How To Rank #1 On ChatGPT By... Gaming It?

Growth Newsletter: https://levelingup.beehiiv.com/subscribe Need marketing help? Visit: https://www.singlegrain.com/ and https://npdigital.com/ Want to recruit great marketers? Find them here: https://marketingschool.io/hire A cafe in San Francisco lets startups pay to brand the drinks, so you can order an ElevenLatte or a MongoDB Greens. Eric thinks it is genius and cheap. Then the segment everyone will argue about: an Instagram reel showing a brand that does not exist ranked number one on ChatGPT for a real query. Neil and Eric take apart how it was done, why AEO is still SEO underneath, what Kimi's download and DAU curve says about Chinese model adoption, and why the Philippines back office grew instead of collapsing after ChatGPT. Key takeaways◾Brand in the places your buyer already stands, not just the feed◾If you can game it, it is a ranking system, and ranking systems are SEO◾AI made offshore work more valuable, employment is up 20% and revenue up 30% Chapters00:00 Intro Etiquette Matters00:07 Branding in Public Places03:22 Ranking on ChatGPT05:34 AEO Is Still SEO06:16 Kimi’s Growth Surge08:41 Marketable Moments Matter12:19 AI Creates More Jobs

• 0:00Speaker 00

  Finding customer information shouldn't mean digging through emails, call notes, and spreadsheets. HubSpot Starter Plan brings all of those details together, so it's easier to track what's happening with every deal and what to do next. It pulls information from calls and emails. Then it gets to work, logging and summarizing them. Right after your call ends, HubSpot pulls next steps from your call notes and drafts suggested follow-ups. And it stays up to date for you, so you can build demand, win more deals, and spend less time on repetitive work. Your records, insights, and next steps are ready for you to act on. And you don't have to miss a beat searching for any of it. Try the Starter Plan for $10 a month. Learn more at hubspot.com. So Neil, did you see the cafe in San Francisco that allows you to brand drinks?

• 0:40Speaker 01

  No, I have not seen that.

• 0:41Speaker 00

  So check this out. So there's a cafe here in San Francisco, right? Where the startups can pay to brand drinks. So 11 labs you've heard of, right? So you can do 11 latte, right? So iced double espresso shot, it's $5.8, so $5.80. And then zero click, okay? You have rippling, there's a berry, berry ripped, okay? 21 grams of protein, 14 mango DV greens, right? 14.5, brexpresso, 7.5. And so I think this is cool. I honestly think that anything that hasn't become saturated yet, like this looks pretty saturated if I'm being honest. But anything that's not saturated where you can kind of stand out. So I have a friend that like, he likes to buy these airplane ads that like fly around the beach and all that.

• 1:22Speaker 00

  Or if you can, I think we had a mutual friend, he bought like, you know, when you land at the airport, there's like, when you're walk, the walkway is all like his, his, um, his company branding. And so sometimes you had the zig where other people are just acting like back in the day, podcast ads used to be unsaturated, where you can stand out. Or back in the day, influencer ads used to be not saturated, right? So if you can attack something where a lot of your people are hanging out. I think a cafe in San Francisco that's popular actually makes a lot of sense because there's a lot of foot traffic that goes through there. And it's people that you want to be seeing your stuff.

• 1:53Speaker 01

  Yes, and being the first is great. If 10 other startups do this or cafes do this in San Francisco, it's, it's not going to be effective for the third, fourth, fifth. Second, maybe can get away with it. Third, it won't be as effective. But the first, whoever's the first to do it, it was great. It was like back in the day, do you remember Million Dollar Homepage? Where some kid was trying to generate a million bucks to pay for, for something? I think college or whatever was selling pixel by pixel. Great model, but going and creating another million dollar homepage is not going to be effective. It was already done.

• 2:28Speaker 00

  Yeah, yeah. Marketing 101, it's, uh, it's, Andrew Chen wrote this, it's the law of shitty click-throughs. Something might be great, the click-through rate might be astronomical at the beginning, just like email in the beginning, right? Then it just goes down. Back in the day, by the way, SMS, I think SMS is still pretty high, right Neil?

• 2:43Speaker 01

  It is extremely high.

• 2:45Speaker 00

  It's still, because it's, it's still very personal, right? But also like ChatGPT, click-through rates might be very high in the beginning. And Neil, do you want to see how this person ranked number one on ChatGPT?

• 2:54Speaker 01

  Yeah.

