Give Me 36 Minutes and I'll Teach You How to Find $1M SaaS Ideas — Transcript
Full transcript
- 0:00Choosing the right SaaS idea can be the
- 0:01difference between making 10,000,
- 0:03100,000, or even a million dollars a
- 0:06month and making nothing at all. But I'm
- 0:08not here to give you another list of
- 0:09hypothetical SaaS ideas, cuz that's not
- 0:12what you need. What you really need is
- 0:14to know how successful founders actually
- 0:16found their idea and how you can steal
- 0:18their tactics to find your own idea
- 0:21today. So, here's what I did. I surveyed
- 0:23over 200 SaaS founders making at least a
- 0:26thousand dollars a month, some making
- 0:28well over a hundred thousand dollars a
- 0:29month, to uncover the exact methods they
- 0:31used to find their winning ideas. This
- 0:34is something no one else on YouTube is
- 0:36doing. People are talking out of their
- 0:38opinion and limited experience, but on
- 0:40this channel, I like to get some data
- 0:42and provide you with some insights that
- 0:44you can't get anywhere else. So, in this
- 0:46video, I'm going to break down the top
- 0:48eight ways these founders discovered
- 0:50profitable SaaS opportunities, and I'll
- 0:52give you examples for each. I've seen
- 0:54these methods work in my own companies
- 0:56and across the hundreds of startups I'm
- 0:58invested in, as well as the thousands of
- 1:00SaaS companies that I've seen started in
- 1:03the broader TinySeed, MicroConf, and
- 1:05Startups For The Rest Of Us ecosystem.
- 1:07There's a lot to cover in this video, so
- 1:09let's dive in. Method number one was
- 1:11used by a shocking 48% of founders that
- 1:14we surveyed, and honestly, didn't
- 1:16surprise me at all. For years, I've
- 1:18said, "Don't tell me your idea, tell me
- 1:21what problem it solves." And for almost
- 1:23half of the successful founders we
- 1:25surveyed, that problem showed up right
- 1:27where they were already spending 40
- 1:29hours a week. Method number one is
- 1:31finding a problem at your day job. So,
- 1:34this is where you spot a problem while
- 1:35working your regular job, either as a
- 1:37developer frustrated by bad tools, or as
- 1:40a subject matter expert who knows your
- 1:42industry's pain points. You already have
- 1:44contacts, you have credibility, and
- 1:46access to potential customers. One
- 1:48example of this is Alex Yumishev, the
- 1:50founder of Jitbit, which is a help desk
- 1:52ticketing system. Alex told us, "While
- 1:54working my full-time job, I realized
- 1:56that many existing tech support and
- 1:58customer support apps were missing the
- 2:00boring features that real teams actually
- 2:02need. Things like SSO, Active Directory
- 2:05integration, user provisioning, on-prem
- 2:08self-hosted deployment for Windows, and
- 2:10many other features. The sleek modern
- 2:12tools had none of that, and the ones
- 2:14that did have the boring features look
- 2:16like they were straight out of the '90s.
- 2:18Cluttered UIs, endless checkboxes,
- 2:20painfully complicated setup flows. And
- 2:23while the market was very saturated, I
- 2:25took that as a good sign. The market was
- 2:28there." Attacking a big competitor can
- 2:30be a benefit on one hand, stealing cool
- 2:33marketing ideas from the smaller
- 2:34companies on the other. Maybe you've
- 2:36spent years working in an industry and
- 2:38you're a subject matter expert. You know
- 2:39all the headaches inside and out, like
- 2:41Zach from Upstream Edge. Zach told us,
- 2:45"I provide a desktop application for
- 2:47forecasting and valuing oil and gas
- 2:49wells that include sophisticated
- 2:51engineering and financial analysis
- 2:53tools. 100 times faster speed and higher
- 2:56level features compared to alternatives
- 2:58are what differentiate my product. I
- 3:00worked in the oil and gas industry as a
- 3:01petroleum engineer for a decade, and my
- 3:04frustration with the current tools led
- 3:06me to learn how to build new tools to
- 3:08solve our daily problems. I had a
- 3:10burning desire to see better software,
- 3:12and I decided to make the product I
- 3:14thought the industry needed." Then
- 3:15there's Lee Dido, the founder of
- 3:17Didomite. Lee told us, "It's the fastest
- 3:20way to get looping video on a public
- 3:22screen. I started working at a company
- 3:24called BrightSign in 2012, who was the
- 3:27market leader in digital signage media
- 3:29players. It's a very hardware-focused
- 3:31industry, and I repeatedly kept seeing
- 3:33the same problems over and over when it
- 3:35came to operational software. Just
- 3:37nobody was thinking about user
- 3:38experience, and I kept socking away my
- 3:40ideas until one day, 10 years later, I
- 3:43thought it was time to jump." Consider
- 3:44jobs you've had, even a long time ago.
