Episode 638 | How to Generate Startup Ideas (Plus 8 Ideas You Can Steal)

Rob Walling:
Welcome back to another episode of Startups For the Rest of Us. I’m Rob Walling, and today I welcome Justin Vincent on the show. We talk a bit about how to generate startup ideas, plus look at eight ideas he brings to the show that you can steal. Justin and I had a great time talking through these ideas. He had sourced them from a few mastermind friends of his, as well as his own brand. He has a tendency for coming up with ideas. I really enjoyed our conversation. If you don’t know of Justin, he is the co-host of the TechZing Podcast, and he also, last time he was on this show, was running nugget.one, which is a community for early stage founders, as well as bunch of business ideas, 4000 he tells me. I thought it was 800, but 4000 different business ideas that he had gathered over a few years. He is no longer focused on that. He’s moved on to a new idea, which we will talk about in the show, so let’s dive right into our conversation about generating startup ideas.
I want to ask you about coming up with startup ideas because A, it’s something that you have done a lot. I mean, I heard you do it on TechZing, your podcast, many times coming up with ideas and even just bringing new ideas, talking through them with your co-hots. But I also watched you start Nugget, nugget.one, which we’ve talked about here on the show before, where you didn’t generate all these ideas yourself, but you did bring them in, evaluate them, and kind of churn them through, and there’s 780 of them or something like that.

Justin Vincent:
4000.

Rob Walling:
Are there? Oh, my gosh. I was looking at the premium ones. So 4000, holy moly.

Justin Vincent:
Yeah. That’s right.

Rob Walling:
And you have generously allowed me to talk about some of them in the YouTube series that I’m doing. About every six weeks, eight weeks, I do a, here are some SaaS ideas to get your mind going basically. And so the YouTube channel, I’ll start showing some of those. But really today I want to talk about A, your process for coming up with ideas, and then you had some ideas you wanted to bring to the show, that you wanted to talk to listeners about.

Justin Vincent:
Yeah. So since we set this up, I’ve just been thinking through stuff. I’ve asked a couple of people for some ideas. And let’s see, I’ve got nine ideas for you. I’ve got two ideas that I think would be good for solo boot strappers because there’s a specific kind of context. I think your context really matters about the idea that you pick. And that’s actually kind of a big part of why I started Nugget in the first place, because I realized I was really bad at picking ideas, so I wanted to just do that thing where I got a chance to test it out many, many, many times. And ultimately, you know what I’ve learned over 25 years? I’ve learned that I need to do a to-do app because that’s where I’m at. That makes the most sense for me. And it’s finally, I’m going to actually build something very good that’s very useful. There’s so many different lessons that you learn as you execute ideas and realize things.
For example, building on Twitter or building on other platforms, it’s like, “Okay, these guys can shift the ground from under my feet.” That’s a really high risk. So okay, I’m going to pick ideas where I don’t build on other people’s platforms, or other things that I learned like with Light, the mobile delivery app. No, don’t pick an idea where you have to manage loads of people because that’s really painful. Just do something that … So ultimately, it just boils down to, okay, for me, the best thing for me to be working on is something that I can just totally do by myself. I can be the sort of one man wrecking crew, and I can bring really high value to people, and I can work with customers and develop features that they like. That’s my sweet spot. Might be different for other people.
So anyway, that was a little side tangent there. So I’ve got two solo boot strapper ideas. I’ve got four ambitious boot strapper ideas for you. I’ve got two VC ideas for you. And I’ve got one absolutely out there idea.

Rob Walling:
All right. Shall we start?

Justin Vincent:
Yeah, please do. Yeah, let’s go for it. Okay. Let’s go through it in that order, so we’ll do first of all … And as I go through them, let’s do some banter as well. So if you just give me your instant impression, then we’ll have a quick little talk about what a go to market might be, what the downsides and upsides might be. Okay, first one, this was something that I actually had a look at and had a go at doing a few years ago, maybe four or five years ago, transcription for team meetings. I did pretty well, I file dated it, I got it out there on a few different sites and I did really well at that first stage of getting consumer interest. But when I actually went to really build it, it didn’t match the expectations. And the main reason is because transcription for meetings is a really good thing to have. You want to be able to take notes. You want to turn them into text as you’re having a call like this or just a regular Skype call with your coworkers.
The issue is that the AI was just not good enough to transcribe audio that had background noise, or had crappy phone connections, or whatever. But now there’s a new game in town, OpenAI Whisper. And I’ll send you a little link to this, but basically it’s completely open and available to use. It’s un-restrictive software. And the key point here is it’s finally good enough to transcribe really bad phone audio quality into good meetings. And so people might say, “Okay, there’s going to be a lot of competition in this space,” and this is something else we should probably touch on as well. Competition doesn’t matter in my opinion when you’re talking about the beginning of seeding something. I mean, look at you and Drip. I mean, you went into a space that had an insane amount of competition, and you just executed and executed, and all of a sudden, you built a mountain just as big as the other people’s mountain. That’s the point.
So actually, competition is great because it means there’s validated market here. So what you can do is you can use a system like this and you could build a custom SaaS app for teams to share and search their notes. They can maybe comment on transcriptions. They can share transcriptions, export them. It could be a plug in at the computer level that plugged into Zoom, or Slack, or many different things. Or you could plug in at the operating system level and just start tracking the audio, the audio stream, and start transcribing from there. That’s the first one.

