when some of the people ask me oh is my idea tarpit I'll be like hey well have you talked to any users and they're like no I just no I thought you would tell me though like it's funny it's like how you been watching the videos I think I think what we say in the videos is you should talk to users and get a customer they're like yeah I guess you did say that yeah [Music] interesting this is dalon plus Michael and today we're going to talk about tarpit ideas again tarp pit the sequel part two part two so we've gotten a lot of feedback from our first tpit video was it about a year it was a while ago yeah maybe a little bit more it's like oh this looks like a nice pool of water no one's no one's here drinking at it I'm going to I'm going to go get a drink of water from this pool right like no danger quicksand what have you learned about well I think a lot of people watch that

video yes and I've met a lot of people in the world real world that have watched it, and the most common question I get is, hey Dalton, is my idea a tarpit? and then they tell me their idea. And maybe the first thing that I noticed was a lot of folks, I feel like didn't exactly get the definition that we had, and they interpret the term tarpet to just mean an idea that I don't like. Again, maybe I'm and Michael, yeah, it's sort of like an idea that came up in the last video, and I think you know we try to define a different way. Why don't you try to give the definition, and you should watch the video, watch yes, but we should, we should restart, right? So, the idea behind the tarpet idea is we would see a lot of Founders applying with the same idea over and over and over again, and often we would fund many of them, and often the idea wouldn't work. And after a while we, we started seeing this trend.

we're like, Wow, there are these certain idea spaces that seem to always attract a lot of applicants and they never work. And like, that's how we kind of came up with this concept of a tarpit idea. And in the video, I explained that like that's how tarpits were for dinosaurs, right? It looks like a body of water from afar, and as the dinosaur gets close, they go to drink it, and it's tar, and they get stuck, and they die. And that's why we have all of these dead Founders in these tarpit pools, all these dead dinosaurs in tarpit pools. What was interesting though was that like you said, people interpret it as almost like a static list that doesn't change over time. Time well, and also that an idea that is bad and an idea is a tarpit are the same thing. A tarpit is a pretty small, if we imagine the VIN diagram, tarpit ideas are a smallish set of ideas. Yes, there's lots of ideas that are neither good nor bad, they're just normal. Yes, yes. And so yes.

there's something about tarpit ideas that are insanely attractive. Yeah, like if some other founders don't come up with them all the time constantly, it is not a tarpit idea. It has to be something that you're like, Yeah, I had that idea too. Yes, I think there's another part about tarpit ideas which like, they sound it's really funny. Most startup ideas, when you tell your friends and they being honest, they're like, That's not a good idea. Tarpit ideas get like a lot of praise. Like the second your friend's like, That's a great idea. Yeah, like a food discovery. Like, you know that problem where you can't decide what restaurant to go to? Yeah, that's a great idea. You pit your friends on like, I totally have that problem. I have that problem all the time. I think that the other thing you brought up is there's a characteristic of a startup idea where it's been done before, right? And I think this is an area where a lot

of Founders, you know, avoid research to not learn things they don't want to learn. But it's been done before a lot, and nothing has changed. Like there's no, you know, I would argue that there are a number of ideas that at any given time in the, you know, last hundred years might have been tar pit. And then technology changed, yep, and suddenly these are not only possible, they're great businesses. But I think that the kind of characteristic of it's been done a lot, a lot of people have failed, and nothing has changed that's very. And there's just like an endless supply of folks that haven't even started a company yet that tend to come up with a similar idea. It just is one of the first ideas to pop into people's minds. Now, I think what's interesting is what's happened since we recorded that video is LLMs, yep, AI. I think that when I'm reading applications, my first thought is like I am open to even classic tar pit ideas no longer being tar pit. Like I'm open, what I love

to see is when a Founder can explain to me why the new technology might make something accomplishable that wasn't accomplishable before. Yep, what I hate to see is a Founder that's not even acknowledging that like many people have tried and failed before. Like it's like, Oh, history that happened before I'm alive didn't happen, wait, it didn't like I've been alive and then that's when history starts, right? And actually, I wasn't even aware so it's like since I've been like 18ish like that's like history starts. The only good music is when I was 18 exactly, literally. I think another area that you brought up is kind of the X for Y, right? You like the co-pilot for, you know, co-pilot's working great for developers, yep. Should there not be a co-pilot for fill in the blank, yeah. And in this one to be very clear, I don't think that these are tarpits fair but they are common and so to be very

specific. Everyone says co-pilot for X and then like we could play Mad Libs, co-pilot for real estate agents, co-pilot for investment bankers, co-pilot for lawyers. Pilots co-pilot for Pilots, co-surgeons, surgeons, yeah, co-pilots for, you know, basically any job. Who doesn't need a co-pilot? Co-pilot for dentists, co-pilot for dog walkers. Who would not want a co-pilot? And so again, these are not tarpits because not enough people have tried them exactly, but they are common ideas, I would say. Again, doesn't mean you shouldn't try it, but yeah, a lot of people just from first principles trying to come up with AI startup ideas, these are the type of things they tend to cook up. Yeah, this is not meant to try to like screw with your dreams, right? Like I think that a lot of people start companies because there's a problem they see in the world they really want to fix. And I think that the depressing thing I've learned is that whatever current

set of technology is available can address only a current set of problems, and like some problems are going to fall outside of that set, and every time there's a new chunk of Technology, we have to re-explore, right let's re-explore what problems are available, and I think that what's cool is that like there are a lot of companies this batch where I would say I would not have funded this idea in the pre- lolm world, okay like I funded a company that's basically doing like completely um AI gener podcasts, and it's like without llms and the ability to do like voice generation, that would I would have never considered funing it with that, I'm like okay let's see, and so I think that's the other thing that you should kind of understand from our perspective, we're definitely in a okay let's see phase compared to like 10 years into B2B SAS, it was like we've seen a lot of, we've seen a lot of forms and