- 3:47The problem I just mentioned was
- 3:48something Lee experienced a decade
- 3:50earlier. Think about those tedious,
- 3:52repetitive tasks that are currently
- 3:54being human automated and that could be
- 3:56turned into software. The next example
- 3:58is similar in that it's automating a
- 4:00task that this consultant did for their
- 4:02clients. Ahmed Babikir is the co-founder
- 4:05of Blue Gamma, which is interest rate
- 4:07forecasts for finance teams. Ahmed told
- 4:10us, "My co-founder and I used to work at
- 4:12a consultancy, where clients would ask
- 4:14us for interest rate data, sometimes
- 4:16multiple times a day. As a recent
- 4:18graduate, my job was to pull the data,
- 4:21format it, and send it over in a
- 4:23spreadsheet every time. After leaving
- 4:25the company and learning about startups
- 4:27a year later, we realized it was a
- 4:28repetitive problem that software could
- 4:30solve very easily." Or maybe at your day
- 4:32job, you see a colleague doing
- 4:34repetitive work that you know could be
- 4:36streamlined, like Steve from Loody,
- 4:38formerly Metro Retro. Loody is
- 4:40collaboration software focused on
- 4:42development teams, helping them with
- 4:43retrospectives, refinement, planning,
- 4:46and so on. Steve told us, "I created an
- 4:48early prototype of a canvas-based retro
- 4:51tool as part of a hackathon at my last
- 4:53real job. The idea came from watching
- 4:55the PM write up the notes after every
- 4:56meeting by hand." So, now let's take a
- 4:58step back and look at what makes this
- 5:00first method of finding a problem at
- 5:03your day job work. The first is, you
- 5:05already have deep context around the
- 5:07problem. You've lived it, you know the
- 5:09workarounds, you understand why existing
- 5:12solutions fall short. The second is you
- 5:14have built-in access to potential
- 5:16customers and beta testers. These are
- 5:17your colleagues, your industry contacts,
- 5:19and so on. The third is that you
- 5:21understand the buying process. You know
- 5:24who signs off on purchases, what budget
- 5:26cycles look like, what objections come
- 5:28up. And the fourth is that you can
- 5:30potentially validate the idea while
- 5:32still employed, reducing your risk. But
- 5:34there are three things to watch out for
- 5:36with this approach. The first is make
- 5:38sure the problem exists beyond your
- 5:40specific company. Your employer might
- 5:42just be uniquely dysfunctional, so
- 5:45validate your idea outside of your
- 5:47little bubble. The second is check your
- 5:49employment agreement for IP clauses.
- 5:51Some companies claim ownership of
- 5:52anything you build while you're
- 5:53employed. And the third is be careful
- 5:56not to use proprietary information or
- 5:58customer data from your employer to
- 6:00build your product. I also wouldn't use
- 6:02any of my employer's hardware, like a
- 6:05laptop. I would have a personal laptop
- 6:06that I worked on, and as I mentioned, I
- 6:08would check that IP clause and be really
- 6:10clean about it. So, with this method,
- 6:12I'd be asking myself three questions to
- 6:15try to get my juices flowing. The first
- 6:17is what tools do I use on a daily basis
- 6:20that frustrate me or my coworkers? The
- 6:22second is what manual process or
- 6:25processes exist today that I think
- 6:27should be automated by software? And the
- 6:29third is, are the incumbent tools in
- 6:32this industry actually old and clunky
- 6:34and not changing, or are they actually
- 6:36pretty good and I just don't like them?
- 6:39So, there were a lot of companies that
- 6:40used this first method. As I said, it
- 6:42was almost 50%, but I don't want to
- 6:44spend this whole video talking about it.
- 6:46So, I put together a completely free
- 6:48resource over at robwalling.com/ideas.
- 6:54I'll tell you more about that later, but
- 6:56one thing I want you to notice is that
- 6:57all of these folks were solving problems
- 7:00that people would be willing to pay for.
- 7:02In the next category, the founders knew
- 7:04that there would be a customer ready to
- 7:05pay because they copied an existing idea
- 7:08or an existing product. This was about
- 7:1010% of our respondents. In some cases,
- 7:13founders saw how much their employer was
- 7:15spending on a solution and thought, "I
- 7:17could do that cheaper," or "I could do
- 7:18that better." Or you might notice a
- 7:20large space with a hated competitor.
- 7:22While I didn't participate in this
- 7:24particular survey, I did exactly this
- 7:26when I was building Drip, and we did it
- 7:28by accident, if I'm being honest. We had
- 7:30built a little email capture widget and
- 7:32the ability to send sequences of emails
- 7:35to people that signed up. What we
- 7:36quickly realized is there was market
- 7:38pull. There were people talking about
- 7:40marketing automation and email
- 7:42automation features that didn't exist in
- 7:45the less expensive email service
- 7:47providers, and every tool that had these
- 7:49features was really expensive, had
- 7:51annual contracts, software wasn't great,
- 7:54and they required you to talk to a
- 7:56salesperson. So, we started building
- 7:58some of those features, and while we
- 8:00didn't copy any existing products, we
- 8:02did enter an existing category, and
- 8:04there were already paradigms in place
- 8:06that we certainly slotted into, and it
- 8:09made selling and marketing our tool a
- 8:11lot easier because when people arrived
- 8:14at the homepage drip.com, they could
- 8:16see, "Oh, this is this type of tool. Why
- 8:19is it different?" And that's all I had
- 8:20to communicate. I didn't have to
- 8:21communicate what we were doing and how
- 8:23novel we were. I just had to say, "We
- 8:25are either an email service provider,"
- 8:27or at some point we were an email
- 8:29automation software, and then we became
- 8:31lightweight marketing automation. And
- 8:33then I just had to tell you, "Why are we
- 8:35better or different than all the other
- 8:38incumbents?" And there were a lot of big
- 8:40clunky incumbents that people didn't
- 8:41like. And that, in addition to the fact
- 8:43that we executed really well, was the
- 8:45reason that Drip grew so quickly and
- 8:47resulted in a life-changing exit for me
- 8:49back in 2016. Another example of using
- 8:52this approach comes from Tim Leland, the
- 8:54founder of t.ly.
- 8:57t.ly is an all-in-one URL shortener and
- 9:00QR code platform. It's built for
- 9:01marketers, developers, creators, and
- 9:04teams. And Tim told us, "I was working
- 9:06for a company on an online giving
- 9:08product, and we were tasked to build a
- 9:10text to give feature. To do this, we had
- 9:12to shorten the URLs since SMS messages
- 9:15are limited to 160 characters. We looked
- 9:18at Bitly and saw it was thousands of
- 9:19dollars to use a custom domain to create
- 9:22short links. So, to save money, I built
- 9:24a simple in-house URL shortener over the
- 9:26weekend that we could host ourselves.
