Maja Voje: AI Upends GTM — Where Humans Still Win | Glasp Talk #61
Sep 09, 2025
44 min read
50 views
What today’s founders need to know about AI, GTM playbooks, and the future role of humans in marketing.
How is AI rewriting go-to-market playbooks—and where do humans still have the edge?
In the 61st episode of Glasp Talk, we sit down with Maja Voje, a bestselling author and one of the world’s leading go-to-market strategists. Maja has helped 10,000+ companies develop repeatable GTM systems, worked with giants like Tesla, IBM, and Microsoft, and authored the Amazon bestseller Go-To-Market Strategist. She also founded Growth Lab, where she builds AI-powered GTM systems and has trained over 8,500 professionals worldwide.
Watch the full episode on YouTube
Maja Voje: AI Upends GTM — Where Humans Still Win | Glasp Talk #61 (YouTube)
Watch now on YouTube, Apple, and Spotify.
We talked:
(3:02) Why is PMF the most critical stage for startups, and how do you measure it?
(15:18) What impact does AI have on GTM motions, and how will strategies need to adapt?
(21:44) Will we still need human marketers in the future, or will AI handle everything?
(31:35) If AI agents start talking to each other (company & customer), how will the idea of an ideal customer profile (ICP/ECP) change?
(36:37) Is distribution now more important than product? Should we launch viral content first, then build products around it?
(46:13) With AI making it easy to build tools, will B2B SaaS have a harder time surviving?
(51:09) When a channel gets saturated, should startups expand to new channels or focus on upsell and LTV?
(53:02) What’s one piece of advice you keep giving founders that most still underestimate?
(54:40) What legacy or impact do you hope to leave for future generations?
Get a quick overview & summary
✨ Want a concise summary before diving in? Install the YouTube Summary with ChatGPT extension by Glasp for instant video summaries!
Key questions & insights
Q: What makes a go-to-market (GTM) strategy different in the age of AI?
Maja Voje argues that the rise of AI has fundamentally altered how startups approach GTM. Traditional, linear SaaS playbooks are no longer sufficient. In today's market, distribution and positioning must be continuously refined, and messaging should evolve in sync with how both algorithms and human attention behave. She encourages founders to experiment early and often, and to let data, not opinions, shape their GTM decisions.
Q: How should founders think about product-market fit and scaling?
Maja emphasizes that product-market fit is not a single moment but an ongoing process. Founders must treat it as a dynamic interaction between the product, the user, and the market context. She suggests using retention as the key signal, supported by monetization feedback and channel fit. When scaling, she recommends avoiding bloated growth tactics and instead focusing on replicable systems that link discovery, conversion, and value delivery.
Q: What is the role of human creativity in AI-driven marketing?
While AI offers powerful tools for speed and automation, Maja stresses that real differentiation still comes from human taste, judgment, and empathy. She notes that tasks like stakeholder management, brand narrative, and emotional resonance require human insight. AI can generate content, but humans must guide its tone, intent, and timing to truly connect with audiences.
Q: What mistakes do startups often make when implementing AI in marketing?
One of the most common errors Maja observes is over-reliance on AI-generated content without human editing or strategic alignment. She warns that automation without oversight can damage brand trust and miss key context. She recommends keeping humans in the loop, especially in areas involving customer relationships, strategic messaging, and brand consistency.
Q: What mindset does Maja recommend for building effective GTM strategies today?
Maja advocates for a mindset of curiosity, humility, and constant iteration. Founders should treat go-to-market as a product itself, something to test, measure, and evolve. She believes that blending AI's efficiency with human insight can unlock faster cycles of learning and market alignment. Rather than chasing growth hacks, she encourages teams to build systems that scale with intention and integrity.
Book: Go-To-Market Strategist: Everything You Need to Reach Product-Market Fit
How to launch a new product and achieve product-market fit without being overwhelmed.
This book is a comprehensive guide to your go-to-market strategy.
GTM Strategist is an actionable, step-by-step handbook packed with tactics and strategies from 54 leading go-to-market (GTM) experts, including companies such as HubSpot, Miro, Figma, Metabase, CXL, and many more.
The book guides you to answer questions like:
Who am I selling to?
What price should I set?
What marketing channels should I focus on?
What growth techniques will help us achieve this?
Transcripts
Maja: Things are changing so much with AI that this is the best time to be alive. Like, literally, the playbooks that we were using before for go-to-market are from the previous era, from the SaaS era, and things are not working anymore. So we are here together on a mission to find something that is working and will enable the next generation of entrepreneurs. Taste-making remains a frontier where humans are definitely triumphant over AI.
Glasp: Hi everyone, welcome back to another episode of Glasp Talk. Today, we are very excited to have Maja Voje with us. Maja is a best-selling author and one of the world's leading go-to market strategists. She has helped over 10,000 companies develop repeatable systems to bring B2B products to market. She is the author of the Amazon best-selling book Go-To-Market Strategist, which has become a global standard in GTM strategy. With over 15 years of experience and collaborations with industry giants like Tesla, IBM, and Microsoft, Maja has led go-to-market initiatives from startup to enterprise level. She's also the founder of Growth Lab, where she helps companies build AI-powered go-to-market systems that have taught over 8,500 professionals through her programs. And today, we will dive into Maja's remarkable journey, her insights into AI-driven marketing, and how she empowers teams to turn strategy into action. Thank you for joining us, Maja, today.