• 2:54Speaker 00

  Okay, so check this out. So this person made a deodorant brand that doesn't exist, and three weeks later, ChatGPT ranked it number one.

• 3:02Speaker 01

  Okay, what did they do?

• 3:04Speaker 02

  Get to strangers. I spent $11 on a domain. It took me one hour of work. I didn't do any weird Reddit stuff, no fake reviews. And here's where I did that. So I've written a tech newsletter for five years. And since June, I've been mapping what I'm calling the AI shelf. So picture an actual shelf in a store. You ask ChatGPT what to buy, and there's a short list of brands it reaches for. That's this shelf, that's the AI shelf. I've run almost 9,000 queries at it. No one can see the shelf, it's opaque. But it sits in front of $6 trillion worth of global e-commerce, if not north of that. So I wanted to know, can you get on the shelf? And you can. So by day 21, I got my first hit. By day 30, ChatGPT with browsing on, named Morrowan. And directly underneath it, this very real brand, one of them exists, one of them definitely does not.

• 3:47Speaker 02

  Now, it only moved two of my 15 queries. Native still fully owns the category. I just found one little dusty shelf in the back of the store that was still open. Long tail is where the opportunity is, no surprise there. But beyond that, it reveals something about my relationship to shopping and AI that I think is going to have a huge impact. So you put something into Google and you get a list of links. You have some media literacies, you get to decide who to trust. What AI did with Morrowan is it handed the copy that I wrote back to me in its own voice. And that voice made it sound vetted, made it sound like it had some authority. But of course it doesn't. And that's before agents own the whole loop from recommendation to purchase. So pretty soon, I'm going to be telling my agent to keep my cabinet stocked. So it's going to pick, it's going to pay, trillions of dollars are going to be moving through this door that currently no one can audit.

• 4:32Speaker 02

  It's very high stakes, almost no one's watching it. I am, I test every month, so follow along. I made up a.

• 4:37Speaker 01

  So all she did was just use text on her own website and it started crawling it. And then she started doing.

• 4:43Speaker 00

  It sounds like she uses her site as like a promotional vehicle. And, um, but I don't know what she meant by like this, um, AI shelf or whatever. So I don't quite understand that, but let me tell you what I would do, Neil. I would just take this Instagram link, I would dump it into my Codex desktop and I would say, make this into a skill. And I don't even need to do anything. And I'd just send it to my team and say, hey, take it from 50%, get it to 80%, 90%, and then we're good to go.

• 5:08Speaker 01

  Yeah, I do think the copy in your website has a big impact. Because no matter what people write about you on the internet, if your website doesn't show that you do it, it is hard to end up, um, getting placed high in ChatGPT or getting placed at all. The second thing is listicles, we all know about that. The third thing is comparisons. When you compare some articles of you versus other premium brands at the recommending, that does well. The fourth thing, we've been doing a lot of this at my agency, NP Digital. We look at all the sources they're citing and we try to get those sources to just include us. See a lot of people like, I need listicles of the 10 best, you know, toaster ovens or laptops. I'm like, you do know if you type in all the queries, they're already pulling from certain sources. What are those sources?

• 5:52Speaker 01

  And why not try to convince those exact sources to also list you? That is an easier way to get cited than anything else. From everything we've tested, if you can just convince existing sources to include you, it beats everything else out by far.

• 6:08Speaker 00

  And guys, I'm sorry to oversimplify this, but to me, AEO is, there's, you have more digital PR. You have more structuring that needs to happen. And it's just more of SEO. So people don't want to hear it. And it's still, I still find it hilarious, Neil, when, when people come to work with us. And like, we got SEO down, we just need help on AEO. And that actually means they don't have SEO down.

• 6:29Speaker 01

  Uh huh.

• 6:30Speaker 00

  Right?

• 6:31Speaker 01

  Yeah, it's just like, oh yeah, what's your revenue through SEO? Uh, either we don't know, or it's declining, or slightly up, or flat. And it's just like, let's see what you're doing. Oh, this is what you're doing wrong. And this is also why you're not doing well on LLMs. Yes, there are nuances and differences, but I would say 80 plus percent is very similar.

• 6:49Speaker 00

  Yep. So let me share this one over here. So Neil, do you see this graph?

• 6:55Speaker 01

  Uh, Kimi downloads and users skyrocket. Yeah, I believe this. I believe it's going to just continually hockey stick up for a while. Uh, and we haven't even really seen the popularity. Unless, you know, something new ends up coming out that everyone ends up loving.