- 9:28This got me thinking other companies may
- 9:30want a more affordable option, but they
- 9:32don't want to build it in-house."
- 9:33Another example of a founder who used
- 9:35this approach is Lilia Tobin. She
- 9:37started BigMailer, which is an email
- 9:38marketing platform with a focus on
- 9:40multi-brand management for agencies and
- 9:42franchises. Lilia told us, "While
- 9:45working on our previous B2C startup, we
- 9:47had grown our email list to more than
- 9:49300,000 subscribers. All the available
- 9:51email solutions seemed too expensive for
- 9:53a product update newsletter of our size,
- 9:56so we built an in-house solution powered
- 9:57by Amazon and decided to make it into a
- 9:59standalone tool. Another option might be
- 10:01to port an existing product to a new
- 10:04country, language, or geography. We had
- 10:07a couple of respondents who copied an
- 10:09idea that seemed to be working in the US
- 10:11and they translated it to the needs of
- 10:13different geographies. One example is
- 10:16Checkout Joy, which provides custom
- 10:18checkout pages for e-learning businesses
- 10:20that connect local payment processors
- 10:21with LMS platforms. The founder, Merijn
- 10:24de Wet, told us, "My wife had a
- 10:26membership business and couldn't easily
- 10:28accept payments from their LMS platform.
- 10:30No Stripe where we live and very limited
- 10:32PayPal usage. Locals don't want to pay
- 10:34in USD, but that was all the LMS
- 10:36offered. I was researching solutions and
- 10:38realized that there are potentially
- 10:39thousands of businesses with the same
- 10:41issue. So I built a quick MVP with our
- 10:43local payment processor and got them
- 10:45live within a week. We could see an
- 10:46immediate bump in conversions for them
- 10:48and it made it easy to do business and
- 10:50quickly get funds into their local bank
- 10:52account." Another founder who took this
- 10:54approach is Simon Thompson, the founder
- 10:56of Hururu.ai,
- 10:58which is an AI answering service to help
- 11:00local Australian business owners manage
- 11:02phone calls, especially during weekends
- 11:04and after hours. It has local Australian
- 11:07voices and sensibilities. Simon told us,
- 11:09"I saw examples in the US, such as Rosie
- 11:12AI, doing the same thing and figured
- 11:13there would be a need for the same type
- 11:15of service, but specifically with
- 11:17Australian accents and sensibilities. I
- 11:19also guessed that it wouldn't make sense
- 11:21for the major players in the US to build
- 11:22Australian accents into their platform
- 11:24and go to market in Australia, at least
- 11:26not anytime in the near future." So what
- 11:28makes this method work? Well, four
- 11:30things. The first is you have proof of
- 11:32demand. Someone else already validated
- 11:35that people will pay for it. The second
- 11:37is you can learn from the incumbent's
- 11:38mistakes by reading their negative
- 11:40reviews and support forums. You can see
- 11:41what they're doing right and see what
- 11:43they're doing wrong because they
- 11:44pioneered it. The third is you can
- 11:46differentiate on price, user experience,
- 11:49specific features, or a particular
- 11:51niche. And if the incumbent has been
- 11:52around for a long time, they've often
- 11:54raised prices and gone up market and
- 11:56that can leave a lot of opportunity. The
- 11:58fourth is if the market is large enough,
- 12:00even a small slice can be a great
- 12:02business if you're a mostly bootstrapped
- 12:04founder. But there are also four things
- 12:05to watch out for when using this
- 12:07approach. The first is copying a product
- 12:09in a small market means you're fighting
- 12:11for scraps. This approach works best
- 12:13when the market is big. The second is
- 12:16make sure you have a real
- 12:17differentiator. It's not just we're
- 12:19newer, that's not enough. So when we use
- 12:21the word copy, what we really mean is
- 12:23enter an existing category where there
- 12:25are already other players. I still think
- 12:27you need to get in and position yourself
- 12:29in a corner of the market that isn't
- 12:31being served very well. The third thing
- 12:33to watch out for is the incumbent has a
- 12:35head start on brand recognition, on SEO,
- 12:38integrations, customer trust, marketing,
- 12:40all kinds of things. And the fourth is
- 12:42be honest about why you think you can
- 12:44win. I'll just execute better is not a
- 12:47strategy and just copying an existing
- 12:49player is very unlikely to work. So the
- 12:51questions to ask yourself are number
- 12:53one, why are customers unhappy with the
- 12:55current options? Is it price, features,
- 12:57support, usability, or something else?
- 13:00The second, is my differentiation
- 13:02something customers actually care about
- 13:04or is it just interesting to me? The
- 13:06third is can I reach customers through a
- 13:08channel the incumbent isn't using well
- 13:10or where there's some type of
- 13:11opportunity? And the fourth is is there
- 13:13a specific segment of the market that's
- 13:15underserved and would love a product
- 13:17built just for them? Method number three
- 13:19is discovering a problem through
- 13:21freelancing. About 10% of the founders
- 13:24we surveyed took this approach. Examples
- 13:26include Ben Walker, the founder of Drum,
- 13:29which is an ERP for small to medium
- 13:31consulting firms. Ben told us, "I was
- 13:33consulting as a custom software
- 13:35developer and found that I was building
- 13:36very similar functionality over and
- 13:39over. I felt that if I could build a
- 13:41very basic version of what I was
- 13:42building my clients, but build it as a
- 13:44standalone tool, I should be able to
- 13:46find some early customers and iterate."
- 13:48John Reynolds, the founder of Civic
- 13:50Review, which is permitting software for
- 13:52local government, told us, "I was a
- 13:54freelance software developer and I was
- 13:56asked to build our product, permitting
- 13:58software, by municipalities. I was asked
- 14:00three times and on the third time I
- 14:02decided, 'Okay, these guys can't find a
- 14:04SaaS solution that fits their needs.