Maja: Thank you for having me. Every time somebody introduces me, I would like to hide under the table because here is what I think about myself, right? I'm not taking life too seriously. So for me, like everything has been just like a fun journey, right? I'm very passionate about this science. I think that things are changing so much with AI that this is the best time to be alive. Like, literally, the playbooks that we were using before for go-to-market are from the previous era, from the SaaS era, and things are not working anymore. So we're here together on a mission to find something that is working and will enable the next generation of entrepreneurs. And I'm just like so excited to hear and learn from you as well, because you guys have grown your subscription on your media channels to half a million people, and I just mentioned that in my country, there are only two million people, so you could have a country in Europe.
Glasp: Thanks so much. Yes. So, first of all, you mentioned like there are three stages that startups need to go through, like technically for problem solution fit product market fit, go to market fit, and scaling like expansion and moat, and let's begin with product-market-fit because that's the most important topic for many startups and…
Maja: Right. So problem solution fit is just like making something work for you, right? This is just like the will it work stage, and this is more of a technical and product challenge than anything, like go-to-market related. Cool. So, after we establish the fact that our product should be working at least to some extent, it's time to just see if we can make them work predictably for the right type of people and get money for that. That's the whole point of product-market fits. So, typically, we measure it with retention, right? Not only will comp with uh will people come to the product once, but they will be so happy, so satisfied that they will return to the product again. This is what we would like to achieve, but the journey to get there is very interesting because there are a couple of other milestones that you need to take. The first one is just like your ability to present the product in a way that the customer will understand and care about. Right? So that's a little bit of positioning and messaging. Am I saying that this is like an AI-generated text writer, or am I saying this is your future knowledge management system? This is something that I would like to test. The second one is just like we should figure out the pricing for it, right? The business model, and I think this is so important for like B2B companies in the AI space, because we actually have to pay for credits that our free users are using, right? So, whereas we can get like these awesome funds from big AI model providers, there is just a need for speed to validate that not only will people use the product, but they will also pay for it. So we know how to make money in this value exchange system. And last but not least, we have to start thinking about how to get other people to use this product apart from my mother, my cousin, and my neighbors. So, how can I reach people that I don't know, right? We have to build sales and marketing channels that will help us promote this product. And these are like the components of product-market fit that have to work together. Now this is fun to test because when you fix something when you like say okay now I will try if for example professional content creators are going to use my AI for knowledge management this is one value equation but if I said okay now I will test if the teachers will be using my solution then I would have to change all the other elements so willingness to pay budget will be different how I describe the product will be different and channels where teacher hang out would be different right so because of that complexity in my 15 years of career I never ever ever seen a company that would not pivot at least one of those elements where they are searching for product market fit so it's okay to change most people would change it five or seven times before kind of aligning those elements to something that is working Any questions about product market fits guys? But I would love to ask you, how did you know that you had product-market fit? When did you know?
Glasp: Our product?
Maja: Yeah. Yeah. For your product, did you even think about this?
Glasp: I think, yeah. Yeah. Kazuki will have maybe different answers but when we launched YouTube summary which is our main product now so it's kind of like extension that summarizes YouTube content with AI tools such as ChatGPT, Claude, Gemini, and when we launched so it went viral on social media and so our traffic was like literally hockey stick, J-curve and when we had a bug so trouble in the product so we got so many inquiries from users. So we couldn't address it at that time. Then we noticed, oh, people are really using our product, and if the product is missing, they are disappointed. So maybe it's Sean Ellis' test, right? So that's when we feel the product-market fit.
