Vollständiges Transkript anzeigen (5.634 Wörter)
[00:00 - 00:48] [Music playing]
Today, I want to talk about a problem that I think every founder in this room is feeling but might not have the words for yet. You keep hearing that content compounds, SEO is a long game, and that if you just keep publishing, the traffic will come. But if search sends fewer clicks, social media suppresses links, and AI answers your prospect's question before they ever reach your site, and more on that later, you can't just publish content and hope for the best. So I'll show you what's actually happening on the web, why the old playbook is broken, and what you can do about it starting next week, or whenever you get back to your desk.
[00:49 - 01:21] So, I will start by painting a picture of the world that we all grew up in as digital marketers. So, for years, digital marketing's superpower was supposed to be measurability. For search engines, we would optimize signals, get rankings and clicks. For social networks, you know, we used to post links to draw traffic. For, you know, news media and blogs, we would pitch coverage to get links and traffic. For ad networks, we would just buy the traffic.
[01:21 - 02:11] Now, you measured all of it with traffic plus attribution to signups, sales, and revenue. The squishy stuff, like impressions, share of voice, that was for the channels where you couldn't really get credit. This was a tidy system. It was predictable, legible to your CFO. But here's the problem now. We call this the "alligator graph." Impressions rise, clicks decline. The jaws keep opening. The core problem now isn't that nobody sees your marketing, it's that more people see it, but fewer of them click. So if this looks like your Google Search Console, you're not alone.
[02:11 - 02:43] And it's not just you. So this is from the Financial Times, right? Condé Nast CEO says AI is a death blow to Google search. Well, major publishers are seeing the same trend, right? Impressions up, clicks down, across the board. This is a structural shift in how the internet works now, and it's not a problem with your content. So you might be thinking, well, search is dying and everyone is using ChatGPT and maybe this particular problem doesn't matter. But search is not dying.
[02:43 - 03:22] In fact, it keeps growing. So this is, this is from the State of Search Q4 2025 report by Datos and SparkToro, my company. So since about this time last year, traditional search, AI tools, and e-commerce search have all grown. But, and this is, this is the important part, a bigger share of searches now end without a click. The pie is getting bigger, but the slice that reaches your website is getting smaller. And that changes what winning looks like.
[03:22 - 04:32] And when we say "search," most of us picture Google, but search is now a behavior. It's not a channel, and it's happening everywhere. Traditional search engines still dominate, right, about 81% of all searches. Google alone is like 74%. But look at the rest. Commerce platforms like Amazon and eBay, it's like 10%, right? Social networks like YouTube, Facebook, Instagram, 5%. AI tools like ChatGPT and Claude, it's about 3%, and scoring fast. So, that means roughly one in five searches is happening somewhere other than a traditional search engine. And on every single one of these platforms, that zero-click problem applies. Amazon doesn't want to send you traffic. Why would it, right? It wants you to buy in the platform. YouTube doesn't want people to leave. So when I say "zero-click problem," I'm not just talking about Google. I'm talking about everywhere people go to find answers. And when they do search Google, by the way, here's what actually happens.
[04:32 - 06:04] So, in 2024, or yeah, we looked at, at over at SparkToro and at Datos, we looked at the clickstream data of 2024. We found what happened after Americans searched on Google? So, out of every Google search, 41.5% resulted in one or more clicks. Okay? 37.1% resulted in nothing, the browser session just ended. The person Googled, "How old is Paul Rudd? 57, can't believe it, he looks so young." They don't need to learn more, right? 21.4% resulted in another search. But that means that 58.5% of all Google searches are zero-click searches. And so, of the clicks that do happen, that 41.5% column there, 70.5% go to organic results. 28.5% go to Google-owned properties like YouTube, maps, Google images. And just 1% goes to paid ads. So put another way, for every 1,000 US Google searches, only 360 clicks went to the open web. That's you and me, bestie. That's us. Those are our websites. Everyone in this room is competing for 360 clicks out of 1,000. And the traffic that does make it through, unfortunately, it's shrinking.
[06:04 - 06:40] So this is from Ahrefs global traffic panel of nearly 75,000 websites. I think the good-ish news is that traffic to these sites dropped -5% in the last 18 months. Paid and direct seems to be making up for some of the losses in search and social. But here's the thing, pretty much all these websites are trying to grow their traffic, right? A lot of them have marketing teams using Ahrefs. So if these sites that are trying are losing traffic, imagine what's happening to everyone else.