- 14:06I'll just make my own and sell it so I
- 14:08don't have to keep rewriting the same
- 14:10software.'" There's also the chance that
- 14:12you discover a problem to solve that's
- 14:14not something you've been asked to build
- 14:16multiple times, but rather it's a tool
- 14:18that you need in order to be more
- 14:20efficient in running your agency. Colin
- 14:22Bartlett, the co-founder of StatusGator,
- 14:25which is a status aggregator, told us,
- 14:27"I was working on a bug ticket for a
- 14:28customer all day and then finally came
- 14:30across the status page for the Facebook
- 14:32Ads API and it had a warning about there
- 14:35being intermittent failures. Turns out
- 14:37that was the bug. It annoyed me that I
- 14:39did not even know the status page
- 14:41existed. Plus the app interfaced with
- 14:43like 10 other ads APIs. I thought it
- 14:46would be really nice to have all that on
- 14:47one screen and get an email when the
- 14:49status changed. So I built StatusGator
- 14:51for myself." Note that it's important,
- 14:53if you're going to do this, that you
- 14:55think very carefully about the
- 14:57intellectual property. In a lot of
- 14:59consulting and employment agreements,
- 15:01you will sign away anything you build at
- 15:04the job and anything you build in your
- 15:06free time, even on your own computers at
- 15:09home. So you you want to be very
- 15:10diligent about this and have all your
- 15:13eyes dotted and t's crossed. Another
- 15:15example is Dump Truck Dispatcher,
- 15:17founded by Joe Walling, no relation.
- 15:20Dump Truck Dispatcher streamlines the
- 15:21operations and logistics for bulk
- 15:23hauling companies. And Joe told us, "At
- 15:26the time I ran a custom software
- 15:27development company. I had a client who
- 15:29wanted this type of software. I kept the
- 15:30IP in return for charging them less for
- 15:33the software." So let's look at four
- 15:34reasons why this method works. The first
- 15:37is you see the same problems across
- 15:39multiple clients, which validates that
- 15:41demand exists beyond one company. The
- 15:43second is the clients are literally
- 15:45paying you to understand their problems.
- 15:47You're getting paid to do market
- 15:48research and customer development. The
- 15:50third is you build real expertise in the
- 15:52domain as you're getting paid. And the
- 15:54fourth is you can sometimes fund early
- 15:57product development through your agency
- 15:59or freelancing revenue. But four things
- 16:01to watch out for include agency work is
- 16:03demanding. Finding time to build a
- 16:05product on the side is genuinely hard.
- 16:07Number two is make sure IP ownership is
- 16:09clear in your clients' contracts before
- 16:11you productized anything, like in the
- 16:13case of Dump Truck Dispatcher. The third
- 16:16is that the transition from freelancing,
- 16:18agency, custom work to actually thinking
- 16:20like a product owner is harder than it
- 16:22looks. You have to learn to say no to
- 16:24feature requests. And the fourth is
- 16:26don't build a product so tailored to one
- 16:29client that no one else wants it. So I
- 16:30have four questions to ask yourself if
- 16:32you're going to take this approach. The
- 16:34first is what functionality have I built
- 16:37three or more times for different
- 16:38clients? What internal tools have I
- 16:41built to make my agency more effective
- 16:43or efficient that others might want? Do
- 16:45I have a clear path to transitioning
- 16:47from agency revenue to product revenue
- 16:49or will I be stuck doing both forever?
- 16:51And the last is are my clients in an
- 16:53industry I actually want to serve
- 16:55long-term? So real quick, if you found
- 16:57any of this helpful so far, can you
- 16:59please give this video a like and
- 17:00subscribe to the channel? My team and I
- 17:02are working really hard to bring you
- 17:04content like this. We put a ton of time
- 17:06and energy into making sure the content
- 17:09is super valuable for you. And when we
- 17:11see those metrics on YouTube, the likes
- 17:13and the subscribes, it helps us know
- 17:14that we're on the right track.
- 17:16So those first three methods all assume
- 17:18you have work experience to draw from.
- 17:21But what if you're early in your career
- 17:23or you want to look beyond your current
- 17:24industry? Method number four is solving
- 17:27a problem for a spouse, friend, or
- 17:29colleague and about 8% of respondents
- 17:33said this is how they found their
- 17:34startup idea. In this case, you're going
- 17:36to want to go on the hunt and try to
- 17:38find problems. So you'll want to reach
- 17:40out to your network. Maybe you talk to
- 17:42your spouse or maybe you talk to your
- 17:43friends to see what problems they have
- 17:46in their day jobs. John Sister is the
- 17:49founder of Quill Therapy Solutions,
- 17:51which he describes as software that
- 17:53generates progress notes and other forms
- 17:55of documentation for mental health
- 17:57therapists in an ethical and
- 17:58privacy-focused way without recording
- 18:01therapy sessions. John told us how he
- 18:03came up with the idea. He said, "My wife
- 18:05is a therapist and she 100% thought of
- 18:08the idea. She saw me working on a
- 18:10different idea that relied on AI to
- 18:12rearrange information and she basically
- 18:13said, 'Hey, could you do that for
- 18:15progress notes instead?'" And she was
- 18:17right. One way to approach this method
- 18:19is to have a chat with someone who has
- 18:21deep experience. Yoav Sigurðsson is the
- 18:24founder of CrankWheel, which is a screen
- 18:26sharing web conferencing and video
- 18:28platform for sales. And Yoav told us,
- 18:31"So my co-founder, Gilsi, had been
- 18:33working for years in various sales jobs,
- 18:35mostly sort of hybrid telephone and
- 18:37in-person sales, like insurance and
- 18:39banking, telecoms, that kind of stuff.
- 18:41And we started brainstorming about
- 18:43what's missing from that kind of space.