Maja: I love this story. So, we established something, and super congratulations, like this is a dream of every entrepreneur that you would launch, go viral, and then figure out how to do the support and all the other stuff. It is a very big scenario. However, for the majority of entrepreneurs, as you know, traditionally, 95% of businesses fail not because they have product issues, but because the majority of them, like 80% of those who fail, have business issues. They have the wrong pricing, the right ICP, and the wrong way to do marketing and sales. So what I'm saying here is literally that with product market fit, you sooner know that you don't have it than when you have it. Because when you have it, you are worrying about other things in business, and you're just trying to make it work the best you can. You're not thinking a lot about product market fit as consumers are returning to the product again. Cool. So the next stage is super interesting as well because at that stage, we have to build sustainable marketing and sales systems. We call them GTM motions. So these are predictable and repetitive ways to get customers into our products again and again and again. And by now, like the majority of companies that I work with, have AI-assisted go-to-market motions, right? So last year, everybody started to prototype to test, like either different purposely built solutions or tweaking their own stuff. So there is a lot of AI assisting us on that journey as well. But typically, how I define those go-to-market motions in my book is inbound, meaning you create content and then, like you create so good content that people discover your product as well, for example. So, it's you communicating with an audience, and then the audience, based on the quality of this content, gets attracted to the product. The second one is outbound. This is what I love. Most people absolutely hate it. But if I had to, like literally get five customers tomorrow for my business, I would be doing outbound. Okay, maybe now I would send out a LinkedIn post, but that requires an audience. But yeah, for most people outbound warm. So to the people that you know, either recommend introductions or if they would like to work together, or cold. So, based on your knowledge, who could be benefiting from your product, who should be buying it, and send them invites to discover your product. This is something that most people hate doing, but I love it, and we can talk about this a little bit more further on. The next one is paid. Now paid is getting increasingly interesting because right now the velocity of ads given, like all the advancements in AI, everything is like AI-generated, and one of my projects or one of my partners literally outperformed in on LinkedIn ads, which performed worse than AI-generated creatives, but like twice as badly. So I really think that just like in terms of ad making, we are there at least for the design and copyrighting part for the setups, we can still discuss this a little bit, but yeah, this is to a large extent automated right now. Cool. Next, we have something very difficult to automate. These are communities. Communities are especially important for tech products and for products that have, let's say, customers that are early adopters, because we cannot shut up about the latest and greatest tools that we have discovered. It's our pride and joy. So yeah, communities are super important, especially in the tech world. And then we have something that is very much like outbound, but a little bit more complex. This is where like the sales kicks in. how we call it like accountbased marketing and then like accountbased sales when you are doing the selling and unlike outbound when I don't know exactly to whom I should be selling this to and I'm still sending this email out to like 300 or 1,000 people with different domains that have been warmed up before not from my prime email with ABS and ABM so with the like bigger sales systems we have a very clear definition who our ideal customer profile is. We have a list of companies that we would like to work with. It's a shorter list usually like around the the smallest one that I have seen was 10 companies but usually it's around 100 something and then we are creating these multiple touch points ways how to first attract their attention how to get them on the meeting and how to sell them something surprise that should be expensive so that we are literally covering the costs of everything what we are doing before. And last but not least, there is the favorite kid on the blog. This is like the new um go-to taste. So that's PLG. That's product-led growth. And just like this notion, try it before you buy it. And then when people are happy with the product, they would be inclined towards committing to more packs, to higher packages, or even recommend it to others. And here is where we have a little bit of an issue, like we talked about before with AI products, right? Because you don't want to cover these credits yourself for an infinite amount of time. You need to start monetizing immediately and especially if you have like if you're a bootstrap company. And if this technology is hard to implement, it's better to first start with a control pilot to earn the right to automate and then expand into the account. Okay, guys, did I geek out too much here? Because this is like literally it's such a broad topic, but if you have any questions, any thoughts, any experience, let's just break it down a little bit more if we have to, yeah.
Glasp: I love that, yeah, and also I follow you on Substack and social media. And you share an image of GTM motions, and you also mention that you miss the partners, I think partners.
Maja: I miss the partners. How could I? Because, tell you what, because I cannot automate them with AI yet. This has been like a pain in the ass for me to just like do a little bit more like AI-assisted partnership programs; there are purposefully developed solutions for that. But you're so right, like big companies such as Clay that recently got like 1 million investment, and HeyReach. HeyReach is a really cool company from Europe. They do like email marketing on LinkedIn. So they are a tool for sending out LinkedIn DMs. They have created like 7 million yearly recurring revenue in just like 14 months or something like that was insane, right? And they were both using agencies as levers of their growth. So agencies brought their customers and helped them with these setups. So that is brilliant. Kazuki, I think that you need a certificate or a medal for correcting me. That was great.
Glasp: Yeah. Should we edit out? Okay.
Maja: I love it. Okay. I forgot one. Excellent. Why? Because I cannot AI it.
Glasp: Oh, thank you. Yes. So you already mentioned about AI the impact of AI, but nowadays it's been two to three years, but people use ChatGPT and less Google, and click on Google results, Google search and how do you see the impact of AI in, let's say you mentioned GTM motion, so GTM strategy how has it changed and how should we prepare for? Some people say that the AI engine, GEO, a generative engine, optimization matters to us more in the future, and yeah, could you tell us a little bit about the impact of AI?
Maja: Yes, whereas I'm not an expert in that specific subject I have seen like huge results from just like my AI first companies that I work with so a lot of them are discovered through just like AI engines instead of Google right and they have been like doing very interesting experiments how to get even more of this traffic one of the companies that I work with, they produced like over 100 articles for indexing in just like June. And after everything was deployed and they tweaked keywords a little bit, they jumped into the top five in their categories suggested by the AIS. So right now, I really think that it's a fun time to explore, like mentions on Reddit being referenced by other creators, making sure that you are just like multi-channel and that you are paying attention to what's going on there. And one of the best tools that I recommend to check that is Brand Trader from hrefs. I have been using this one. I will soon publish a list of B2B companies that are best indexed in my vertical. And yeah, that has been super fun for me. I have been working with experts who have been having great results with it, and I have been experimenting with doing experiments for my business myself, but I don't consider myself an expert on this subject. So, okay, and Kazuki, if you have some good practices here to learn as well, do you do some GEO experimentation? What is working for you guys?