Rob Walling:
All right. So my first question to you is: Who needs transcriptions of their meetings? And the reason I ask is I have this thing called the Five PM Framework, where it’s five Ps and one M. It’s the problem, the purchaser, the pricing model, the market, product market fit, or product founder fit, and the pain to validate it. And it’s a mental framework of thinking through. It’s idea evaluation, just a checklist of a bunch of questions and stuff. I’ve talked about it eight episodes ago or whatever. And they’re in that order, it’s problems, number one, purchaser, number two, blah, blah, blah.
I don’t have this problem, and so I’m wondering who does. I’m not saying no one does. I know someone does. But why is it that I run two teams of however big they are, six and five? And we have team meetings and we just never record them. We either take notes, or we remember, or whatever people put on their to-do list. So who does have this need?

Justin Vincent:
That’s a great question. And I think it’s about notes and getting it into … When you just take notes, it’s kind of a bit limited because you’re in the middle of … You’re thinking about two things at the same time. You’re thinking about taking the note, capturing what they just said, and you’re thinking about what they’re saying right now as you’re writing the note, so you’re a little bit bifurcated mentally. So I would imagine that for the person taking the notes, it would be super helpful to just have this stream that they could just really quickly type in the keywords, just quickly go through it, and just go, “Okay, get that out into the bullets,” and then also be more present in the meeting. I think that would be my answer.

Rob Walling:
So your ideal customer, or in this case, the purchaser would be someone who works at a company where they tend to take a lot of notes in their meetings. And again, that’s not me. That’s not our company. I don’t know why. We’ll have meetings and it’s like, “Okay, so now we know there’s an action item.” We assign it to Rob. It goes in my Trello board right then. And then we sent it to Tracy, it goes into whatever her to-do list app is. So yeah, there’s really relatively few. I guess, you know what the difference is, those are update meetings and planning meetings. Brainstorming meetings, we do take notes. That’s the difference. So it’s when we’re trying to come up with this whole new concept for a new type of MicroConf event. We run through a bunch of stuff. We put it in Notion, actually, and we just have bullets and we’re moving them around and blah, blah, blah. So maybe is that the case where this works better?

Justin Vincent:
I mean, I think so. I mean, I’ve had hundreds of those kind of brainstorming kind of meetings. And you write stuff on white boards, and you say a lot of stuff, and a lot of it gets lots. And then a couple of days later you’re like, “What was that thing I said?” And you kind of look for it, but the white boards erased. The notes are gone. It’s like having an undo button.

Rob Walling:
Yeah, where you can go back to it. Yeah. So I can imagine producer Zander, producer Ron and I are having a brainstorming meeting, and we record it, and I can send it to rev.com for a buck a minute or 45, whatever, 75 cents a minute. Or I could send it through … Castos has AI generated transcription built into it for nothing, 10 cents a minute or whatever. And there’s just some services. So what is the idea that you have, how is it specifically good for meetings? What is built on top of it that is more than just me taking an audio and getting a transcript?

Justin Vincent:
Let’s just go back to the previous question you asked. I could do this other thing. Why should you use your product? That as a basic reason to stop to someone doing a product doesn’t make any sense to me because new products come out all the time. Why should someone use Drip when they can use Constant Contact? You know what I mean?

Rob Walling:
Drip was a better product. I mean, it was substantially better. It was noticeably better.

Justin Vincent:
So it would obviously be painful to send in … It’s annoying to have to like … Okay, you need to extract that audio into a file in some way. You need to send it. If it’s a purpose built tool, then it’s going to basically automatically do all that stuff for you. You don’t have to think about it. It’s operating like the NSA. It’s like-

Rob Walling:
Actually, have you ever jumped in a Zoom meeting, and then the person shows up, and then their AI bot jumps in? And I think that’s what it … I don’t even know what it’s doing. I think it’s recording and transcribing. Yeah.