[06:40 - 07:05] Here's another angle. Same story. Now this is Bloom's analysis. Zero-click searches keep climbing. AI overviews have doubled. Traffic is down 10%. Meanwhile, searches are up 15%. And somehow search is still the largest traffic referrer. More searching, less clicking. That is the world we live in now.
[07:05 - 07:51] It's Meta too. So this is Meta's own data. I just made the chart. This is from their most widely viewed content reports, so from 2021 to 2025. 97.3 of all US post views on Facebook go to updates that do not include a link that sends people out. Facebook is burying posts with external links. Their own Meta Business Suite, they started advising business accounts to put links in the comments. Yeah, my friend here is nodding. Yeah, they did that. And here we thought that was just what LinkedIn bros did, right? Linked in comments. The platforms don't want to send you traffic. They want to keep people on platform.
[07:51 - 08:46] So let me zoom out and show you what the rest of the web looks like. So this one is from SimilarWeb's global panel in January 2026. Right at SparkToro we looked at the top 5,000 most visited websites on the web and grouped them into five buckets. Google alone, that's the first bucket. They get roughly as many visits as the next 13 sites combined. Those next 13 sites are like YouTube, Facebook, Instagram, ChatGPT, X, Reddit, Wikipedia. All of them together roughly equal Google. Then the third category, it's about 170 more sites, and then about 1,300, and then the remaining 3,500 sites. The concentration at the top is staggering. And most of us, we're in the bottom tier.
[08:46 - 09:23] Here's another way to see it. Same data. It's Google versus literally everyone you think is big online. And this is the playing field. This is where your startup lives. Somewhere in that long tail, competing for the scraps. So if the old game was optimized for traffic, we have to acknowledge that we're now competing for increasingly scarce attention from increasingly dominant platforms. But here's what I actually want you to take away from this data. It's not that the web is dying, it's that your homepage has changed.
[09:23 - 11:17] This is your homepage now, a Google search result. This is where people find the, they form their first impression of you. It's not your beautifully designed website, search that bit. This is also your homepage, a ChatGPT answer. Your prospect asks a question, AI gives them the answer that may or may not mention you. You didn't get a click, you didn't even get a visit. Your brand was either there or it wasn't. Oh, and by the way, I saw our friend Cole from Formspree. I'm a customer, I'm a customer via Claude. I I saw you in a Claude response. I didn't go to your website. I mean I did eventually to make my account, but I was just like, "Nope, got it in Claude, I'm going to become a customer." Right? This is also your homepage, right? What you and your employees and your customers say about you in social media. Your founder's LinkedIn post about why you built the product, your customer's tweet about how you solved their problem. The point is, your brand's first impression now happens across dozens of surfaces before anyone ever visits your dot com. Sorry, Cole, I have no idea what the Formspree website looks like. So you need to be present and credible on all these surfaces. That's the game now. Platforms prioritize native content. Platform suppresses links because they want users to stay. Your content has to deliver value in the feed itself—not behind a click. The new game: earn attention through resonance, not bait.
[11:17 - 11:33] And I'm not saying never share any links. No, I'm saying don't make click your only mechanism of value. We'll come back to this. Okay, so the web has changed, but there's a second problem that makes all of this even harder for startups. And it's about measurement.
[11:33 - 12:04] For 25 years, marketers convinced executives to invest in digital channels because, unlike offline, we could precisely measure the ROI. We trained our CEOs and our CFOs and our boards to believe that every conversion could be attributed to the channels and tactics in that buyer journey. That's over. Here's why.
[12:04 - 12:27] [Music playing]
Hey, Rob Walling here. If you're watching this and thinking, "I wish I could be in the room for talks like this," you can, and you should. So much of the magic at our events happens in the hallway track. That's where you connect with 2-to-300 like-minded founders who are trying to solve many of the same problems you are. I hope you'll join us at our next event. Head to microconf.com/events to grab your ticket. Our next one is in Iceland in September of 2026, and then we'll be in Austin in April of 2027.
[12:27 - 12:57] For reasons, I'll try to go fast. First, cookies. Only about 30% of users accept cookies, and Safari rejects third-party cookies by default. Your tracking is already missing most people.
[12:57 - 13:06] Second, ads and analytics blockers. Somewhere between 20 and 60% of browsers block analytics tracking entirely. For tech-savvy audiences, which I'm guessing is a lot of your customers here, it's my audience, too, that's probably closer to 60%. Be honest here. How many of you use an ad blocker? Yeah, look at us. We are the worst, we all do it.