Glasp: Yeah, we could, yeah. Okay. Yeah. First of all, so we do AI-generated content. So yeah, based on YouTube content, we tested it out, and since the YouTube transcript is not on Google, it's not on the Google search engine. So we didn't copy all the transcript, but based on the YouTube transcript, we made an article, then it's going to be indexed on Google and we see a lot of traffic from Google as well as you know ChatGPT or Gemini and I'm not sure if it's directly related to user acquisition but we see traffic so that's one thing.
Maja: And one of the shortcuts, for like at least what is working for my clients, is to create these comparison articles, right? So, it's like your Google versus competitor versus competitor versus competitor. These are very popular for quick indexing as well. And yeah, people who are solution-aware usually think like these are the queries that they would be using. So, let's link a couple of excellent GEO articles in the comments so that the audience can learn more about this. But I hope that we just get people thinking that this is important, and this is getting increasingly important, and once ChatGPT launches their proper shopping. I think that even ads are going to be broken in Google. So yeah, Chip could be changing in a mega platform. They are playing the cards really, really, really nicely. Brian Balfour said that first to give like full reference but I'm waiting for agents so hard because the only thing that is missing for me like on JPT are MCPS and just like agents I would love to have like a mega platform where I could do everything so I don't have to go and like have a cluster of seven different tools that I'm using. Oh my god. Sorry, that was a little bit hardcore in terms of like geeking out about stuff. You can cut it if it is unnecessary.
Glasp: Oh yeah. Yeah.
Maja: I like that you had an oyster shack. You're okay with this type of subject.
Glasp: Yeah.
Maja: Okay.
Glasp: As I mentioned, so yeah, it's unscripted. So yeah, we're happy to learn, and also, yeah, we check AHF almost every day, and we see a lot of backlinks from SaaSHub, which generates comparison articles. So, I think they automatically generate content. So just by comparing similar products, we see a lot of traffic and links from them. Yeah.
Maja: Yeah. I love AI-generated content. I have been using it a lot myself because I'm not native and wherever like I have to produce like high quality output of course I will hire an editor but for majority of my LinkedIn post post formatting and just like stuff that is derived from like webinar transcripts or something like that I think it would be a waste of time and just like a huge bottleneck in velocity if I wouldn't be using these type of tools. So the way how I usually like start with the implementation of these tools is I start to think a about the workflow right so what is like my workflow looking like where am I wasting the most time what do I hate most and then I start to build a knowledge base so that I don't leave things you know to the garbage in garbage out system but I know that the system is like taking information intelligence from the carefully created knowledge base then it's just like prompt engineering and fine-tuning it and making sure that we have frequent feedback loops with it so that it hopefully gets better and better and better the more we are using it and this is exactly my love relationship with ChatGPT the problem is that I'm using it for so long that if I started with clo right now which is objectively according to like my colleagues much better in content creation I couldn't because by now I'm completely like stuck there like it's so well trained it's working amazingly well for me and I feel that I'm locked in.
Glasp: I see. So related to AI, do we need human marketers or human growth managers or, you know, product marketing managers in the future? In terms of marketing or growth, a marketing strategy is needed. So, for example, Google can do everything. So they have Chrome, they have Google Analytics, they have VEO, and they can make an advertisement. So eventually, all of the marketing activities can be on Google, one platform, and so machine learning can also generate content. So in that sense, will we need human marketers or growth marketers?
Maja: Okay, whereas I love AI, I get pissed off daily when I get like chat, jeoperated briefings, chat jeopardy generated blog posts, and called emails that were not read by a single human being to review. And my usual response is just Have you read it? Like, seriously, have you read what you have just said to me? Just like this, taste-making remains a frontier where humans definitely triumph over AI. The second thing is stakeholder management, especially when you are working with larger orcs and you have to coordinate like gazillion stakeholders right there. Okay, I have a terrible joke. No AI is stupid enough to handle us. But people like to communicate with people, and there is an emotional component to it, and just like this sense of proximity, developing the relationships we have emotional needs as well. So, as long as people are transacting with people, I think people are needed, but the important question is where? So, where humans ' human intelligence outperforms artificial intelligence, some people would say, like in strategic thinking, but I have seen better GTM strategies generated with ChatGPT than I have seen slides before. I have seen a huge improvement and advancement. I do believe in the art of coration, though, because at the end of the day it's all about trust and proprietary intelligence. Even if the model was trained on all the data that you can grasp on the internet, you as a human still have a proprietary taste, proprietary experience, and just like what my weird people in marketing would describe, the vibe, the energy, right? So I do think that, like I like to chat with ChatGPT, but also I get very mad because he's reinforcing me for things that I should not be praised for, such as great job you completed one exercise in August. I like you should be saying what the f like you should be like exercise more and stuff like that. So yeah, at this like interaction, it's just not there yet, as we are comparing it with humans, but a lot of technical founders, such as you guys, must be like product people, all right, more technical people or marketers.
Glasp: me technical, more like product technical.
Maja: You.
Glasp: I'm more business side, I would say, yeah.
Maja: Cool, and I would love to hear it from you, how do you see it, because I'm probably like 10 years older than you are, and I think that, like, management is very important, creation is very important. I'm trusting AI that much. I would love to see, just like hear, how you think about this, where do we need humans in marketing and sales?