Justin Vincent:
I don’t know. I don’t know.

Rob Walling:
There are sales tools that do this, where they will transcribe and then analyze how often the person says, “Um,” or how often, which keywords when used a lot, tend to lead to sales or whatever. So yeah, this is interesting.

Justin Vincent:
But it’s also about the character of the app. It’s about who you are as an entrepreneur creating the app and how you connect with your audience. That’s a huge part of it as well, so you can build a really interesting and strong audience just from the character that you are, how you promote it, the reasons why you do it, so all that plays into it as well. This is where I was talking about why I love the fact that you’re kind of passion driven about the market comp stuff is it’s not just about executing a business. It’s about you, you the founder are in the middle of that business, and that’s a huge part of why it’s going to succeed or fail, product founder fit, just like you said.

Rob Walling:
And specifically because you said these two ideas, the first two, are for solo boot strappers. These are more lifestyle boot strappers. This is not a multi million dollar ideas as you’ve presented it. It could feasibly be, but probably not. It’s more like a: Can I get this to 10K a month? Can I get it to 20K a month? Right? And for that, this could be it. And so for someone listening to this, if I were thinking about a next step is I would try to talk to as many people as possible to find out who has this problem, what they’ve done to solve that problem, much like out of The Mom Test. You have this problem. What else have you done? And if they haven’t done anything, maybe it’s not worth it. How much would they pay for it? These are the conversations I’d be looking at to have next.

Justin Vincent:
That’s true. Yeah, okay, so let’s move on to the next one. We’ll count that one as transcription for team meetings, idea number one.

Rob Walling:
Right.

Justin Vincent:
So number two, this comes from a TechZing listener on the TechZing Discord, Will Whit, online time capsule. And this again is for solo boot strappers, so all the caveats that we just said apply to this. So basically, online time capsule, upload files that get encrypted and give them a release date that would automatically trigger de-encryption at a point in time, days, months, years, or decades into the future. You could release it to a specific email address or you could release it to a public release stream on the main site. So on a public page, it shows a list of upcoming releases and it allows people to specify if the title is visible for the release or not. So the key point here is this could be useful for journalists. It could be useful for promotion. It could be useful for PR, or it could be useful for your family for stuff [inaudible 00:12:27].

Rob Walling:
Fascinating. And so it could also send a Tweet, feasibly. That could be the release of it. Right?

Justin Vincent:
Yes, it could send a tweet, yeah. Absolutely.

Rob Walling:
Or a Facebook post, or post to Reddit, or post to Hacker News. Sorry. I’m just [inaudible 00:12:40].

Justin Vincent:
So basically, it’s fully encrypted data release on a time. But it’s different to a dead man’s switch. It’s different to a dead man’s switch.

Rob Walling:
Yes, because this is based on time rather than a particular event. It’s not like when I do … Because I actually thought I have some information I want to relay to my family when I kick the bucket, but I don’t particularly want to give it to them now. And I’m wondering how to do that in a way that is easy. But this isn’t that, that’s the event, in the event that I die. This is more like a date based thing. Right?

Justin Vincent:
To do what you want to do, there’s a site called deadmansswitch.com, which would do what you want to do. But then this becomes super interesting idea. Who is the person behind this? Do I trust them? Is this really going to be released in 50 years?

Rob Walling:
Will it be around? I mean, that’s the other thing. As a solopreneur, how many of these apps do we see built and then gone within six months because they didn’t generate enough revenue? And so yeah, if I book it for five years. So what problem does this solve? The one thing you named that is someone who … Because I also can’t see myself using this, but maybe I’m not thinking creatively enough. So I’m wondering who would. Would a journalist? Let’s say I’m under embargo, which means I get a story, I can’t release the story until 8:00 AM on Thursday. Normally, you go into your CMS, you schedule it to go live 8:00 AM on Thursday on your WordPress instance or whatever. How do you think about this working for journalists?

Justin Vincent:
I mean, I didn’t until just this second, but-

Rob Walling:
Until you said it. Yeah. Or anyone else. Who has the problem?

Justin Vincent:
I think the point is it’s about … I mean, you could really use this for nefarious means. Couldn’t you? You could release different information bombs. You could use it in a bad way just as much as you could use it in a good way.

Rob Walling:
Absolutely. Especially can it be anonymous? We haven’t talked about that.