[13:06 - 13:19] Third, multi-device journeys. There are about 3.6 devices per person worldwide. Forget about tracking individuals pre-login across all those devices.
[13:19 - 13:37] Fourth, privacy regulation. GDPR, CCPA, LGPD. What's possible and legal in the US is not legal most anywhere else. The compliance burden alone, that makes persistent tracking impractical for most of us here.
[13:37 - 14:06] And that's before we even get to the really messy stuff like dark social. So last year, we ran an experiment at SparkToro. We sent over 1,100 visits across 11 social networks and measured what Google Analytics reported. 100% of the traffic from TikTok, Slack, Discord, WhatsApp, and Mastodon, if you care, were, were reported as direct. No referral information at all.
[14:06 - 14:39] So your analytics literally cannot see where those visitors are coming from. 75% of Facebook Messenger traffic was hidden, 30% of Instagram DMs. Even LinkedIn, hit it 14% of the time. So if you have a Slack community, if people are sharing your stuff in WhatsApp, like in the MicroConf group here, if your product gets mentioned in Discord, you'll never see it in GA. It just shows up as direct. And you have no idea what's working.
[14:39 - 15:58] And then there's the platforms themselves taking credit for the sales that they didn't create. Now this is from a peer-reviewed paper that was published in IEEE Access just last month. This is the Dropbox study. Just curious, who, who knows about this study? Oh, interesting, breaking news. Okay, so this is, this is this couldn't have come at a better time. So Dropbox ran month-long blackout experiments. They turned off entire ad channels and measured what actually happened. So mobile advertising, this showed an attributed return on ad spend of 1.53. Search engine marketing, looked even better. About 2x. And that, actually, this looks pretty healthy. But when they measured the causal ROAS for mobile ads, it was 0.7. That means that their own mobile ads were literally destroying value, not creating it. It's not that dissimilar from their SEM performance. Causal ROAS was 0.92. Still underwater.
[15:58 - 16:38] Now that's the difference between a channel touching the sale and a channel actually creating the sale. So when Dropbox, they they reallocated $25 million away from low incrementality spend, their portfolio lifetime value to customer acquisition cost improved by 53%. Incredible. So this is a publicly-traded company with rigorous methodology that attributed outcomes can overstate causal impact by 2 to 10 times. Okay, so the web sends fewer clicks, the platforms keep the audience, and the measurement tools that we built our industry on are unreliable. What do we do about it?
[16:38 - 17:11] This is the part where we talk about what we can do on Monday, or whenever you get back to your desk. Five things. Build on rented land. Yep. I know for years we've been told, don't build on rented land, and I look, I get it. Um, my colleague, Rand Fishkin, he had over 462,000 Twitter followers, built over 16 years. I grew my account there to over 130,000. Then Elon Musk came along and you know the rest. And pretty salty about it.
[17:11 - 17:53] But here's the reality. For every, like, one visitor to your website, 100 people will get to know you on platforms that you don't control. That's where the attention is, and you have to be there. Now what this means practically, you need to publish native value on platforms where your audience already pays attention. Write posts that teach without requiring the click. And where you do this depends on your audience, right? Maybe it's LinkedIn, it's X, Reddit, YouTube. This is important. No two companies' marketing mix should be the same. Even in the same sector, right? Because you have to figure out where your specific audience pays attention.
[17:53 - 18:56] Number two, keep one owned channel strong: Email. Rent matters, but founders still need one reliable owned channel. Look at this data. Email open rates in 2005, 30%. In 2024, 34%. Click rates have bounced around between 2 and 4% for 20 years. 20 years everyone's saying, "Email is dead. Slack is going to kill email. Gen Z doesn't use email." But these stats are pretty much unchanged. Email is the one channel where you control the relationship. The algorithm can't suppress your reach. The platform can't hide your content. I mean, I mean, actually, you could get stuck in the Google promotions tab, which is not great, but hopefully most of us are marketing to other business emails, or maybe that's not as big of an issue. I don't know. But you own the list, that's the point. So yes, build on rented land, but make sure every rented land effort feeds your email list.
[18:56 - 19:50] Number three is about how you measure all this. Drop traffic as a KPI; measure what matters. Traffic can be a fine secondary metric, but it's a bad thing to optimize for unless you sell CPM ads, and I'm guessing most of us here don't sell CPM ads. Here's one of my favorite examples. So HubSpot's organic search traffic dropped 80% after Google's algorithm updates and AI overviews rolled out. This is last year. 80%. It sounds catastrophic. And yet, HubSpot's Q4 2025 revenue hit an all-time high. The stock jumped, right? Traffic and revenue are not the same thing. So if you're judging your marketing by traffic, you're measuring the wrong thing.