- 18:45We rediscovered a lot of sales
- 18:46enablement tools that already existed
- 18:48and I'd been working on real-time
- 18:50communications in Google Chrome at
- 18:51Google. So I innocently asked him at
- 18:53some point what people use when they're
- 18:55on the phone with a prospect and they
- 18:57need to show them something. As it
- 18:59turned out, they didn't use anything
- 19:01because it's too complex for a lot of
- 19:03prospects and you don't want to take a
- 19:04good sales call and ruin it with some
- 19:06tech issues. That's how the idea for
- 19:08CrankWheel was born." What makes this
- 19:11method work are the following four
- 19:13things. Number one, you have a direct
- 19:15line to someone experiencing the
- 19:16problem, instant access for interviews,
- 19:18feedback, and iteration. Number two, you
- 19:21can deeply understand the problem
- 19:22without having to cold call strangers at
- 19:25least initially. Number three, your
- 19:27relationship creates accountability.
- 19:29You're more likely to follow through
- 19:30when someone you care about is counting
- 19:32on you. And number four, they can
- 19:33introduce you to others in the field for
- 19:35validation and early sales. But there
- 19:37are things to watch out for. The first
- 19:39is your spouse or friend might have a
- 19:41unique problem that others in their
- 19:43field don't share, so you want to
- 19:44validate beyond them. The second one is
- 19:47they might be too nice to tell you your
- 19:48solution isn't good, so push for honest
- 19:50feedback. The third is mixing business
- 19:53and personal relationships can get
- 19:54complicated. So you're going to want to
- 19:56set expectations early. And finally, you
- 19:58still need to understand the market.
- 20:00Their expertise plus your building
- 20:01skills isn't enough on its own. If
- 20:04you're going to attempt this method,
- 20:06you're going to want to ask yourself
- 20:07four questions. The first is, what does
- 20:09my spouse, partner, or close friend
- 20:12complain about in their day job? The
- 20:14second is, do I think they can introduce
- 20:16me to other people who potentially work
- 20:18in the same field and have the same
- 20:20problem? It'd be nice to have five
- 20:22intros, for example. The third question
- 20:24is, am I building this to solve a real
- 20:26problem that I think can be a great
- 20:28business, or am I doing it to solve the
- 20:30problem of a person that I care about?
- 20:32And the last one is, do I have the
- 20:34patience to learn an industry that I'm
- 20:37not already familiar with? Method number
- 20:39five for finding startup ideas is to
- 20:41find a problem online, and only 3% of
- 20:46respondents used this approach. So this
- 20:49is where, instead of going to people you
- 20:50know, you are looking for ideas in
- 20:53forums, in online groups, whether it's
- 20:55Facebook or Reddit or private Slack
- 20:57groups you're in. Maybe you're lurking
- 20:59in support forums looking for problems
- 21:01that aren't already solved. And this
- 21:03method often gets touted online, but
- 21:05it's really hard to do, as evidenced by
- 21:07the fact that only 3% of folks who
- 21:09responded to the survey actually used
- 21:10it. It didn't show up much in our
- 21:12research, and only a handful of
- 21:13respondents had used this method. Couple
- 21:15examples are Tolu Akinola, who is the
- 21:18founder of String Thing. String Thing is
- 21:20a Shopify app to enable buyers to select
- 21:22a specific delivery date, for example,
- 21:24for gift purchases. Tolu told us,
- 21:27"Someone had recently asked me to help
- 21:29set up a Shopify store, and I noticed
- 21:31they had an open API, so I decided to
- 21:32build a Shopify app. I combed through
- 21:34the Shopify support forums to find
- 21:36issues that merchants were complaining
- 21:37about. Once I came up with a short list,
- 21:39I looked in their app store to find
- 21:41which apps addressed the issue. I
- 21:43prioritized issues that had apps with
- 21:45lots of reviews, but also a lot of
- 21:47negative reviews. This helped identify
- 21:49problems that were pervasive but not
- 21:50well addressed." Another example is Rami
- 21:53Koufach, the founder of Hovercode, which
- 21:55is a dynamic QR code generator. Dynamic
- 21:58meaning the scan destination can be
- 22:00edited after printing, and you get
- 22:02analytics. So Rami told us, "It was a
- 22:04mixture of having somewhat of a
- 22:05hypothesis or interest and keyword
- 22:08research. I found QR codes interesting
- 22:10generally, and after chatting with a
- 22:12friend who works in marketing, I started
- 22:14to think that marketers were still in
- 22:15the early stages of making use of QR
- 22:17codes, and there was still opportunity.
- 22:19I did some keyword research with Ahrefs
- 22:21to see if there was demand for QR code
- 22:23tools and saw huge search volume, which
- 22:25was enough for me to get involved in the
- 22:27space." What makes this work? Well,
- 22:29there's there's a bunch of things, but
- 22:31let's cover four. One is you can
- 22:32discover problems in industries you've
- 22:34never worked in. Public forums give you
- 22:36unfiltered access to real frustrations
- 22:38and complaints. You can validate demand
- 22:41through search volume, number of
- 22:42complaints, or size of communities. And
- 22:44lastly, negative reviews of existing
- 22:46products are a goldmine for
- 22:47differentiation. But there are things
- 22:49you should watch out for. The first is,
- 22:51people complain about a lot of things
- 22:52they wouldn't actually pay for. The
- 22:54signal-to-noise ratio can be low. You
- 22:56lack the context that insiders have. You
- 22:58might misunderstand the problem or the
- 22:59market dynamics. The third one is, this
- 23:01method takes a lot of time and patience.
- 23:04Most people give up before finding
- 23:05anything good. And lastly, selling to
- 23:07people you've never met in an industry
- 23:09you don't understand is genuinely hard.