Justin Vincent:
Yeah. It could be like with WikiLeaks kind of usage.

Rob Walling:
When you first started saying that, I was like, “Is this a WikiLeaks link?”

Justin Vincent:
Because the other point is, other point with the ideas that I’ve come up with here, we could’ve come up with really sensible ideas. I could’ve just said, “Okay, build a CRM,” because essentially, if you want to make money and you want to guarantee to make money, go into a huge market, compete with other people, and just build something that people are already using and this definitely makes money, like a CRM. Just take your own niche version of it. But the problem is that’s kind of boring to talk about on a show.

Rob Walling:
It’s boring and it’s also very, the table stakes now are very, very high. To build Drip today, years of development as a solo developer that you cannot get anyone to sign up, so those are amazing markets if you have funding, or if you have multiple people working on it. But yes, these ideas are certainly less boring than just talking about that. So yeah, let’s look at our next. You have three or four for the more ambitious boot strappers.

Justin Vincent:
I’ve got four ambitious boot straps, four ambitious boot straps. This next one is, a lot of you talk about my process. So for me, a lot of ideas that I come up with are just as I’m walking around and it just seems like, “Huh, that should be the solution for that problem.” So this one is an ambitious boot strapper project, and it is pest control using drones. So where I live in this big complex, there are seagulls pooping and congregating on our roofs.
And the complex owners have spent tens of thousands on a laser system to get them away, and those lasers are really not very good, whereas what you could do is you could simply just put a drone that parked on the roof. It has a little parking spot, and when it sees a bird, using AI, it just sort of flies up and sort of acts aggressively and it chases it off. And the innovations for that are, it would need AI to check, to sort of decide when birds were there. It would need to be kind of quiet, and it maybe could emit some kind of noise to annoy birds or something like that. But that I think is an interesting, ambitious bootstrap.
Oh, the other thing is you, just generally speaking, you could use drones for pest control in many ways. So you could use it in a field to get rid of crows that were eating crops. You could use it to get rid of, I don’t know, foxes, or rabbits, or whatever, many different applications.

Rob Walling:
Interesting, this feels less like a boot strapper idea and more like something I would raise funding for. How would you bootstrap this?

Justin Vincent:
Well, because you’ve got to many builder kit drones these days, it’s not super expensive to do that. You’ve got so much access to AI stuff, data modeling and stuff like that. So I think it’ll probably take more than one person. I think it could take a couple of people. Look, by default, if something is an ambitious bootstrap, it can be a VC. It can be VC funded. But the problem with VC funding is, as you know, you gradually lose agency and control of your business, and use lose more and more percentage of it, and you end up working for someone else. So what would be nice if you were going to do something like this would be to try not to get into that position.

Rob Walling:
So obviously, the problem is obvious. What you’re saying is pest control, I used to life in Temple City, and there were these parrots. You know the parrots. I’m sure you’ve heard them. Yeah, they’re very loud.

Justin Vincent:
Yeah. We’ve got them in Pasadena. Yeah.

Rob Walling:
Yeah. The rumor is they escaped from somewhere at some pet arboretum at some point, and they bred. And they’re these wild parrots. They’re so obnoxious. So I think the problem is pretty obvious, and you’re saying this is not a hardware play. You don’t make your own drones. It’s a software play. Is that right?

Justin Vincent:
I’m saying it’s a whole, it’s a holistic play. Obviously, as something scales, you get better. You bring more stuff in house. I mean, that’s what Elon Musk shows us. If he’s proven anything, over time to become more and more successful, bring more stuff in house. But yeah, in the first place, you buy it, you don’t build, so you sort of cobble the pieces together, and then you just work out. Because when you get deep into a problem, like the list problem is I’m doing this text editor thing, the deeper you get into it, the more revelations you have. Holy crap, if I do this kind of thing, I could make it amazing because of these little granular tweaks. Well, that’s what any startup’s like. And I think this would be like that. The components would be the hardware. You’d need to get the good hardware to do it. You’d need to have a good AI to do it. And there’s also, there’s a charging station and just it needs to be quite autonomous. That’s another kind of point about it.

Rob Walling:
That’s a thing. I agree. I see the money in this one not being in selling to individual apartment complexes, but in going to facilities, like let’s say a campus, like Facebook campus in … Is it San Jose, or Campbell, or wherever they are? And they have all these buildings, and you know that they have bird problems and pest problems and whatever, and selling to … Start with them because that’s a big sale. But you kind of get your proof of concept going and then you figure out people. I’m even thinking whenever I’m around Europe, Italy, or France, or whatever, there’s just all these pigeons everywhere. And they have all these spikes that look ridiculous and don’t actually … The pigeons still land everywhere.