like yeah, so I think that like that's another fun part of this. And I guess AI-generated video and audio, we have seen YC companies working on those since 2016. 27 like the early generative AI stuff we funded. We've been funding that consistently, yes, and it seems like it's getting better. Yes, the quality is getting better. Yes, it's one funny thing about these things is uh, they always train them on YouTube video. And guess, guess what, guess who has a lot of YouTube video of them. And so, all these new demos that we see, it's us, yeah, in the videos. Quite embarrassing. It is a little bit embarrassing, but they're getting, they're getting admittedly kind of good, yeah. Um, well, I was thing, I played like, I played this personalized podcast. It's called pocket pot. I played this one to my wife, and she's like, oh, like you're listening to NPR or like some other thing. And then it played a quote of me talking to the reporter, and she's like, oh, did you give an interview?

I'm like this all is this is. I didn't I'm learning for the first time that I'm on my own podcast. And so it's it's. I would say it's fun for us to see all these new things. But like still if you can't draw any connection between the two technology and why your problem can be solved today. And it's one of these ideas that have been tried many times tricky yep. Maybe the last thing we should talk about is like where what are the flavors of tarpits. You know we were talking about this before. There multiple flavors of tarpits like what do you what do you see. I think there's the two most common vect are one just a belief that the world should work differently than it does better. And this is all coming from a good place but it's basically wow the way X works is bad we should do it why and everyone simultaneously needs to adopt the scheme. So I'd put the the idea of coordinating with your friends to see who you can.

hang out with and do event planning is one of these. Where wouldn't it be cooler if all your friends wanted to hang out all the time and you could coordinate with them all, have more fun. We'd all have more fun, we wouldn't be sad. Yeah, it, and so the way you want the world to work is that everyone's like, yeah, I want to install a new app to see what all my friends are doing and hang out with them. Sadly, it does not appear that people want to do that, or at least no one that's tried this idea has figured it out yet. I'm sure someone is watching, they me like, but I'm going to figure it out. Um, God bless you. But that's, again, it's coming from this really good place, this utopian sense of changing the way that we live our lives. Yes, and running into that, it's just hard to change human behavior. It's a perfect example of tarpet because it's the perfect example of something where like nobody would say no, yeah, no one want to discourage you, yeah, you're.

working on something that's really good. Yeah, so that's number one. What about number two? Number two is um, I don't know, you did a better version of this earlier. Yeah, I think it's tricky. I think it could be summarized as like some form of arbitrage, yeah, or getting rich very quickly. There's like a limited window where like I have some secret knowledge and like I can get very rich very fast. And like I would be working for Google, but because of this moment in time, I can kind of like fast forward through the entire startup thing and like kill it. And I think that um, that also sounds really attractive. It's like right, like you have to work for 15 years to make your startup work, but like if I just do the right thing at the exact right time, game on, grand slam, home run, I can be rich, you know, in Miami. Like yeah, this is where you just end up doing an idea that's like whatever is raising money or you find some sort of arbitrage where you know a lot of

adtech stuff was like this or uh anyway. Like you see these momentary things where you want to just immediately make tons of money from it. There were lots of people um influenced by Wall Street bets uh during the pandemic where a lot of startup ideas were like someone saw Wall Street bets and was like hey I'm gonna make a startup out of this somehow like that was the that was the input. My favorite one is like I'm doing so great on Wall Street bets and my friends are all seeing it and they just want to give me their money so I can invest it for them so what is the like fastest way I can create like an investment advisory business excuse me so I could just just shove my friend's money into the market as fast check out Michael. I don't know I don't know why you're live like I was going to pit you that idea. I I got some great stock tips and I'm just like God like am i w like am I watching the car crash in real time I I don't see.

the problem it's like don't markets always go up forever isn't that how isn't that how graphs work aren't graphs like this when you're young yes the market only goes up forever if it's your first time and I'm thinking about this poor Founders friends who just dump all their life savings into their investment advisory business and the market I even know what company you're referencing well there's many I mean there's there's more than I'm thinking of yes I know I know what you're saying so yeah so that's like you know the mercenary so wrapping it up right I think maybe the big takeaway here is this is not a static list of ideas maybe ideas that we had mentioned in our last video are no longer tarpit Technology's changed maybe ideas we made fun of in this video won't be tarpit 6 months from now technology changes if it look looks too good to be true like if you're getting too much positive feedback if it feels like why

does this already exist? It seems so easy. That's not like the shape of things that we see working right. Like maybe you should be a little skeptical. Well, it's funny when some of the people ask me, Oh, is my idea tarpit? I'll be like, Hey, well have you talked to any users? And they're like, No, I just no. I thought you would tell me though. Like it's funny. It's like how you been watching the videos. I think what we say in the videos is you should talk to users and get a customer. They're like, Yeah, I guess you did say that. Yeah, interesting. I thought I would just pick a video and idea based on what you said in the video. All right, Colton, good chat. All right, sounds good. Thanks, man. [Music]

Tarpit Ideas: The Sequel

Videos to help you build a successful startup. Subscribe for startup advice, founder stories, and a look inside Y Combinator.

What is Y Combinator?
We invest $500,000 in every startup and work intensively with the founders for three months. For the life of their company, founders have access to the most powerful community in the world, essential advice, later-stage funding and programs, recruiting resources, and exclusive deals. 

Visit ycombinator.com to learn more.

Video Team:
Zach Both - Video and Content Lead
Ryan Loughlin - Senior Producer