Glasp: If it's B2B sales, so humans are needed for sure, right? So, it's like, as you mentioned, stakeholder management for sure, but we aren't working on B2C, then yeah, as long as the AI can understand the whole concept of our product, and so it doesn't hurt our brand or marketing, or you know, people, so not sure if people are needed in marketing.
Maja: Definitely my brother is in ecom and he has like I think that they are makings like 10 million a month right now I don't know I would have to check with him exactly so they are sell selling like men's wear and he AI generates everything before he had like a very big creative team for doing the ads now like every creative that they are testing out is like um I generated. However, he said influencers when I as this question, right? But influencers like organically the communities that they have built, the trust that they have built. So this is what he's still paying for, having content creators creating like the creatives as well as doing marketing and sales for them. Sorry that I interrupted you. It's just like this example.
Glasp: Yeah, I agree with this community and content creators like influencers are important. I would say the community should be human, but not sure about content creators or influencers. Because there will be VTubers or like AI or digital clones, you know, influencers in the future. Not sure if people would like to see this kind of content, but eventually, technically, they can do it. Or I can be replaced by my digital clone. So, I am talking at this Glasp Talk, and I can be a digital clone, right?
Maja: Right. Yeah. Yeah. It's the intimacy. It's something else that we need, right? How about you, Kazuki? Are you firing? How are you thinking about this?
Glasp: Yeah, I think in the future, as Kei mentioned, I think AI can take care of more like a marketing part and content generation, as you know, many startups already do, and then see the results and learn, experiment, and do another test, like kind of iterate in the process. I think AI and AI agents can do most of the parts. And regarding the AI clone, I love the Reid AI. We talked to the creator behind Reid Hoffman's AI, and kind of that's like Reid Hoffman himself interviewing with the Reid AI and an AI clone of himself. So that's a really interesting concept and so that I think the new way to spread knowledge and insights and with more people because let's say if I want to reach out and talk to Hoffman like a famous figure and I think he doesn't, I don't have access to him and he won't make time for me but if I can talk to AI maybe I can ask many questions to him and his AI clone and so that I can you know learn from him. In that sense, it's a new way to distribute and share knowledge.
Maja: Scale your brain, right? This is super interesting. Thank you so much for highlighting this. Really got me thinking, however, then I would like to think that for every trend, if we can like AI, everything, there is also an anti-trend, right? And it was so funny because I have a cousin who's much younger than I am he just celebrated his like 17th first day and we were figuring out like really hard it was hard question for us as a family to decide what he will study for high school right so the boy didn't care he was like I don't like school and we had present him a couple of options like how this world will evolve how he where he can have the best chance to succeed right and we were like literally in the dilemma the adults in the family between like him becoming a developer and he becoming like a carpenter so somebody who is making stuff from wood right and he then decided that he wants to work with wood I think it was a very smart decision because these guys like at least here in Europe they are making some serious cash are they building the next unicorns no but lifestyle based on ambitions that he has displayed so far these jobs are going to be like super AI safe and I just think that like as we are optim optimizing everything with AI, we have this need as humans to connect with other humans somehow and maybe intimacy and maybe like this labor that is not appreciated right now will become important because I love to think that for every trend there is an anti-trend. Maybe then I can have like a bunch of grammar mistakes and my post will look human and people will be, yeah, your human posted. Probably not, but it could be.
Glasp: Yeah. And I recently saw an interesting post on X or LinkedIn. I forgot where, but now AI generates content and sends it to someone's inbox, and AI reads the content. It generates somebody, and no one leaves it. And what's the point of the process like that? Yeah. So, and also.
Maja: Yeah, I've already liked this on LinkedIn, to be honest. Like I'm living in a dilemma, like should I comment lata lata-generated comments and then like other creators say to me, yes, a comment is a comment, it still counts. So sometimes I just have a little bit of fun with it.
Glasp: Yeah. And regarding, back to the the topic about product market fit and ICP or finding like ECP in your voice you know customer profile and as you mentioned like if let's say we assume more companies or startups use AI or AI agent to reach out to like a clients or negotiate in you know like situations in that sense the concept of early customer provider ECP changes because of AI because okay if we reach out to client then if clients use AI it's not we talking to humans we talking to AI and AI decides if they should pass it on to human or not so in that sense I'm curious about how ECP the concept changes over time?
