky, Dennis. Often when I do a feature it is because there was some screenshot, which I could not do for Kajabi. Ah, yes, Dennis. We actually, we run these little test sites that our users can put in place, but they run on this kind of odd, intricate iframe setting, which I understand you taking steps inside a nested iframe is a little odd. Yeah. But if I can somehow support that little odd iframe setting that you put in place, there's just more screenshots I can take for you. We go work on that, then I can increase that particular metric. Hey, you know what? I would like really to do a, all kind of portrait type screenshots that is being spit out. Yeah, I know it's very much for us, but I'll tell you why we need it. Yeah, we should probably do customized mode. So you can set up your own type of screenshot dimensions. Okay. Then I can use some more screenshots for you. So it all falls back to, You want me to do some new feature? I can see how that increases that number. Let's get cranking. And we can very easily sort it as in Oh, I got, 42 requests, but which one will give me the most? Then I can just sort it. Adil Saleh 37:28 Loved it. And a lot of these customers on the higher end bigger companies that are like thousands of screenshots going on, every week big user on the multi-product, customers. So how about that? Is there any kind of workflow automation sort of thing not on a support level, but on a success level to, measure all of this based on their journey, how they're able to adopt to the platform and how, like whether they're potentially using it. A lot of these users in these kind of tooling, they are not potential users. And you know them inside out and if they're big enough, you care about them. Dennis Mortensen 38:00 So good question, and I'm not sure we've necessarily fully nailed this, however as another side note, so in our prior life, so in our first, Third venture, we did enterprise analytics. So initially we did analytics based on log files. Thereafter, we did analytics based on, You're putting a pixel on a page, running a little JavaScript you know how that works? Very similar to Google Analytics. We got acquired by Yahoo. So we are intimately familiar with how to instrument an application and get what we think are stellar analytics for your usage. So we have our current application, heavily instrumented, have a good understanding of who came from where, looked at what, and what actions did they take, what actions would I liked them to take, and why was it that they didn't succeed or why did they succeed? And how can I use that to at least guide our kind of product design phase. We are not finished yet, but we are certainly not not blind to what's going on. But even with that said, I wrote a book on it as well, by the way, on your little bucket. Yeah. Adil Saleh 39:17 Please tell us more on that too, because audience might be listening and check them out. Oh, because this is, Dennis Mortensen 39:22 I wrote that ages ago, but it is a fun little kind of thing for, I think many of us will have “write a book” on your bucket list. I just ended up writing a little early, and that's fine for Wiley. It's not as romantic as you think, by the way. It's just hard work. But that aside nothing really beats, and again, I'm not saying anything new here, but nothing beats just looking at a recorded session for where one of your users are trying to accomplish a task alone in your application. I said there's no. Yeah, no numbers, dashboards that will humble you as much as seeing Susan stumble around trying to solve something for where you run a thousand little AP experiments, done a thousand little other things and you thought the whole thing was just perfect. And you can just see that they're not perfect for Susan. That I can see. I certainly try to do that. Everybody else will do the same, but it's certainly a, a very humbling experience. and continue. Adil Saleh 40:31 Yeah. Yes. What I've noticed, like a lot of these Europeans, like from European origin, they're so big in analytics. Like I've done 160 or 170 episodes here. And there's like barely 20, 25 platforms that came surely from Congress, from Europe, and more than 50% of their analytics like Xian X and they're building something really cool. So tell us more about what's the reason, like why you guys will hear so much about analytics and data and, visibility and, all of this. Dennis Mortensen 41:02 It's a good, it's a good question. So we built that latest venture one, the one we got acquired by Yahoo for in Budapest. So we spent four years there. But of course like, like anything else, if you get a little bit of an industry, in a particular vertical. What you're building is not just a good software, but it's a whole host of experts. Those experts will have different flavors for how it's gonna make this potentially better, and they'll branch out and do their own things, and all of a sudden what you have is just a grass root movement of people being, particular interested and mindful of how to kinda do this best, And perhaps that's that's what happened back home, but I've certainly spent a decade and a half in that in that particular space. And you never really escape it, which is nice. It's kinda learning how to.. Adil Saleh 42:06 whatever you do, it comes as a parcel. Dennis Mortensen 42:08 Yeah. And you are almost confused when you see people ride blind and not have anything in place. Also, because it comes so, so cheaply. Today, both, you gotta do the initial instrumentation and even getting some of the initial reports but still, you do see some people kind just ride around blind. Adil Saleh 42:28 Love it. Love it. So now, I know that a lot