• 7:12Speaker 00

  I'll shed the, I'll, I'll shed some context on this one, just for people that can't see the screen. So what I have pulled up is, uh, Kimi K3, which is one of the, the best, um, you know, uh, open source weights out there. So their downloads and users skyrocket. And so you can see here, the number of downloads shot up from, this is a daily, by the way. So it shot up to 1.8 million, um, downloads. And I think, uh, daily active users shot up from about, let's say, uh, you know, 75K, 70K or so to about 1.4 million. Um, and then it started to kind of, you know, peter off a little bit, right? So my point of even showing you this chart over here, um, is when you think about the marketing behind Kimi K3, it's not like they did some crazy advertising campaign, same, Starlink, like, they, they run ads, but it's not like a lot, right?

• 7:58Speaker 00

  Frictionless Growth Marketing, hosted by Sonya Thompson, is brought to you by the HubSpot Podcast Network, the audio destination for business professionals. Frictionless Growth Marketing, hosted by Sonya Thompson, shows how modern brands grow by removing the friction in their marketing and customer experience that pushes customers away every single day, often without leaders even realizing it. Each episode delivers practical insights, real brand examples, and conversations with marketing leaders, showing brands how to win with the new rules of growth, identity-driven marketing, cultural relevance, and frictionless customer experiences. Listen to Frictionless Growth Marketing wherever you get your podcasts. But to me, real quick, if you want to acquire customers faster and more efficiently this year with the latest strategies and tactics,

• 8:40Speaker 00

  then check out singlegrain.com. That is my ad agency, again, www dot singlegrain.com. Check it out. And if it seems like a fit, we'll get in touch and help you with a free marketing plan. Neil, I, like, I've heard about people, um, avoiding certain routes just so they can fly with Starlink on the plane, okay? And then because the product is so good, and then you look at Kimi K3, because the product is so good, people talk about it being, you know, close to Fable levels, Fable 5 levels, which I think is, is a little bit of exaggeration, but it's, it's like, it's close, right? The point is, these things are getting so good, people are like, there's a clear problem to be solved, whether it's Starlink or, or, um, you look at Kimi. And, and on the flip side, like I mentioned earlier, uh, when you have such good marketing around something like a, uh, Open Claw had all the mindshare in the world, right?

• 9:25Speaker 00

  But they couldn't execute anymore, they became unreliable. And so Hermes started to take their mindshare, right? And then, uh, now it's, now you have other harnesses doing similar things, you have Grokbot, that's, that's kind of similar. And so I think the fundamental piece of it at the end of the day is this. If you have an amazing product or service, great, you can polish it. But if you have a poo poo product service, if you're, you put gold, you polish a turd, it's still a turd at the end of the day.

• 9:50Speaker 01

  Yeah, with what Eric was just sharing on the screen right now, it's another marketable moment. A16 does this. We've been doing it for a very long time and we do it in a different way than A16Z. We take a lot of our own proprietary data, our own research, our own surveying and create charts from it. Eric was showing a chart from A16Z. He took 100% of the data from Sensor Tower, I believe it was, ran the A16Z, cited the source, which is fine, and it's getting A16Z name out there as a venture fund, even though they're already well-known and they probably don't need their name out there anymore. But the point I'm getting at is you can create marketable moments for your business, whether you have proprietary data or not.

• 10:34Speaker 01

  A16Z is a prime example of this. I've seen a lot of the other charts where they're just taking data from like places like SimilarWeb. There's a lot of things that you can do like this to go and create brand recognition for yourself by just aggregating the data, putting it in a pretty format, and then sharing it with the world. You may not get as many views as you want, but if you're creating the right type of charts on the right type of data, it may really focus on your ICP and that can drive a lot of revenue in the long run.

• 11:01Speaker 00

  Yeah, we should spend more time on marketable moments because this is something I'm pushing the team on a lot more now. And we've talked about last week that every, every person is becoming part-time engineer and part-time marketer. Okay?

• 11:13Speaker 01

  Yeah.

• 11:13Speaker 00

  So I told the engineers, I said, "Hey, guys, like, look, you're all going to become part-time marketers now. I'm going to show you how to do it." And so whenever we launch a new clickflow feature, for example, I have this shared on the screen over here, you can see that we have a little video kind of outlining, this is a hyperframes video, and then people can spin up these videos very quickly. And you can see, you know, "Oh, this is what this new feature does over here," right? And then actually, this one got 26K views. It actually got 40K views on X because I did like a quote tweet. And so there's that, right? But who did I get this idea from? Well, you see a lot of it on X already. So Matt Van Horn, he's always launching new stuff. And so today, he launches an agent cookie for Grokbot, right? And he always has good music tied to it. So let me play some of the music over here with this, this video.