[19:50 - 20:53] So what do you measure instead? Here's the framework, across content platforms, search, social, and your website. Track four layers: audience. Followers, keyword volume, viewers, listeners, returning visitors. Are you building an audience? Reach, impressions, visibility, views, minutes, email subscribers. Are you being seen? Interest, engagements, product page views, comments, shares, votes, branded search. Are people engaging with you? And then sales, conversions, and conversion rate. All the things above lead to this. These are all leading indicators that should eventually lead to sales. Incremental sales lift, that is your primary KPI. Everything else should tell you whether you're on track. Now you're thinking, how do I prove this to my board? Or how do I know it's working?
[20:53 - 21:50] You can't fully attribute, but you can correlate. Track the leading indicators alongside business outcomes. So here's a simplified example. Let's say your LinkedIn, your LinkedIn impressions from no-link posts went up 52% this month. Okay? Well, did branded search volume go up? Did email subscribers grow? Did demo requests increase? You're looking for patterns over time, not single-touch proof. Alright, it's fit. Did we, did we do more of X this month? And did it, did Y go up? Right? For us at SparkToro, we know that zero-click social content, for us, is strongly correlated with signups over the next two to four weeks. We can't attribute it in GA, but we can see the correlation clearly.
[21:50 - 22:43] Right there's this other, um, B2B SaaS company, Dreamdata, right? This was maybe last year, two years ago, they started their employee advocacy program, where employees were empowered to just post as thought leaders on LinkedIn. They post about sales, you know, attribution, revenue, life at Dreamdata. And since doing that, their number of demos booked increased by a lot. They don't, they don't know like which post drove X amount of demo requests, but they know that this program has resulted in a lot more people reaching out to learn more about the product.
[22:43 - 23:51] So this is how you report to your board. And I think that this is a more honest way to do it. It's more credible. And it also proves that your organic and brand teams are doing their jobs well. Alright, number four is the one that I'm most excited about. Remember that search captures demand, but the public evidence creates it. This is so what I'm about to show you, this is a more realistic buyer journey that you're never going to see in analytics. Okay, step one. Someone sees your LinkedIn post. Then they hear you or your founder on a podcast. Then they read you in a newsletter mention. Maybe a peer drops her name or your branded name in a Slack group. Then they Google your brand name, and convert. Google gets the credit because your analytics will say organic search, but steps one through four created the demand. Google just captured it. So your job is not just to rank, it's to influence the public record.
[23:51 - 24:45] Reviews, forums, discussions, search snippets, third-party mentions, and the pages AI systems site and summarize. Because if you, if you don't influence the public record, someone else will. Okay, two stories. First one. My friend Ross Simmons, over at Foundation Inc, he analyzed B2B SaaS search results and found that Reddit outranked every vendor simultaneously on 50-66% of shared keywords. 77% of the search volume that Reddit won came from generic category keywords. So not best or review or alternative terms, and as queries got longer, Reddit's advantage grew even more.
[24:45 - 25:35] So what does this mean for you? It means buyers are forming opinions in subreddits before they ever visit your website. So if you're not part of that conversation, authentically, not spamming, right? You're invisible at the moment of decision. Second story. My friend Will Reynolds, over at Seer Interactive, he discovered that one negative review theme, it was high account manager turnover, that was surfacing 67 times in branded AI outputs. Perplexity, ChatGPT, all of them are repeating it. And by the way, he's had his agency for like 24-25 years, that was from one negative review, which at the time when he saw it years ago, he was like, "Oh, that's one person." But AI systems go deeper than humans do, so AI treated that one data point as a trend and amplified it.
[25:35 - 26:10] So, it sucked. So what did Seer do about it? They started publishing real employee retention data. And after just two citations, the LLMs stopped referencing the negative one. Now I'll, I'll admit that this wasn't super durable. Like if they didn't refresh, if they didn't refresh this data every like several weeks or so, LLMs would pick up the the old negative review. But the overall lesson is: if your strongest proof points are tracked in Slack, in your finance reports, or they're in your head, then they're not part of the public record.