- 23:12Some questions to ask yourself are, am I
- 23:14finding the same complaint repeated by
- 23:16many people, or is this just a one-off
- 23:17gripe? Would people pay money to solve
- 23:20this, or do they want to vent? Can I
- 23:21actually reach these people to sell to
- 23:23them, or are they scattered and going to
- 23:25be hard or expensive to find? And am I
- 23:28willing to invest the time to deeply
- 23:29understand an unfamiliar industry before
- 23:32I start building? Among our respondents,
- 23:34there were a couple of people who had
- 23:35grown nice marketplace businesses. And
- 23:37if you're interested in building an
- 23:38add-on for an existing platform, my team
- 23:41has pulled together a list of over 75
- 23:43marketplaces at
- 23:44microconf.com/marketplaces.
- 23:47In addition, you can look at Reddit
- 23:48communities related to specific
- 23:50industries or professions. You can look
- 23:52at Facebook groups for niche
- 23:53professional communities, support forums
- 23:55for existing products where you want to
- 23:57look for repeated complaints, review
- 23:59sites like G2, Capterra, or app store
- 24:02reviews, and finally keyword research
- 24:04tools like Ahrefs, Semrush, or even
- 24:06Google's keyword planner. So I have one
- 24:09real-world example from someone who
- 24:11completed the survey. And in this
- 24:13example, they actually noticed something
- 24:15while they were out in the real world.
- 24:16So Steven Gladney is the founder of
- 24:19OrderNerd, and OrderNerd aggregates
- 24:21third-party online ordering apps like
- 24:23DoorDash, Uber Eats, Grubhub into a
- 24:25single platform to make it easier for
- 24:27restaurant owners and staff to manage.
- 24:30Steven told us,
- 24:31"We already had our team, a sales guy
- 24:33and a developer, and we're actively
- 24:35hunting ideas. One day my co-founder was
- 24:37at a restaurant and saw a wall of
- 24:38tablets and began talking to the staff
- 24:40about it. The pain in their voice was
- 24:42palpable. I laughed because, man, when
- 24:44you when you hear a pain point like
- 24:45that, it's it's often a a good sign. He
- 24:47texted me a photo and said, 'It might be
- 24:50the beer talking, but I think this is
- 24:52the problem we need to solve.' Given
- 24:53that I had previously sold to
- 24:55restaurants for 5 years, and he had over
- 24:57a decade of experience working with
- 24:58third-party APIs, we knew we were the
- 25:00right people to tackle this problem."
- 25:02Obviously, this is a unique example that
- 25:04might be hard to replicate, but it's a
- 25:06good reminder that by remaining curious
- 25:08and observant, you might find an idea in
- 25:10a place you wouldn't expect. Method
- 25:12number six is to scratch your own itch,
- 25:15and this one was made so popular in the
- 25:18early and mid-2000s
- 25:20by Basecamp, by DHH and Jason Fried,
- 25:24because they scratched their own itch
- 25:26and it worked for them. And so they
- 25:28would often talk that maybe that's the
- 25:30best way to start a startup. And while
- 25:32it's certainly not a bad way, only 16%
- 25:36of our respondents did this. This is
- 25:38where you build something to solve your
- 25:39own problem, and you discover that other
- 25:40people have the same issue, and you can
- 25:43find them, and they will pay for it. So
- 25:46there's a lot that has to come into
- 25:48play. So there's a lot of things that
- 25:49have to line up in order for this to
- 25:51work. You can't just say, "I have a
- 25:53problem, therefore that's validation."
- 25:54It's not. You validated that you will
- 25:56have one customer paying you $0, which
- 25:58is yourself. So when I think about
- 26:00scratching my own itch, which is I I've
- 26:02done I built Drip was a originally
- 26:04software to scratch your own itch. And I
- 26:06took that idea and went and asked 17
- 26:10other founders that I knew if they had
- 26:11the problem. Then when I got 10 or 11
- 26:12yeses, then we started building. So if I
- 26:15was going to do scratch my own itch, I
- 26:16would want to validate there other
- 26:17people out there with this issue.
- 26:19Examples of founders who scratched their
- 26:21own itch include Olga Khusar, founder of
- 26:24Dialog Shift, which is an AI application
- 26:26for hotels. It's chatbot, phone AI,
- 26:29email AI. Olga told us, "We came up with
- 26:32the idea as hotel guests ourselves,
- 26:34frustrated by how difficult it was to
- 26:36get simple information about what,
- 26:37where, and when things were happening
- 26:38during our stay. Our initial thought
- 26:40was, how great would it be to just send
- 26:42a WhatsApp message and instantly get
- 26:44answers to all our questions about the
- 26:45hotel. For example, when is the next
- 26:47yoga class? Till what time is breakfast
- 26:49served? Can you send me the daily
- 26:51program?" One more example is from Scott
- 26:53Brownlee, the founder of Simple Booklet.
- 26:56Simple Booklet is marketing,
- 26:58communication, and support collateral
- 27:01presentation that includes distribution
- 27:03and tracking across the web. Scott told
- 27:05us, "I had sold an earlier business in
- 27:07the early 2000s, then went sailing with
- 27:09my wife for 11 years around the world.
- 27:11Getting access to information about
- 27:12places was difficult, and usually was in
- 27:15the form of paper brochures, lookbooks,
- 27:17and guides. I really just wanted to scan
- 27:19a QR code or get it on my tablet or
- 27:20phone to access. And since every print
- 27:23brochure started as a PDF, how could we
- 27:25make that more accessible?" What makes
- 27:28scratching your own itch work? Well,
- 27:30first of all, you understand the problem
- 27:31intimately because you've experienced it
- 27:33yourself. You're motivated to solve it
- 27:35because it genuinely bothers you. You
- 27:37can be your own first user, which speeds
- 27:39up iteration. And your passion for the
- 27:41problem often comes through in marketing
- 27:43and sales. But there's a few things you
- 27:44want to watch out for. The first is, you
- 27:46might be a market of one. I referenced
- 27:49this earlier. Just because you have this
- 27:51problem doesn't mean enough other people
- 27:52do to make a business. The second is,
- 27:54you might build what you want rather
- 27:57than what the market wants. So stay open
- 27:58to feedback. The third, your use case
- 28:01might be too niche or too specific to
- 28:02your situation. And the fourth is,
- 28:04founders often underestimate how
- 28:06different their needs are from typical
- 28:08customers. So, questions to ask
- 28:10yourself. Number one, [snorts] how many
- 28:12other people have this exact problem?