of these founders, they try to build something AI native and some sort of AI component, and I love the way that you're sitting back and leaning back and, watching like how are we everybody's performing and only focus on your industry, your product, your use cases, and trying to do best with that. Although what's your it's one thing like that makes you excited this year. I know we are pretty much done with this year, but takes a really passive and two months could be anything. So how do you see this timeframe for yourself, like from a business standpoint, opportunity standpoint? What makes you excited around growth? Dennis Mortensen 43:09 We are still early enough for where the sales cycle and the number of steps in our sales funnel are not fully cemented. So there comes a point for where you know each step by heart, you know exactly what it takes to put people in at the top and take them from step, seven to six to five to four, and so on and so forth. We are still running some of those early experiments, so don't yet have the answers. I am looking to get to the inflection point where we can start to see, patterns. Patterns which we can then take advantages of. Now we'll be seeing are dots in some sort of scatter plot. And we're trying to kinda figure out how can I group some of these dots into a particular panel. I'm hoping that end of this year and early next year, I can start to do some of those groupings, but, early days in, in that regard. So we don't have that down just yet. But that's the exciting part. And I don't, and I don't mind that part. Adil Saleh 44:12 Yeah. A lot of times you have to too much visibility is also not good, especially in the times when there's so much noise. As, you're wired like 15, 20 years down at like yourself. I'm just like your student. I'm here to learn. And even whenever we meet, my, my mission is only going to learn and take anything away from you. But my point is like 15 years down, you with wired like an entrepreneurial mindset and you think about monetizing things like in, not in a bad way, like in a good way, like costing, making business, all of that. So how do you cope with this with a startup as young as this and with having that kind of mindset and background and looking all of these folks doing around you might have met a lot of them on Monday as well, on Tuesday. So like how is that going? What is your mindset these days? Just trying to catch and people listening to, to go about business. Dennis Mortensen 45:10 I'll make this my my closing question. So I am in time for my next thing. That is certainly the scar I got from running a meeting scheduling company for six years, which is that I could never turn up late because if I turn up late, not only am I tardy, I also suggested my product wasn't any good. So I had to be in time for six years straight. I think that just yeah got cemented. I think my closing, answer to that would be to perhaps I'll add a couple of points. One, I think you need to play the game for the love of the game, not for some potential trophy you might get in the end. As in, there's no great sportsman, who came into some particular sport for where all I wanted to do was win this particular trophy. They probably arrived 'cause early on, probably even as a kid, I just liked playing football and we played football every single day with our friends, and one day you grew up and you ended up being a pro footballer and good on you, or tennis player, whatever your fantasy might be. So I think you do need to do it for the love of the sport, not the trophies. The second one is, If you do it for the love of the sport, there's probably certain stages for where the sport is more or less exciting than other stages. I just happen to be a particular founder who's very much fond of the whole zero to one part or the zero to 20 or 30 employees part or the, we have no money, no customers no hope shitty software. We are probably gonna die moment. And all of that I find almost romantic. The whole against all odds. And some will just find the whole, Hey series C. All we need to do now is scale. We need to figure out with the velocity. How much the losses we can add to this particular machinery that we have. I find the early days romantic, I think is the best word. I can I can attach to it, so I keep coming back to that even though, somebody might say, I did it once and I'll never do it again. And I'm very happy we had a good outcome or okay with us, not gonna get into the other side. I'll cross it off my bucket list and go work at McKinsey, but nah, I, I love that whole against all us part, which we are in right now. Adil Saleh 47:34 Yes founder require, and you cannot do it without falling in love with it and every single day and waking up, and that's something that drives you every single day, love it. So it was very nice having you Dennis, and we'll keep in touch for sure. And I'm sure there's so much for people to learn first and to me as well. And I'm fortunate to, have you on today. Dennis Mortensen 47:59 Time well spent. Adil Saleh 48:02 Thank you. Thank you very much. Have a good rest of your day. Adil Saleh 48:08 Thank you so very much for staying with us on the episode. Please hear your feedback at adil@hyperengage.io We definitely need it. We will see you next time and another guest on the stage with some concrete tips on how to operate better as a customer success leader and how you can empower. Engagements with some building, some meaningful relationships. We qualify people for the episode just to make sure we bring the value to the listeners. Do reach us out if you want to refer any CS leader. Until next time, goodbye and have a good rest of your day.