[26:10 - 27:38] And if they're not published, they're not helping you. Okay, number five. This is how we do it. Treat content as a Service (CaaS), right? We're all SaaS QTs here, hopefully we like this CaaS. Now this is how I run content teams across both B2B and B2C. Every piece of content has a job to do and a client it serves. Case Study, that's for Sales, success metric: win rate and cycle time. A How-to Guide, that is for Customer Success, metric: ticket deflection and feature adoption. Benchmark Report, that is for corporate Comms and BizDev. The metric is backlinks and speaking invites. If your content can't tell you what job it does and who it serves, it's not ready to ship. So when I ran, I used to run content over at Fitbit B2B, this shift was the real unlock. Because once we started doing that, sales started coming to us first, instead of last. And our output ended up going down, because in having that tighter loop, we didn't have to create as much stuff, but our impact went way up. There were like days, months where I only focused on creating case studies, which sounds kind of lame if you're a content marketer, right? But that ended up moving the needle a lot more, because our biggest money came from enterprise accounts. Enterprise accounts don't care about your how-to blog post, right? They want the case study that shows them that your product actually works.
[27:38 - 28:44] So here's the intake form that's going to make this operational. Use this for every asset, 10 fields. If you and your internal client can't fill this out, then you can't make the asset. Client team, so Sales, CS, Product, corporate Comms. Two, job to be done. Audience, so the role, the ideal customer profile, right? What stage they're at. Number four, desired asset type, moment of use, desired action, you know, what you want the consumer to do. Source material and subject matter experts. Success metric and target, distribution plan, shelf life and refresh trigger. Right, this is a lot, but the point is, this will help you optimize for actual business outcomes.
[28:44 - 30:13] So if you can't do all this, you know, with your internal stakeholder, then you don't get to make the content. Now, let me tie all this together into something, you know, tightly actionable that you can start doing next week. Here's the weekly loop, five steps. One, pick 2-3 channels where your audience actually pays attention. Not where you wish they would click, where they actually hang out. Uh, wink, wink, nudge, nudge, you could probably create a SparkToro account, um, that would tell you where your audience hangs out. Anyway, step two. Publish 1-2 zero-click assets per week. Standalone value, no click required. Three, repurpose each idea into multiple native formats. So one insight, maybe it just starts with a LinkedIn post, maybe you get some pretty decent engagement that causes you to think about other things, so then you then, flesh that out into a longer blog post, then maybe you also repurpose that for your newsletter, and then maybe you also repurpose that for a podcast talking point. Four, capture demand. Email signup, branded search, demo request, direct traffic. That's how you collect the value from all that rented land activity. Five, review it monthly. Look at your correlation dashboard. Repeat this weekly, review it monthly. Adjust channels quarterly. This is something a founder with no marketing team can start doing on Monday.
[30:13 - 31:13] Now, one more mental model before I close. Think of your content strategy as a banking system. Right? Every zero-click post that delivers standalone value is a deposit. You're building algorithmic capital, right? Goodwill with the platform and trust with your audience. The insight, the data point, the story, the how-to, the hot take. Those are all algorithmic deposits. And then every once in a while, you make a withdrawal. Subscribe to our newsletter, book a demo, create a free account. That ratio is roughly five deposits for every one withdrawal. Earn the goodwill, and then spend it wisely. Don't make the click your only mechanism of value. Make the value obvious in feed. Make the action obvious with a link.
[31:13 - 31:52] Let me leave you with this. The job of marketing is no longer just to drive the click. It's to create understanding, trust, recall, and preference, wherever your buyer already is. The founders who embrace this early will grow faster, because they will have started building influence across the services where their buyers actually spend time. The companies that win will not be the ones clinging hardest to old attribution. They'll be the ones building influence across the modern discovery journey.
[31:52 - 32:27] Alright, that's all I have. Uh, best way to keep in touch is my newsletter. The QR code is there. Um, I send it out every Tuesday, so you'll get it tomorrow. Where you'll also get updates on my Zero-Click Marketing podcast. If you have any questions or want my slides, or if you want to send me hate mail, and don't worry, I read them all, um, you can email me at amanda@sparktoro.com. Um, is there too many calls to action on one slide? You have the call of the children don't have shoes, I apparently have too many shoes, so.
[32:27 - 32:41] [Applause]
And I'm rarely, so if you have questions... Amazing. The best marketer in the world, and she came to hang out with us. I love this for us. The worst. And now I'm just going to steal your shine, cause here I am with you. Yeah, when you answer questions, do we have time? No, but what I want to know first, like real quick, cause we only have time for three questions...