Justin Vincent:
Exactly.

Rob Walling:
So there’s in those public squares and all that.

Justin Vincent:
It could definitely be a VC. I mean, it could definitely be a VC idea grow at a very fast pace and it could be a massive business. And it’s not just the birds thing. It’s also farming. For organic farming, think about that. One of the hardest things about organic farming is not using pesticides or whatever, I don’t know, whatever chemicals they use to keep things away.

Rob Walling:
Yeah. This one’s interesting. I do think if I were to do it, I would raise money right from the start. I would not try to bootstrap it. Not saying you couldn’t, I just think when you bootstrap you go slower because money in your personal life saves you hours, money in your business saves you years. And so this is one that is probably inevitably going to happen I think. This is just going to be a thing, so I need to get there faster. How fast can I get there by raising money right?

Justin Vincent:
Just to what you just said, this is inevitably going to be a thing. I mean, that is actually, that’s true with pretty much any idea. If it’s worth any salt, it’s going to be a thing, so just another reason why I get kind of irritated when people say, “Oh, you can’t do it because of this and because of that.” Okay, next one, prerecorded live interviews. Here’s an example of it. I think it could be used in multiple ways. This one comes from, by the way, comes from Jeremy Nunan, who you matched me with in my mastermind group, who I really appreciate. I’ve got two great people in my mastermind group, Chas and Jeremy.

Rob Walling:
Awesome. MicroConfmasterminds.com.

Justin Vincent:
Yeah. So screening job applicants with prerecorded questions, but getting live answers. So the key point here is applicants go to a page, you can talk to them by text, or you can talk to them by video in a prerecorded way. And then ask them questions, and then they respond to the interview, but they don’t really have time to go away and Google it or whatever, so you can get their sort of moment of thinking answers, but you can do it at scale. So think about you could really scale the applicant inbox review process. You can get way more than just a resume. For example, let’s say you wanted to create a really deep pre qualifying applicant funnel for developers. The first one could be simple resume level, just ask some basic questions. Then the next one gets deeper on technical stuff. And so you can get those higher quality candidates in an automated, scaled way. And there’s a lot of different customers for this. Customers could be companies. Customers could be agencies, many different customers.

Rob Walling:
That’s interesting. So you’re saying me as the interviewer, I would pre record five different questions of, tell me about yourself. What are your biggest strengths? There would be better questions than that. But I record those individually, and then they would be in an interface like QuickTime, or Loom. I mean, it’s obviously a built interface, but they basically … Or almost like a Zoom thing, where it comes up.

Justin Vincent:
Yeah.

Rob Walling:
That’s interesting.

Justin Vincent:
And it’s live, and so the other thing is, by the time the … Because think about how much time is wasted just getting to the right people. So by the time your staff, who are your really good people who work for you, who are expensive, by the time they get to see anyone, they’ve got a show reel to look at. They’ve got a show reel. So that just helps everyone in every way of that hiring process.

Rob Walling:
Yeah. There’s something I like about this and something I don’t like about it. And the thing I really like about it is the efficiency, like you said, and I love asynch, oh, my gosh. I want my whole life. I want my whole life. I wish you and I could record this podcast asynch because just finding time on both of our calendars, I guess you and I are probably an exception because we actually have pretty open calendars. But a lot of times, I’ll go to try to record a podcast and it’s like, “Well, the next time we can do it with a busy person is a month from now.” And it’s like, “I’d really like asynch.” So I like those things.
Couple things I struggle with, with this, is I think as a candidate, I might feel like this is the dystopian future, 1984, cattle call, really couldn’t even take the time to chat with me. There’s always the charitable view of technology, and the uncharitable interpretation is like, “Well, if big tech just put me through a process, inhuman,” so I think that some people could feel that way.

Justin Vincent:
Let me just say, so that could be set up with expectations. That could be sort of kind of managed through expectations, and even the person asking the question could say that. And by the way, just so that everyone knows, on every idea I’m personally undecided, every point, until I see it’s successful. So I’m not pitching it as I definitely believe in every idea, I don’t know. I’m in the same position as you. I’m sort of like, “Yeah.” But I think it’s interesting.

Rob Walling:
Oh, totally. Well, that’s why it’s helpful for us to play the opposite side. I mean, sometimes on this podcast, I will literally play devil’s advocate, where I am saying things that I don’t agree with because if we both agree, it’s not that interesting.