- 28:13Can I find them and talk to them? Am I a
- 28:16representative user, or am I an edge
- 28:18case? Would I pay more money to solve
- 28:19this problem, or do I just find it
- 28:21mildly annoying? And lastly, can I
- 28:23separate my personal preferences from
- 28:25what the market actually needs? Since we
- 28:27did all this research, I'm actually
- 28:29going to take a lot of it and put it in
- 28:31a new book I'm working on on finding,
- 28:33choosing, and validating your idea,
- 28:36finding product-market fit, and
- 28:37launching a profitable SaaS business. If
- 28:40you want early access to chapter
- 28:41previews, launch updates, that kind of
- 28:44thing, hop on my email list,
- 28:45robwalling.com/
- 28:47emails. Method number seven is taking
- 28:50advantage of an emerging or a
- 28:52fast-growing technology. This is every
- 28:55developer's dream, where AI becomes a
- 28:57thing, or DHTML, remember Ajax in like
- 29:00what, 2006, where suddenly you could do
- 29:02stuff in a browser you couldn't do
- 29:03before. And you see it and you think,
- 29:05"What can I build with this?" And while
- 29:06it's every developer's dream to play
- 29:08with that new technology, it's very
- 29:09rare, at least according to our survey,
- 29:12that people actually turn these into
- 29:14businesses. Only 3% of respondents went
- 29:17this route. And when I say an emerging
- 29:18technology, this can be AI, it can be a
- 29:20brand new API, it can be a new platform,
- 29:23like when I think Facebook opened up for
- 29:25development and uh you know,
- 29:28Squarespace, HubSpot, like all these
- 29:29marketplaces, Shopify. When these open
- 29:32up, you can look for practical
- 29:34applications before the market gets
- 29:36crowded. Examples of this include Rom
- 29:38Shingale. He's the founder of WA
- 29:40Notifier, which is an all-in-one
- 29:42WhatsApp marketing tool for businesses.
- 29:44Rom told us, "I used to run a WordPress
- 29:46agency, and setting up email marketing
- 29:49forms was one of the recurring things we
- 29:50used to do for businesses in Western
- 29:52countries. With WhatsApp gaining
- 29:53momentum and WhatsApp marketing starting
- 29:55to pick up, I made an assumption that if
- 29:58email and email marketing is big,
- 30:00WhatsApp marketing can also become big.
- 30:02This was especially for businesses in
- 30:04countries where WhatsApp was popular and
- 30:05a primary channel for communication like
- 30:08India, GCC countries, South America,
- 30:10etc. And then right around 2000, when I
- 30:13was looking for SaaS ideas, WhatsApp
- 30:15opened up their gates for new API
- 30:17partners. So, the timing was right and
- 30:19that's how I got started. Another
- 30:21example is from Alex Charles, founder of
- 30:23Gen Text AI. Gen Text AI is applied AI
- 30:27add-ins for Microsoft Word, expanding to
- 30:29other AI applications. Alex said, "I
- 30:32worked as a consultant and economist and
- 30:34used the Microsoft Office suite heavily.
- 30:36As soon as Chat GPT came out, I realized
- 30:38it would be extremely useful if
- 30:39connected to Office and if it could
- 30:41reduce hallucinations. That's how the
- 30:43idea came about." Then there's Tiago,
- 30:45the founder of PodSqueeze. PodSqueeze is
- 30:47an application we actually use to help
- 30:49produce my podcast Startup to the Rest
- 30:51of Us. It's an all-in-one for the
- 30:52production and promotion of podcasts.
- 30:55So, you upload audio or video and it
- 30:57generates social media posts, clips,
- 31:00summaries, transcripts, and more. Tiago
- 31:03said, "I had my podcast and got really
- 31:04frustrated with all the post-production
- 31:06work. In other words, creating the show
- 31:08notes, promotional content. So, back
- 31:10then Chat GPT had just been released and
- 31:13we thought that even if the AI
- 31:14transcripts made mistakes, GPT was smart
- 31:16enough to get that content and generate
- 31:18good text assets. That's how we
- 31:20started." So, what makes this work?
- 31:22Well, first of all, early movers can
- 31:23establish themselves before the market
- 31:25gets crowded. New technology often
- 31:27creates new problems that need new
- 31:28solutions. The third is you can ride the
- 31:30wave of attention and interest in the
- 31:32technology itself. And lastly, customers
- 31:34are often more forgiving of rough edges
- 31:36when the technology is new. But there
- 31:37are some things to watch out for. First
- 31:39is timing can be extremely difficult. If
- 31:42you're too early, the technology isn't
- 31:44ready or the customers aren't educated.
- 31:46too late, then you're one of dozens or
- 31:48hundreds of competitors. Second thing is
- 31:50the hype cycle can mislead you. Just
- 31:52because people are talking about a
- 31:53technology doesn't mean they'll pay for
- 31:55whatever you're building. The third is
- 31:57the technology might not mature the way
- 31:59you expect, leaving you with a product
- 32:01built on shaky foundations. And lastly,
- 32:04big players can also move fast in hot
- 32:06spaces. So, you might get crushed by
- 32:07well-funded competitors. The questions
- 32:10to ask yourself are, is this technology
- 32:12actually ready for real-world use cases
- 32:14or is it still experimental? Second one,
- 32:17are businesses already spending money in
- 32:18adjacent areas or is this entirely
- 32:21speculative? Do I have a unique angle or
- 32:23insight or am I just chasing hype? And
- 32:25lastly, can I build something useful
- 32:27quickly before the window closes? And my
- 32:29eighth and final method for finding SaaS
- 32:33ideas is building on an existing
- 32:35product. 4% of respondents told us this
- 32:38was their approach. So, this is where
- 32:40you're working on one product and you
- 32:41discover a bigger or a better
- 32:43opportunity. Maybe an internal tool with
- 32:46outside demand or an adjacent problem
- 32:48that your customers keep mentioning.
