I Tested Sol 5.6 vs Fable 5 on Real Marketing Work (Clear Winner)

0:00Neil Patel:Sol 5.6 versus Fable 5. What's better for marketing? Yep.

0:06Eric Siu:So I basically ran a test and your mileage may vary. In fact, I think your mileage may definitely vary. So Neil, have you done like a head to head test yet or no?

0:15Neil Patel:I have not.

0:17Eric Siu:Okay, so I'm going to speak to this and then we can react to this. So I basically, so I ran a test with Sol 5.6 versus Fable 5 to test which one's better for marketing. and I actually was really surprised by the results. And so Sol 5.6 is ChatGPT, obviously. Fable 5 is owned by Anthropic. These are the frontier models as of this recording. And so when I tried Sol 5.6 for an analysis on the best, if I were an investor, what bets would I choose if I were running Amazon? Okay, so Fable 5 said it would bet on the same-day grocery service for Amazon, right, as like an investment. They would like bet all in if they're running Amazon.

0:59Eric Siu:Sol 5.6 was like, I would bet all in on the Amazon, the supply chain service that they just launched recently where you can basically use their infrastructure to ship out more, ship out goods, right, using their infrastructure. And then it kind of outlined how big the industry was. So I'm just saying from like a technical analysis point or like an investment analysis point, Sol 5.6 chose the one that had the bigger opportunity opportunity whereas fable five missed the the big opportunity so that's one that i reward to to solve 5.6 to start okay so number two what i did neil was i had and i wish i could share my screen right now but um i had fable five made make a website and i saw 5.6 make a website the thing is solve 5.6 it made the website which is like a scrolling website meaning that every time you scroll your mouse down it was zoom in into this world the world would get bigger and bigger and bigger.

1:49Eric Siu:And I did it in one shot. Fable 5 kept asking me more and more questions. And so as I continue to run more and more of these tests, I realized that Solve 5.6 is like the employee that you just tell it to do it and it just does it. Whereas Fable 5 always has to keep asking you questions back and forth, back and forth, back and forth, right? Which is really annoying. And by the way, Solve 5.6 is cheaper too. The third example I'll give is that when it came to making thumbnails, I just told Solve 5.6, I was like, hey, take the transcript. So for example, Us recording this marketing school right now, I can just take the transcript and I just say, hey, Sol 5.6, look at my YouTube channel and take this transcript and I want you to make the thumbnails.

2:27Eric Siu:And literally we published those thumbnails and they're getting like a 6-7 % click-through rate, which is good for B2B. Whereas Fable 5, it failed, right? It needed a lot more context and it needed a lot more connectors. And again, it was just like, it was too much back and forth, right? So I can't remember the fourth example off the top of my head right now, but based on my, I read four different tests, and keep in mind this is just one person, I found that Sol 5.6 was not only cheaper, but it got the job done faster without all the back and forth.

3:01Neil Patel:So you would say Sol is better?

3:04Eric Siu:I find myself using Sol 5.6 a lot more now because Fable 5 is a lot more token hungry and by the way, I'm not going to say that Fable 5 is bad lad you can definitely use it, it's really good for planning it's really good for adversarial code reviews and things like that but since I've been using ChatGPT more in the last few months with Codex I find it to be more convenient for me and the skills that I'm making, all these things ChatGPT's image gen is really good as well Fable 5 and Claude in general I don't use that harness as much and this might be different for someone that uses the harness a lot more when it comes to Claude

3:41Neil Patel:yeah dude yeah it's in AI I've always seen with these models never to get hooked on one one company will have a better model one month and the next month someone else releases a better model and then you just end up going back and forth and then what you end up finding is different models are better at different things and your team just adapts. But I think the key here is don't get, don't have your marketing reliant on one model or one company and their stack. Things change. You need to adapt with it, especially in AI. It's like every week shit changes.

4:19Eric Siu:So Neil, let me, let me show you something real quick. I don't know if I can, I'm going to share my screen with you real quick. So can you see this?