Maja: Okay, I think that you are like one step ahead of me in this thinking. Here is where I am at the moment. So beforehand how we were building this like outreach list or lead list right we said hi I want these companies from the US that got 10 million founding in the last three months and then like the system speed up a couple of like email addresses and that was pretty much it then we were like emailing them hi I'm Maja I'm really good at making GTM you sure need GTM because now you have founding let's talk better but you know know the logic right we were doing it like based on their graphics so this is an AI company based on signals they recently got founding and then like let's hope for the best and let's account that I can personalize and do outbound copyrightiting really well the problem is that everybody has the access to these signals right so like as company gets founding it will get 200 of these emails because this signal could be bought by anybody. What is really interesting now to explore is where we can find proprietary signals. So instead of me saying okay, I sell to like car repair shops in the US with at least 50 locations, I would be thinking who, and I don't care about the profile right now. I just care about the idea of them being interested, of them displaying the behavior that my AI will recognize as purchasing behavior based on previous patterns. This is how I could start creating my proprietary predictive intelligence on who would be interested in my products. But we are not there yet. I mean we are in some installations with some like bigger implementations but for majority of folks especially like pre-product-market fit the safest ways how to do this is a you just like develop a case study and then you say to people who are in that vertical high I saved this energy company $600,000 would you like to see how you can implement AI agents to your support systems that would be one way and the Second way is to just like forget this prime idea of segmentations and go broader and reverse engineer what was actually going on. But based on the parameters that are more behavior-based than this, like typical OOH product managers, liked it. Let's make sure that the next iteration of my marketing and sales is all about product management. Because people who like it, I don't have to force this label on them. I could be like treating them as separate units. So this is where I am in my thinking at the moment. I have no idea how this will transact as our like agents will be interacting with each other. But since they will be trained on our knowledge access assets from our past call recordings and stuff like that, I think we can figure it out.
Glasp: Definitely. Yes. Yes.
Maja: We'll see. But it's a fun time to work. It's really fun.
Glasp: Yeah, definitely. Yes. Thanks. And yeah, so I was going to ask something. So, regarding the burning needs, in the B2B space, the startup needs to find the burning needs so that users have a huge problem. They want something like an immediate solution so that we can sell the product to them, right? But in the AI era, I think it can be will sorry, sorry I was going to ask something, but yeah I forgot, but yes yeah okay then, have you ever seen the case of Cluely? So I was curious about the GTM also distribution. Usually, people start making a product with their founder's mission or burning needs, or after a user interview. Then, they make a product, find a product market-fit, then move on to go-to market strategies but if like GTM or distribution are more important, then, why don't we focus on making like viral content on TikTok on YouTube, then, launch a product related to it, see the metrics, then oh yeah this is a use case where people use then why don't we adjust it so it's kind of like GTM strategy or distribution product like, how to say, hierarchy is different or changing recently? How do you see the market?
Maja: I love it, like seriously, this is how we were doing testing back in my hardware days, so we didn't create products hardcore because that was too expensive, just like the molding, the tooling costs you like 50k, even like before you started to like even think about having this product for the prototype. So yeah, we did a lot of those, like fake gateways when we just like did a render of a prototype of the product, and then we attracted a bunch of traffic there. So we bought like Meta ads that was pretty much it and then we were just like measuring conversions cost per conversions and then saying oopsie this is not like available or we like gave refunds to people to just like demonstrate that there could be a sustainable what we called back then funnel before we got to the decision if we will invest into making this or not. Now, as you mentioned, and I think this is genius, distribution is the queen. I am approached by, and probably you are as well, by probably like five or 10 makers per month that they would like to just piggyback on my distribution channel, which is insane in terms of opportunity but also in terms of the business models that we are developing right as content creators. So the reason why I started to build my audience is not that I don't have anything else to do, and I would love to spend 20 hours a week creating content. It was to do demand generation for the book, right? Because I have put a lot of love, a lot of thought into the book, and I wanted to succeed, and I saw LinkedIn as audience building as a nice way to we can create a sales channel from this. The rest is history, but this was the beginning. So I literally had to build my own distribution channel for my products as well. And now, as it is evolving, it's just like a business decision. How do I want to proceed with this business? Right? There are a lot of opportunities. There is advertising, there are sponsorships. These are my products. These are like consulting businesses. I could be building like mine as well, or co-branding them. I'm sure that it's the same for you. So, just like building the audience first and then launching the product is already in use in many different places and forums. What is super interesting is that not as much with tech products as, for example, in ecom or even like I explained in the hardware space. So it's a completely legitimate way to think about this as long as you are connecting this with a business model, which is a strategic decision that you must take as a founder, and when it comes to visions, I mean, I always admired people who knew what they would do in their lives. I still don't like seriously for me everything is like a really fun journey and it's just like this wild exploration thing but some of the people are just like wired differently like at when they are 18 they know this is exactly what I want to build this is my vision I want a factory that will produce chairs or something like that I never had that so in a way I was kind of forced to just like think okay how to do marketing before ever building a product practically all my career.
Glasp: I see. Yeah. Interesting and insightful. Actually, I was about to ask What is your GTM strategy for your media? So I prepared that question, but I think you are building an audience first and trying to find a product.
Maja: No.
Glasp: Because, yeah, you are doing Substack, you are publishing a book. So what is a GTM strategy?