[32:41 - 33:04] Okay, is, uh, when was the last time anyone in here, like, posted on LinkedIn as a founder? Like within a week? This morning. This morning. Of course. I mean. This Alex Boyd, everyone. Uh, okay, it was like 10 people, so you need to be doing it all the time, right?
[33:04 - 33:48] Yeah. Actually, I have more to say on this. Ideally, you'll be doing this, like, 2 or 3 times a week. I think people used to think, "Oh, LinkedIn, you got to post twice a day." No, post less. Because LinkedIn, the way their algorithm works is, they're trying to find, they're trying to give the right content to the right person. They're kind of trying to tamp down virality in exchange for higher-quality engagement.
[33:48 - 34:04] And so part of that kind of means, like, high signal-to-noise. Um, the other thing is, if you post twice within like 18 to 24 hours, your previous post is going to get throttled. So, yikes. You're, if you're intimidated about LinkedIn, just think of it as, like, you only have to do it three times a week. That's it.
[34:04 - 34:49] Yeah, and you signed up to do all of this. Anyway, maybe now we only have time for two questions. Uh, we have our question boxes. We have a question right there.
[34:49 - 35:54] Yeah, speaking of LinkedIn, is it important to, or better to, post as a person or as the business profile? Which is most effective? Oh, that's a, it depends, right? If, if you are just looking for raw engagement, a person's page will always perform better. Right? People want to engage with people, not brands. Um, so the way I do it is, I treat the company page as sort of like the newsfeed. Everything the company does just goes on that page, and then my personal page is a little bit more of my perspective of that company news.
[35:54 - 36:51] If that, if that kind of helps. That way you kind of look as the company page is sort of the de facto, like this is the record of things. And your personal page, you don't announce everything, right? Like you don't talk about every single blog post your company does, but it's a little bit more point-of-view led. Nice. Who else we got?
[36:51 - 37:12] Okay. Just a quick follow-up: video or text?
[37:12 - 37:40] Depends on the platform. Um, use the on the platform like the one that is video or text first, right? So like, Instagram is probably going to prefer video and images. LinkedIn, like, I've been seeing that, I mean, obviously, an embedded video will perform better than like a YouTube link.
[37:40 - 38:09] But I will see that my reach on an embedded video is like 10x lower, or 8x lower than a text-only post. So I would say, I think especially if you're just getting started and you you just need to get it done, just do text because it's a lower barrier to entry. Yeah. Founders hate being on video. That has to be on LinkedIn. I know. Video is the worst. I hate it.
[38:09 - 38:51] [Applause]
Do we have one more? Is there one more? Yes, right here. Working on it. Thank you Justine.
[38:51 - 39:00] Thank you. Um, do you have any recommendations on how to use AI to create content? Like I, I'm sure all of us here use it, but does it get deprioritized now by platforms, or is it more like people don't want to engage with that because it's built made by AI?
[39:00 - 39:46] I I see it as, I think the platforms are trying to be smarter about detecting what is and is not AI generated. So I think it's, like, LinkedIn for instance, they've been pretty big in talking about their algorithm is trying to reward high-quality engagement, and an AI-seeming post that just kind of restates the post is not going to be deemed as high quality.
[39:46 - 40:22] So there's a little bit of that. Um, if you're asking to about, like, how do you create content or marketing stuff with the AI, I would say with caution, but use it as an editor who can give you feedback. Like what I like to do is I'll give it like a finished a finished asset and be like, "Hey, poke holes in this theory. Where do you think I need more examples?
[40:22 - 40:53] Where do you think it gets confusing?" And sometimes it'll point out like, "Actually, this is kind of weak. You need a you need a clearer example for this claim," stuff like that. So I would say do that. But the copywriting itself is, is you, right? Like. Yeah. Okay. Yeah, that's the big thing.
[40:53 - 41:15] I think the the dangerous thing is that if you start using AI to create good-enough content, you're just going to forget what good actually is. Because it'll because it's supposed to please you. Right? The LLMs are like they're they're literally designed to give you the next like statistically likely word. So it's supposed to look good.
[41:15 - 41:27] So it's important for me to maintain my voice and my IP, and just use it as, like, "Hey, can you stress test this concept?" Thanks so much for saying that. I'll Venmo you later.
[41:27 - 42:15] [Applause]
Also, hire Leann if you need a copywriter. Oh my god. What, where did they come from? The best copywriter in the world, guys. She's here in the room, guys. This, well, this has really worked out for me being up here. Can we give it up for Amanda one more time? [Applause] And now I'm just going to steal your shine, cause here I am with you. Yeah, when you answer questions, see you there.
Video