4:34Neil Patel:yeah I can see it okay so

4:37Eric Siu:not bad right like basically by the way this is just me talking to chat hey by the way shouldn't all the clawed icons be Fable 5 icons instead and it's just going back and forth these are the thumbnails so and I'm sharing my screen right now so those of you watching can see it but Neil do you see like these thumbnails they are I wouldn't say they're the best in the world but they're not bad right

4:57Neil Patel:yeah you just whipped them out they're not terrible

5:00Eric Siu:yeah I whipped them out so 6-7 % click the rate is not bad and by the way when I asked it to switch the Claude logo over here it switched it to the Fable 5 one like this looks pretty legit doesn't it yeah okay so let me show you more when I go back to my profile I actually can pull up the examples here on cell 5 versus Fable 5 over here Okay, so 5.6, what's better for marketing? So basically this website that I'm sharing right now, this was the world that it made, okay? One took about 30 minutes. The other one took about a couple hours or so. So Sol 5.6 took about 30 minutes to make this website.

5:37Eric Siu:And this is what it looks like. So Neil, keep in mind, when you're zooming on the site, it goes deeper and deeper and deeper for single brain over here that you can see. And for thumbnails, you can see the difference with the thumbnails over here. So these look pretty good, I would say. And then, oh, by the way, Neil, It did a really good job. Okay, this is the fourth thing. So when it comes to clipping for marketing school, I said, hey, I want you to put a hook at the beginning of each of these, and then here's the entire export. So I could literally take this export that we're doing right now, and it'll basically find like 10 different moments, and it'll put the hook in the beginning, and then, because you need a hook in the user, right?

6:12Eric Siu:And then it'll put the meat of the content, and it'll have a transcription. I used to have to go back and forth with my Hermes all the time to get it right. I didn't have to do that. So my point is, it gets to the point a lot faster. and that's what I would say, right? So it works a lot faster. It costs less tokens overall. I even have it running a recruiting loop as well, and this Amazon analysis as well. So all that to say is, I'm not going to say, again, your mileage may vary. I think both of them are very strong frontier models, but I've just found it to be, with the tests that I've ran, to be more helpful in general.

6:49Neil Patel:Dude, yeah, I'm with you. And on a random note, I just got this message on my WhatsApp. So one of my team members says, check out our user acquisition on Ants of the Public. 62 % comes from direct. Does this make sense? Then he says, the same is with Ubersuggest. Does this make any sense? 62 % of direct traffic is super high, you know? And I said, it does, not super high. When you build a strong brand, most of your revenue ends up being direct. When I look at all big companies, and I'm not saying my companies are big, But when I look at all big companies that we work with, the number one source of revenue is direct.

7:24Neil Patel:It's because they built massive brands. It beats out anything else. And the key is good product, good brand, be in business for a long time. It's, you know, doing all the stuff you did for years, SEO, paid advertising, social media, it all helps build a brand and all feeds into each other. but if you're measuring your marketing and if your direct sales and direct traffic isn't increasing year over a year um you're doing something wrong yep totally uh and by the way

7:58Eric Siu:i mean how long would you say you've been building your neil patel brand for

8:06Neil Patel:uh intentionally or not just in general intentionally intentionally intentionally i would say 20-ish years

8:17Eric Siu:wow dude 20 that's two decades guys i just want to call that out and i think for me what i think it was it's probably been maybe 13 years or so so this stuff takes time and we've said this before on this podcast if you want to start to build a brand you might not see anything for the first three years or so um and that's usually the case for whatever reason usually is around three years. Sometimes he might hit lightning in a bottle, but even if you look at Mr. Beast, he struggled for such a long time to get it going. And I think it took over three years to really get it going for him. And I think it's just, to even get 10 ,000 subscribers, I think that took him quite a while, longer than he thought he would take.

8:59Eric Siu:And if you look at his videos in the beginning, they're really bad, right? And so I think you have to embrace the suck, you have to embrace the cringe. I would say, again, if you're starting from scratch right now, go out there, create in public right now, build in public and then other people will start to follow you. You'll find like-minded people. You might even find partners to work with as well. I think there's one last topic we can hinge on here, lean on here before we hop out. So let me share my screen again. Do you see my screen?