Maja: Okay. So first, you have to understand my business. No, you have to. I will explain it to you anyway if you are interested. I have three business units. The first one is media. So the metrics there, of course, are the side of the audience, but more importantly, in my case, the quality of the audience. You will not believe how many times I get like, do people get into my DMs and actually ask about these implementations that we are doing? Have I really tested this? What were the results that they tested it, and they were a bit disappointed by this one, right, so like the intimacy of me and what I'm how I'm connected with my audience is much larger than just like oh I will send you this like letter once a week and you have fun with it or not, so yeah. This is like not the size game; this is the quality game, if you ask me. By now, the second thing that I'm doing very passionately is consulting workshops, just like the standard service business, right? I never wanted to call it an agency because I do have a little bit of an issue with how big the teams should be and whatnot with the AI changing and stuff like that. So yeah, I never wanted that. But for the consulting, we have been working a lot with corporate ventures, for example, lately, and with whisies, this is super super super interesting because you learn new stuff and you meet a lot of people. So that's again feeding the content engines, and then we have the digital products business unit. This is like the funnel. These are the online products. So there is a book, there is a checklist, and there is a master class at the moment. But I'm thinking really hard about how to transform this into something that is going to be subscription-based. But I don't want to be built like a community or paid Substack because I don't think that this is like a proper and it's a proper way to monetize like I know people that are making 200k per month but just like on Substack it's a proper way to monetize and it's not my preferred way of doing business. Right? So, as a founder, this is not something that I would very much want to have in my business. It takes like a certain work that's not very near and dear to me. So, yeah, I was thinking a lot about how we could do the productization more aligned with what is going on on the market right now. And I'm sure I have said this publicly before if I hadn't added prompts into the checklist in February I think that I would be having huge issue with digital funnel monetization right now because the way how we do go to market even in early stages everybody's prompting I cannot say to you now we will fill in the spreadsheet together and we will spend four hours filling it a dock people will go nuts like seriously the way how we expect to work change so much that we have to but and that is really important for my generation remain relevant. Remain relevant. Are we still relevant? This is the question that we should be asking ourselves, but not every single day, because that would be depressing, but at least once a quarter, right to really ask yourself if what I'm doing is still relevant to the reality that I'm in right now.
Glasp: Mhm. Yeah, totally makes sense.
Maja: And what are you thinking? How will you monetize? That's a super interesting discussion with other creators as well.
Glasp: Yeah, the monetization one idea is already done. But you know, so since you have a framework, you wrote a book about GTM, go to market, and you have a framework, and you categorize startups into each category, right? So you interview founders as you do, then share the podcast and also newsletter, kind of use case, kind of newsletter, then having the sponsorship, that's what I think is very debatable, I think the monetization system, I think you can do, I think you're already doing, but you know that's what I thought.
Maja: If you ever want to interview a guy who sold like a newsletter to HubSpot, I have some very good contacts, very, very, very great exit strategies for ike niche subtext and newsletters. It's really interesting because for companies, I'm just doing a newsletter project with one of my clients, right? And we have been creating this mood board. What are our people reading? What is relevant to them? And we actually find way more creatives that come than company newsletters. Why? Because, like, company newsletters are usually just like goddamn boring promo. They are not really great at knowledge distribution. Most of them are, but the majority of them are struggling to keep things interesting. So yeah, there is also an opportunity to exit your newsletter if you want to. I don't want to have the mental capacity bandwidth to deal with this right now, but it's a fun, fun thing to think about.
Glasp: Yeah. And sorry, now I remember my question. Good to know. Yeah. Sorry. Yeah. Because now thanks to AI many companies can build I mean especially startups can build you know tools by themselves using AI so meaning if they have a burning needs they can build the tool and solution by themselves using AI coding tools and so on not using they don't need to actually they don't necessarily buy the tools from others they can build it by so in that sense I think the threshold or cut off or the burning list is getting lower and meaning isn't it how to say Are we having a hard time in B2B SaaS space because if anyone can build the solution by themselves thanks to AI that's my question.
Maja: Yeah, I love this question, and let me walk you through a couple of examples in which cases I would definitely love to buy a tool over building my own tool first. So I definitely cannot buy as much data, and I like to train my own signals. It wouldn't make sense for me, business-wise, and I will happily pay for the access to curated databases and just like train on the best examples in order to do my outbound, so that I'm not messing things up. The second thing is just like, and this is a brilliant example. All of my examples are B2B, though, because in B2C, I do imagine that things are different, but you will be the experts in this one. Okay. So the second example in B2B is actually the call intelligence, and this is the question that you will have to answer as a founder. How is your solution better than ChatGPT? And if you say something unconvincingly, if people will have like the slightest idea that this is something that they can DIY, that you don't have any differentiators, this is a problem. So, where to get these differentiators, and here I would just like to state a company's momentum. So they are a call recorder in a normal category-wise, but they are like actually GTM intelligence, right? So they are connected with your CRM. They are like training their own models with their own agents. It is so much more. And I actually asked Jonathan because you know I'm a very poly person and I asked him like oh but okay I can upload my recordings to my GPT and I can ask it questions too and he was like haha haha right I was completely ignoring the concept of knowledge base of how like agentic and like how artificial create intelligence is created how important pre-training is how important post-raining is in this type of questions. So I love this definition. It comes from OpenAI that ultimately, there are three differentiators of AI products. The first is the model itself. Do you have proprietary data or synthetic data, or are you like user data or something like that that would give it another output? The second one is close integrations in workflows because if I have a CRM, it's very much more likely that I will use a call recorder from that specific CRM than if I had to pay for another thing or something would not be compatible, right? So close and frequent interaction in workflows, and there is the third one, which is feedback loops. Imagine when I am like, have I mentioned that I'm using JGP for content writing instead of J code, which is legitimately better. Okay, I'm hooked because I have been spending so much time with it. I have been fighting with it. I have been voicing it. I have been prompting it. I have like made it work so hard that now I have created so many feedback loops that, for me, it's the ultimate best tool to create AI-generated content in my specific use case. And these are like the differentiators that we could be considering. So, build versus buy. These are the differentiators. Can that make sense? Is that an asset that we can create? And on the other hand, is it cheaper to just like buy it and pay 10 bucks for a month for it? What is my building cost? How frequently the usage is? Right? Because potentially yes. I could be building my own. What could I build? Okay. Extractors for leads from LinkedIn. I could be doing that, but then like if I have to go back and fix everything, then something has changed on the LinkedIn algorithm that would create a lot of overhead for me and my team. So this is not my core business expertise. This is not what I want to do in my life. It's much easier and better for me to just like have somebody else change this lead extractor whenever LinkedIn changes its algorithm.