• 11:58Speaker 00

  Okay, so he really likes rap, right? And so he'd always send me like a new video and he'd like text me like, "Hey, here's, what do you think about this new rap song?" Right? And so it can, it allows you to inject your own personality too. But here's the thing. So Neil's talking about charts and things like that. I think charts are very good. I think videos are good too, but beyond that, I think it also depends on what you're publishing too. And so we have the skills dojo where we have all these skills. Well, the easy thing is just keep promoting amazing skills that we find on X and then add them to the skills dojo and give them credit. And we can keep promoting that over and over and over. That never stops. I can make short-form content for that too, right? And so there's that. Our API gateway where we, we allow you to connect different tools, different connectors, have different permissioning, security, governance, all that stuff. That is stuff that I can promote as well. And so that makes it more exciting for the people creating it

• 12:44Speaker 00

  and it makes it easier for us to continually have this marketing machine. And this whole concept of marketable moments, I actually got it from our mutual friend Patrick Campbell, where when he was running, I think ProfitWell back in the day, it was every month, right? This is like years and years ago, right? There's no AI, right? And years and years ago is like, every month, we need to talk about a new thing. We need to keep our audience excited. We need to, you know, get them, get them on top of this. And I think now, one of the things is, as a marketer, I can launch a lot more of these on demand because I'm a lot less resource constrained with the ability to do these things.

• 13:17Speaker 01

  Yeah, dude, we all can just do so much more that we couldn't do before. And there's no more excuses for just not getting stuff done.

• 13:28Speaker 00

  I got a question for you because we're talking about getting stuff done. This can be the last topic here, Neil. Do you think there are net more jobs in Philippines now? Or less jobs based on how the last few months have been going?

• 13:39Speaker 01

  More jobs. I think by far, there's more jobs because people want cheap labor to go and do all this stuff for them.

• 13:45Speaker 00

  Yep. So, so I was trying to throw you off by throwing in something random at the, at the end of it so you couldn't guess. I wasn't trying to lead the witness, but let me, let me show you this. So check this out. So what I have on the screen over here, this is from Trung Phan. And so he's like, "Look, since the launch of ChatGPT, the Philippines offshore industry, so 8% of the country's GDP, I didn't know it was that high, has actually grown much bigger. So employment in IT and business outsourcing is up 20% to 1.9 million workers and industry revenue is up 30% to 42 billion. So AI is helping offshore workers. So if you look at this graph over here, it's a back-office bonanza, okay? So Philippines IT and business services. You look at employees, 2016, it was a little over a million.

• 14:28Speaker 00

  And then now it's, it's close to 2 million or so. Revenue back in 2016, this looks like a little over 20 billion. It's called like 23 billion or so. And now it's, it's 42 billion, okay? So what's happening now is AI is helping offshore workers in the Philippines do new, more complex jobs. They're picking up work, training AI models, or supervising AI agents. Hospitals in America are increasingly outsourcing the checking of insurance eligibility and filing of medical records to Filipinos packing AI tools. So more high-value work, such as accounting or engineering, is going offshore in part because AI is a leveler. You can now teach someone something complex. You can teach someone complex stuff more quickly

• 15:10Speaker 00

  and it can still get done more cheaply than America or Europe. And so people might think, "Oh, man, you know, all these BPOs, all these offshoring, like, they're going to go out of business because we don't need them anymore," right? But nope, more. And we talked about this for a while, more.

• 15:26Speaker 01

  I think it's all going to be more for a while, you know? And I even watched a lot of the news channels and like, "Yeah, you may not need as many, as much memory." Well, what happens when there's robots and all this kind of stuff? Like, the memory usage will be more. I'm not saying memory, the profit for a lot of these companies will be the same. I'm just saying they'll need more.

• 15:45Speaker 00

  Do you think you'll ever fall in love with one of your robots?

• 15:49Speaker 01

  If it can clean 24/7 and have the house spotless, yes, because I'm so OCD.

• 15:57Speaker 00

  All right. Well, I think that's a good place to end, guys. We will catch you next week and yeah, have a good weekend.