Justin Vincent:
Right, right. Yeah.

Rob Walling:
So I’m coming up … I’ll come up with a pro, you can come up with a con. I don’t have an issue with that. I know that when we’ve tried to have developers record a 60 second video of just tell us. And we’ll get very few applicants because developers don’t want to. Other roles, like sales, customer success people who are more used to being on camera don’t have as much of an issue with it. So I do think there’ll be a hurdle, like a hurdle, and maybe this is better for certain roles.

Justin Vincent:
Maybe it’s just audio.

Rob Walling:
That might be even better too because I have another thing I heard, and then we’ll move off. I don’t want to spend too much time on this. But someone had a hiring process where they did require video, and someone brought up the idea of diversity, and of someone not wanting … That some folks shy away because it’s like, “Well, when they see me on camera, if I’m not a cis white male, then maybe I’m lower on … ” Right? So but audio could potentially get around some of that, although you’d still have gender stuff. Whatever, yeah. I don’t think we need to go too far into this, so interesting idea. What’s the next one?

Justin Vincent:
Okay, next one. So we’ve had to interesting, that’s all I actually want. I just want to see if you find them interesting. That’s the main point for me. Okay, so number three, this is actually for me, this is something that I’ve posted in Nugget. It’s a problem that I had thought about. I never really got into it because I think it will require a significant amount of work, but I do think that there’s a lot of potential here. A special diet builder, okay, so special diet builder. So step one, scrape every food recipe on the internet. Step two, create an interface to build any diet you want, gluten free, vegan, nut free, paleo. Step three, present recipes in a unified format with a link to the original. You can sell this to nutritionists, to the general public, medical professionals. Or you could create multiple landing pages and do an ad or an affiliate play. But a key point is, with a food framework, if you just limit to a specific food framework, it’s much easier to stick to an eating and diet plan.

Rob Walling:
I’m much more bearish on this one. See, I don’t like B to C stuff. You need buckets of money to churn as high, blah, blah, blah. Would copyright be an issue with this, or scraping all the stuff? And then even if you link to it. But you’re selling it to them, yeah.

Justin Vincent:
Well, you’re selling it to nutritionists, and you could sell it to nutritionists. You could sell it to … I mean, there’s a consumer aspect, which you could sell it to consumers. But probably another aspect would be to sell it to businesses, so it’s both.

Rob Walling:
All right. I’m just going to squint my eyes at you and say, “What’s the next one?” I don’t like this one. That’s my least favorite of all.

Justin Vincent:
Okay, yeah.

Rob Walling:
This one and online time capsule, they’re together at the bottom of the list for me.

Justin Vincent:
Okay, I’ll give you online time … Okay, so this one I just thought of. This is my most recent one, so I’ll just read through the spiel and then we’ll talk about it. Casting director talent finder, this is one of many possible ideas that there are for things like Midjourney and DALL·E2. But it’s just something I’ve been thinking about a lot recently because there’s so much of it out there. I’ve seen some blogs recently where they imagine what cartoon characters would look like in real life by working them up in DALL-E or Stable Diffusion. And the end result looked like real people, real and very well done. And I’ve also seen how easy it is to create people’s faces based on a description.
So what I was thinking is you could build a system for a casting director talent finder, like a SaaS solution. Essentially, it would have two steps. One, generate the face that you are ideally interested in for your project. So you go through your whole project and you develop all the faces. And then step two, basically, you use facial recognition to search through IMDB and all the different talent websites where actors are available, and then sort of pull them back. So some notes about this, the movie business is all about the look. That’s the primary driver of talent search. This would increase the number of close matches for casting directors significantly and save them an insane amount of time. It’s a nice B to B revenue model. It could be sold as two separate products or plans. Number one, the ability to build the face, and number two, the ability to find the actors. The key point here is that it could save hours of legwork and get better results.

Rob Walling:
That’s super interesting. This is what I love honestly about running TinySeed, is that we get these applicants and I’m able to look through just ideas I would never even know about or think about, and this idea is one I would not know about or think about, but I like the use of just the roundabout, it’s the look to AI, to reverse image search is really what it is, which I think is great. So here’s where I like this one because it’s fun and it would be the validating this is not that hard. Right? If I were to try to, next step from this, I need to talk to 20, 30 casting directors. I need to say, “If we built this, does this make sense?” And find out. Is this not within their process? Do they not start from the look? Do half casting directors don’t give a crap about the look, and they give a crap about something else? That’s what I want to know because I know so little about this industry and this process, that I want to know how do they actually do it across 20, 30 different casting directors.