Glasp: But in the future, will it change? Do you do you think?
Maja: Yes.
Glasp: Thank you. Is your time. Okay. So, just yeah, sure.
Maja: We should be wrapping up. If you give me a two-minute breather, I would very much appreciate it to just like read what David read to me.
Glasp: Okay. Okay then. Yeah. Before wrapping up, I have one more question. So yeah, you are mentioning a product-market fit to GTM, then scaling. But eventually, if the channel is saturated you we need to think something differently. So is that something like we need to expand to different channels, or do you recommend a product or company to do upsell, or like increase lifetime value? Do you have any framework?
Maja: Yeah, the quickest and the easiest way for me is usually an upsell. Can I change my business model? Can I, like, change my packaging to charge existing buyers differently or more? That would be the first thing, the easiest thing to test. Then, like opening new markets either geographically or like demographically in B2C stuff. This is interesting to explore. You can always be doing testing with Meta Ads and with Google Ads. I don't know what works better for you and on social media. So that's something, and then like the first thing that we usually consider are new product launches or massive product improvements. So it's either like a channel, it is a business model, it is like a new market, so a new ICP, or what else, what else could we do if things are not working well?
Glasp: There's that time. But yeah.
Maja: But we'll be running out of time, right?
Glasp: Yeah.
Maja: But this is how I would be thinking. So, like when you have a channel that is saturated, of course, you should always be testing, like at least 20% of your time in new channels if you see like diminishing returns and something declining. So you should be prepared for this. It doesn't happen in a day. And then consider entering new markets, new endeavors, and new product launches. It's always fun in GTM. It's never boring.
Glasp: Yeah. Yeah. Thanks. I remember that's what Sean Ellis also mentioned. 80% put your effort into existing channels, 20% explore new channels always. Yeah, that totally makes sense. Yeah, the number is nice. So, we really. Oh, yeah. Yes. Do you have any advice that you have already shared many times? Do you have any like what's one piece of advice that you give repeatedly to founders or but feel most people still like underestimate? Do you have any?
Maja: Yeah, think about early monetization because if you let a bunch of people free into your product, you could be optimizing it based on the feedback of people that will never pay for it, and that could be potentially problematic. And the second one that I usually just like this is a little bit of a more intimate one is take your time to worry. So there are so many things that will be grown every single day, and you don't have to be reactive. It's actually the worst thing to be reactive. You, as a founder, need to be firm in your personality. You are a leader of your team. You should not, I mean fragile. You should be honest or something like that. But just like working on your personal stamina and make sure that you are separating whatever is wrong in business not to mess up with your performance right here, right now. Make sure that you are super present, that you are focused, and that you can say no whenever you discover that you don't have the resources to do it today or tomorrow, because it's very unlikely that next week is going to be any different than even more busy. So learn the self-mastery of just like handling yourself as a decision maker in charge of many humans and AI.
Glasp: Yeah, priority and focus always. Yeah, that totally makes sense. Yeah. And this is the last question. So, since Glasp is a platform where people share what they are reading, learning as a digital legacy, we want to ask this question. What legacy or impact do you hope to leave behind for future generations?
Maja: Okay, so I would just like to make this go to market less mysterious, right? I don't think it's rocket science. I think technical founders can totally DIY this. I don't think that marketers should be as concerned as just like well, will we be replaced by AI or something like that. I would just like to talk real and make this journey a little bit more manageable and comprehensible for future generations, and it was for me I was afraid of developers for 14 years in my career, now they are reading my newsletters, ha uh so yeah just like to make this work a little bit more manageable and comprehensible for folks.
Glasp: Yeah, I think that's a great go-to-market strategy, I think.Yeah. Yeah, thank you, thank you so much for joining today. We learned a lot.
Maja: Thank you for taking the time. It's very late when you are, and I'm so impressed by your story. It was great talking to you. I think that we can create some awesome value-added for your audience together.
Glasp: Yes. And for definitely. Yes. Yes. Thank you.
Maja: Thanks so much. Cool. Take care, guys. Good luck with everything.