- 32:50This obviously wasn't a huge group, it
- 32:51was 4%, but we did have a number of
- 32:53respondents who were working on a
- 32:54product at various levels of success and
- 32:57then in doing so discovered a bigger
- 32:58opportunity. It's worth noting these
- 33:00weren't necessarily pivots, but new
- 33:02products altogether. So, Ivana is the
- 33:05founder of Authored Up, which is
- 33:07LinkedIn content creation and analysis
- 33:09integrated via Chrome extension. And
- 33:11Ivana told us, "We were working on a
- 33:13startup for hiring managers and
- 33:15recruiters. We had a product developed,
- 33:16but we were working in a vacuum, not
- 33:18talking to potential customers, no mom
- 33:20test. After we almost finished the
- 33:22product, we wanted to get feedback and
- 33:24beta users. We tried networking, sending
- 33:26LinkedIn DMs, and pitching. We realized
- 33:28we can try creating content on LinkedIn,
- 33:30but we couldn't find simple tools to
- 33:32help us out and that's when we built
- 33:33this internal tool." Refael, who's the
- 33:35founder of Super Data, which is an API
- 33:38that lets developers and no-code makers
- 33:39easily extract data from websites and
- 33:41platforms like YouTube, TikTok,
- 33:43Instagram, and X Twitter. Refael told
- 33:45us, "It's my own itch which turned into
- 33:48a business. I needed to scrape YouTube
- 33:49transcripts to add a feature to my other
- 33:52product. Over time, I also used the
- 33:54scraper to build other small products.
- 33:56It became de facto an API servicing
- 33:58multiple apps. At some point, I decided
- 34:00to make a small experiment and try to
- 34:02monetize it on RapidAPI and from there
- 34:04it snowballed." And lastly, we have
- 34:05Ruben Gomez, the founder of SignWell.
- 34:07SignWell provides easy-to-use document
- 34:09signing with enterprise-grade security
- 34:11and compliance at an affordable price.
- 34:13Ruben told us, "The idea came from
- 34:15running another business and
- 34:17seeing/hearing about the dissatisfaction
- 34:19of customers with solutions in the
- 34:21e-signature space and trying ourselves
- 34:24to give them good options for that part
- 34:26of their workflow." So, what makes this
- 34:28approach work? The first is you're
- 34:29already in the market. You're talking to
- 34:30customers and you're seeing real
- 34:32problems. You have momentum, skills, and
- 34:34sometimes revenue to fund exploration.
- 34:36The third is you might discover that a
- 34:38tool you built for yourself has broader
- 34:39demand. And the fourth is the new
- 34:41opportunity often leverages what you've
- 34:42already learned. Things to watch out for
- 34:44include chasing shiny objects. It's a
- 34:47real risk. Make sure the new idea is
- 34:49actually better and not just more
- 34:51exciting.
- 34:52The second one is splitting focus
- 34:54between two products can kill both of
- 34:55them. Third is you might be running from
- 34:57hard problems in your current product
- 34:58rather than toward a real opportunity.
- 35:01And the fourth is sunk cost in your
- 35:03current product can cloud your judgment.
- 35:05So, I have four questions I think you
- 35:06should ask yourself if you're going to
- 35:07take this approach. Number one, is this
- 35:10new product idea genuinely bigger and
- 35:13better or am I just bored with what I'm
- 35:14working on? Number two, can I validate
- 35:17this quickly without completely
- 35:18abandoning my current product? Number
- 35:20three, do I have the resources to pursue
- 35:22both of these ideas or do I have to make
- 35:24a really hard choice? And number four,
- 35:27what would have to be true for this new
- 35:28idea to be worth the switch? So, as a
- 35:31recap, here are the eight methods that
- 35:33the successful founders we surveyed used
- 35:35to fund their profitable SaaS ideas. Let
- 35:38me leave you with one number, 72.
- 35:41That's the percent of founders who found
- 35:43their idea through their work, be it day
- 35:45job frustrations, expensive tools they
- 35:48knew they could build better or cheaper,
- 35:49patterns from agency clients, or
- 35:51opportunities discovered while building
- 35:53something else. 72%.
- 35:56That means your SaaS idea might be right
- 35:58under your nose. But you need to know
- 36:00what to look for and that's why my team
- 36:02and I created the 9-to-5 audit. It's
- 36:04totally free. It includes even more
- 36:06founder examples we couldn't fit in this
- 36:08video and it walks you through thought
- 36:10exercises to help you discover SaaS
- 36:11opportunities hiding at your work day.
- 36:14You can get your free copy at
- 36:15robwalling.com/ideas.
- 36:17I hope you find it helpful. Once you
- 36:19found your SaaS idea, the next step is
- 36:21validation and that's exactly what I
- 36:23cover in this next video. If you found
- 36:25this video helpful, I'd appreciate it if
- 36:27you give it a like and subscribe to the
- 36:28channel. Thanks to all the survey
- 36:30participants for helping out with this
- 36:31project. Thanks for watching and we'll
- 36:33see you next time.
About this transcript
This page contains the full transcript of Give Me 36 Minutes and I'll Teach You How to Find $1M SaaS Ideas by Rob Walling, generated from the public captions YouTube serves with the video. The transcript has 7,185 words across 1,079 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.