Justin Vincent:
They must just have to look through thousands of photos.

Rob Walling:
Maybe. I don’t know. I don’t know how they do it, and so I’d want to find out, and not just from one because I guarantee there’s … Is it casting directors? Or is it … Yeah, no, I guess that’s what it is. Right? It is, that’s the right role.

Justin Vincent:
But it could be casting directors, but also, it’s the other people, like the main director and that kind of stuff. So it could sort of create this central hub around the casting of it. You could have a dashboard for the director, for the producers, and then I think there’s something there. This could be a VC kind of thing, or it could be a bootstrap kind of thing as well.

Rob Walling:
Yeah, it’s interesting. All right. What’s next?

Justin Vincent:
Okay, so we’re onto the VC ideas now. I’ve got two of them here, one of them from Chas, who’s also in my mastermind that you set up, and then another one that I just thought of. And I genuinely have no idea if this is any good, or this is a complete pile of crap, I just don’t know. But what I see is, I see VCs in the media kind of complaining about stuff like founders having parties, stuff like that. So I had an idea to create a cash burn alert for VCs. So you use Plate to plug into all the startups’ accounts, look at the spending heuristics. If a huge party spent on ridiculous stuff happens, alert the VC. VCs can offer potentially better terms on the deal if you’re prepared to just at least have this basic monitoring.

Rob Walling:
Oh, man. I love your stuff. This is so funny. So this is monitoring. It’s like, “I’m going to invest and I’ll give you an extra million dollar of your valuation if you allow us to … ” [inaudible 00:30:30].

Justin Vincent:
If I could just check that you’re not being a total child with the money that I give you.

Rob Walling:
Oh, you’re killing me. So yeah, that’s interesting. So you can get alerts, that’s the key thing.

Justin Vincent:
Yeah, it’s alerts.

Rob Walling:
Because obviously, VCs have the information right, so they can look at your books. But do they want to every quarter comb through it and look at every expense? No, they don’t. And so you’re saying you’d set up some type of way to look.

Justin Vincent:
You’d have good heuristics, but just say, “Eh, eh.” They hired a yacht with a hot tub.

Rob Walling:
Oh, my gosh. How would it know that though? It would just see a big expense. It could be the office rent, or it could be payroll, or it could be a one time expense to invest [inaudible 00:31:12]?

Justin Vincent:
That’s implementational details. That’s not my problem.

Rob Walling:
Oh, man. You hand waver. Dude. Okay. So I’m just going to shrug my shoulders and say, “Hey, it’s VC backable idea.” I could see it. I’ve seen weirder ideas backed by venture funding.

Justin Vincent:
This could help de-risk investment potentially, if anyone was prepared to [inaudible 00:31:33].

Rob Walling:
It’s pretty rare, honestly, the big launch parties and the yachts. It is so rare, so few companies do it, so that’s where I’m a little bit like … I really like it. All right. What’s the next one?

Justin Vincent:
This next one is, I’m afraid this isn’t funny. This is Chas’s. This is really serious. This is a proper business. It’s the least fun of all the ones we had, but it’s probably the most serious that you could actually turn into something big. Basically, ETL data pipeline modeling tool, a no code tool that can ingest and transform huge amounts of data and export it in a transform format to other systems with parallel processing, IE, it’s a pipe between two systems. It kind of sounds like Zapier, but that’s actually different because that just works one event, one line at a time. This is for bulk data movement, so think Yahoo pipes, but industrial scale.
So examples, let’s say you have to migrate one petabyte database from Postgres to a different structure, no SQL database, rather than writing scripts and spinning up lots of infrastructure, you just point this tool at it and it will auto scale that massive data slurp and save on developer time because that’s what normally happens with these kind of situations. You just have to create a whole bunch of stuff that you never use again. So ETL extraction transform the load, and it is really slow waiting for big data to move around the place, really, really slow. I think that a lot of companies, just to pick something out, Facebook or whatever, just a tool that could just really scale and make it fast to move massive amounts of data around could be very, very helpful.

Rob Walling:
This feels like an almost boot strappable business, at least to start with because you could built it on a smaller scale. I mean, first thing I would do as a founder is I would want to find out. What does the landscape look like? Not to say, “Oh, someone’s doing this, so I’m not going to build it.” That’s not what I’m saying. It’s: Who’s doing it? Who’s adjacent to this? How are they doing it? What’s working for them? What’s not? What’s my angle to get in? How am I going to be better, or different, or something? Right? And this is a piece of software that maybe you start with it only does MySQL to Postgres. What are the biggest, what are the most common, what’s the single most common database migration today from one to another? If I were to boot strap it, if I were to venture, you build 10 at once because you have 10 million in the bank. But if I were to take it slow, it’s like a lot of people moving from MySQL to Postgres, let’s see how we do that. I like your no-

Justin Vincent:
No code.