9:27Eric Siu:Yes. Okay. So there's a lot of talk about building an AI native services company or AI native whatever company, right? And so this is a post that our team put out in terms of the structure of how this team might look. And so I'm going to keep it very high level here. I'm really talking about any AI native company in general. And I think the way, whether it's your business or my business, Neil, I think both of our businesses need to evolve. And so what I mean by that is your team is maybe, they're this human judgment layer over here. So they have these strategy calls, taste calls, client trust calls, all these calls over here, right?

10:08Eric Siu:Then you have these operating rules, these processes that I'm showing on the screen over here. And so I think the way you structure a company now from an org chart standpoint or even how you structure it from a process standpoint is going to be very different. Because you're going to have a lot of different skills that your team might take on and the agents might take on. You're going to have different knowledge and memory layers that the agents might take on as well. You're going to have an agent fleet, some that might do paid media work or SEO work. and then you might have all these loops. So there's all these things that you have to consider.

10:37Eric Siu:And Anil, I won't pretend to say I have all the answers, nor will I say you have all the answers. But I think the key thing now is that we start to talk about this because we're all kind of just figuring this out as we go. But I don't think people can kind of just sit still and think that they're going to get by with the same original structure. So how are you guys thinking about how you guys will evolve this over time? Or are you even thinking about this at all right now?

11:01Neil Patel:I'm not thinking about it, but my team thinks about it. So it's just something that I don't spend my time on. We just divide and conquer and we focus on what's the most important for us as a company and individually to get done. And I don't worry about some of the stuff that my team handles because I trust the people that are working on that stuff.

11:24Eric Siu:So do they even bring this stuff to you at all? No. Oh, what a life, Neil. What a life.

11:35Neil Patel:so they tend not to bring me much stuff. I tend to bring them stuff.

11:40Eric Siu:All right, so I wanted to take a moment to tell you about my podcast co-host, Neil's agency called NP Digital, and they work with a whole host of global companies or a global organization. Also, Neil has SEO tools such as Ubersuggest and Answer to Public. All you have to do is go to npdigital.com to learn more, and we'll see you on the other side.

12:00Neil Patel:By the way, we should talk about that. my time on just growing revenue. I don't actually spend most of my time on operations. I've genuinely focused majority of the time on revenue.

12:11Eric Siu:Well, Neil, let's talk about that because I think that's great. So the reason this is important, I think it's worth double clicking on is because if you, let's say Neil has an amazing executive team, right? Which he has said many times that his team is really good. And he's been very complimentary of them, right? Now, here's the thing. If you have to continue to tell people what to do versus them, like them kind of just figuring it out, that probably means you haven't hired right, right? And so I would say if there's one thing Neil's gotten right with MPDigip the last seven, eight years or so is he's hired correctly.

12:43Now, when you have to push people and pushing is probably you're always going to have to push them,

12:48Eric Siu:but when you have to tell them what to do strategically, that means that you have to dedicate your brainpower to that. And that means, unfortunately, whether they're good people or not, and I tend to think that most people are nice people, it's not a good thing for your business because then your mind becomes split in all these different directions. The good thing is Neil's just focused on, let's say, the M &A piece, right? Or he might just be focused on driving more revenue right now, doing more deals, right? He focuses on what he's good at and he hires people for the things that he's not so good at or that he's not interested in doing.

13:19Eric Siu:But I would say that my observation of Neil over the years is that he continues to keep the energy high, he keeps the temperature high, He continues to push people. But if Neil, you hired me for sales and you have to keep telling me what to do, what ends up happening to me in a couple weeks or a month or two?

13:36Neil Patel:All I was asking is, when you're saying, hey, what would you do if I had to keep following up with you? Are you saying this is your organization or you're working for my organization?

13:49Eric Siu:I'm working for you.

13:52Neil Patel:There would be a problem.

13:55Eric Siu:Yeah, and very quickly, right?

13:58Neil Patel:Yeah, very quickly. But my team and my culture is very sales driven. So my co-founder, Mike, messages me at 5.41 a.m. All right, I'm on PST. He's on PST as well. He sends me a screenshot. Good news for Friday, X company signed. And then someone else on the team, amazing. They're mentioning the team member's name, absolutely brilliant. this is a country where we made changes to recently because we felt it wasn't working out, especially on the sales end, right? When I said I made changes, you can assume it is people changes. So, and I didn't, my management team did because they also are very sales driven and it wasn't working out.