Rob Walling:
No code. Yeah, no code, sorry. And I see a need for this if it doesn’t already … If the problem is not already solved really well, this is a fascinating idea, and I could see it both being boot strappable and VC because you could … This is one that where you could make enough money to make it make sense, even at a small scale, versus some of the other ideas. A lot of VC ideas, they don’t work at small scale. That’s the point of them. Right? This is a dev toolkit, but also the whole service to do it as well.

Justin Vincent:
It’s the service that … But let me just, I’ll play a little devil’s advocate about this one. I was sort of thinking about it because I’ve done these kind of things a lot as a developer of 25 years building big systems. One issue that I was thinking is, okay, you make the middle part really fast, so you scale up, you do parallel processing, but you’re kind of limited by the input and the output points. If that database you’re slurping from is slow, or if the database you’re inputting to is slow, that’s sort of your limitation. So you could make it really fast in the middle, the transform stuff, but you might be a big limited on both ends by what’s already there. But maybe that’s fine, it’s just still this point of I think a lot of cases are, it’s slow in the middle. It’s the transform stuff. It’s transforming so many petabytes, and you have to build parallel system that transforms that and keeps it all in order and stuff.

Rob Walling:
All right. So you’ve brought eight ideas, thanks so much, two solopreneur, four ambitious boot strapper, and two VC. Although, I think you could kind of bootstrap one of them. But do we want to go high and low, top one or two favorites? Do you have favorites? Or are you just agnostic to all of them? Because see, I have opinions. Right? That’s what I bring to this, is through my lens, the ones I think that are most interesting, through my lens of boring because I like boring businesses. Right?

Justin Vincent:
That’s a really good point that you bring up, because of thinking through 4000 ideas, I am agnostic to all ideas. And I believe that any idea can be successful until it’s proven otherwise. And that’s kind of why I believe in the whole fail fast concept. Just go out there and test it and speak to people, and just fail fast and move through them as quickly as possible.

Rob Walling:
Yeah, awesome. Do you mind if I do my top two?

Justin Vincent:
Please do, yeah.

Rob Walling:
So my favorites of this, I mean, obviously, I really like the ETL data transfer tool because it’s technical. It’s just nerdy enough, it’s just boring enough, and I’m sure it solves a problem someone has, so I really dig that one. I also, that casting director talent finder is just fun and interesting. And I’m not saying I would go out and build it, and I personally wouldn’t even be the person to build it. But if someone is in that sphere at all, where you’re around filmmakers, casting directors, the movie industry, it’s just a fascinating use of a new technology, and it’s that get there early. Right? One of the advantages I talk about in startups is getting there early, and this could potentially be something where you could get there early. There’s obvious headwinds, obvious behavior changes, blah, blah, blah. We could talk through the negatives, but I dig that one quite a bit.
For me, the funniest one was the cash burn alert for VCs. I don’t know why that tickled me, it’s hilarious because it’s so out there, and I’m going to throw that. And then probably my least favorite, the ones that I don’t quite understand the … I think I just don’t understand the problem they solve is the recipes, it’s B to C, and the online time capsule one, where even transcripts for meetings, where I don’t have the problem myself, I’m sure someone has that problem. And the drone pests, I wouldn’t do the drone one because I don’t like hardware. But it’s a problem. And so how are we going to [inaudible 00:37:34] each of these applicant pre recorded interviews, it’s problem, so each of these I can see, except for again, I’m not saying there’s no problem, it’s just of these eight ideas, those are probably my two least favorite for me. Well done, sir.

Justin Vincent:
Oh, sure.

Rob Walling:
Justin, thanks so much for bringing all your amazing ideas to the show and discussing them with me. If people want to keep up with you on Twitter, you are at Justin Vincent. And of course, nugget.one if they want to see the project you used to be working on, and now see trylist.io for what you’re working on now. Thanks again for coming on.

Justin Vincent:
Thank you. Thank you so much for having me.

Rob Walling:
Thanks again to Justin Vincent for joining me. Thank you for listening this week and every week. It’s great to have you here. I appreciate all your support on social media, Reddit, Hacker News, all the mentions, they mean a lot to me. This is Rob Walling signing off from episode 638.