How To Start A Startup (The FAST Way!)

Welcome to How to Start a Startup. In this course, I'll share with you the exact steps I took to build my own business, Contact Out, to eight figures in revenue. Today's video is about the process of executing on a startup. The startup process can be broken down as LIFT, learnings, ideas, and tests. The faster that we can go through this iterative process of coming up with business ideas, validating and testing our business ideas, learning from our tests, refining our business ideas, and doing this iteratively, that's how we ultimately succeed at our startup. Let's dive into how to get startup ideas. I'd start by picking an industry with a big market size that I'm passionate about. I'd find a hundred companies. I'd study the sales, their products, their marketing. I'd talk to ex-employees, ex-founders, and industry experts. And I'll talk to at least a hundred users to really understand their problems. Through this process, I'll be able to become an expert myself and use all this information to create a product and solution that is 10 times better than whatever else is already out there. I go through this process in detail in my video, how to get startup ideas, but let me give you the highlights. I'll start by looking at a hundred companies in my industry recruiting by going to the Y Combinator startup directory, or by searching recruiting on Crunchbase, where there's 2000 companies, or by looking at the latest recruiting companies on Product Hunt. For every one of these companies, I'll study their entire product, go through every screen and user flow, and assess what are their strengths and what are their weaknesses. For example, I studied ZoomInfo, Apollo, and Lucia. I'll study their marketing. Tools like SEMrush can tell me where competitors are getting their traffic from. Rockerreach, for example, gets a lot of traffic from Google for pages that provide contact data for people like Bill Gates. So we've decided to create our own page for Bill Gates' email. I can study my competitor's advertising copy on the Facebook ad library, for example, Lucia. I can study social media accounts and learn from their content in order to create our own content. I can look at Apollo's affiliate program in order to create our own affiliate program and sign up partners like Greenhouse. I'd learn about sales process by booking sales calls with all of my competitors using a different email. For example, I've recorded sales calls with ZoomInfo, HireEasy, Apollo, Entelo, and Rockerreach. I'd also collect sales collateral from Connectifyer, from Entelo, and from TalentBin. So we'd take the best parts of all this and then we'd create our own sales process from this. I'd reach out to ex-employees at competitors and ask them to mentor me. For example, Santosh, who was the COO at Apollo, agreed to advise us, giving us proven strategies for growth. And eventually, Santosh agreed to join us as co-CEO. You can also use sites like intro.co for mentorship with billionaire founders. For example, I scored a mentorship session with Alexis O'Hara, who is the founder of Reddit. Finally, I'll talk to at least 100 users and ask them what are their biggest problems, what is their day-to-day workflow look like, and what tools do they currently use? What we're trying to figure out is, what would users pay a lot of money for? We could just ask them this, or a nicer way to put it would be, what's something that's really valuable for your company that you're not currently getting from your current software vendors? From this research, I'd develop a bunch of ideas around the problem that we're solving, recruiters finding it hard to reach candidates and hire them, LinkedIn messages having a low response rate, and other prospecting platforms providing only work email. Our unique value proposition being that recruiters prefer personal emails and mobile numbers for a higher response rate, and marketing via cold email as well as SEO like our competitor Rocketreach. A price point of $2,000 as well as a sales process that we developed by combining the best practices of our competitor sales pitches that we listen to. The next step is to test and validate our business ideas. We want to ask ourselves, what are the biggest risks and what is the fastest way to test and learn? There are a lot of risks and unknowns like, am I solving a real problem? Is my solution 10 times better than the alternatives? Will people buy this? How would I reach customers? And how would I build the product? A common mistake is to start building the product without validating whether users would buy it. For example, I spent one year building a time management tool called Focus A Lot, where I built a desktop application for Windows and for iOS and I learned two programming languages. When I finally launched, I found out that most people don't want a tracker on their desktop that tracks their productivity and application usage by the minute. They find this invasive. I could have found this out by conducting user interviews instead of wasting one year building the product. Google Glass is another example where Google spent a billion dollars in product development costs only to find out that people don't want to wear robot glasses that makes them look socially awkward and records all their friends to make everybody uncomfortable. What you can quickly validate is, am I solving a real problem and is my solution 10 times better than alternatives? For example, Google Glass was not solving a real problem. Nobody wanted to wear Google Glass. For a contact out, we spoke to 10 recruiters and the consensus was that it's difficult to get in touch with candidates, especially software engineers. Candidates don't show up to interviews and this is despite companies spending thousands of dollars on LinkedIn recruiter and tens of thousands of dollars on recruiting fee. So is our solution 10 times better? LinkedIn has low response rates and recruiters say that they prefer personal emails, but to be honest, we didn't really get a clear answer here. Here's some more examples. Lumini helps customer support reps save time. The first iteration was a simplified user interface for Zendesk that included keyboard shortcuts. The response to this was lukewarm. People were willing to try it. They're like, They then pivoted to a full customer support automation where with the click of the button, a customer support rep can issue a refund and Lumini will go into Stripe and process the credit card refund. It will go to FedEx and lodge a return slip that is then emailed to the customer. And then it will go into the accounting system to log the refund and reconcile the accounts, saving the customer support rep up to 15 minutes of time. There was an overwhelming response to this where users signed up and pay thousands of dollars for the product before it was even built, which is complete night and day compared to the first response. And off of that, Lumini managed to raise $20 million in venture funding. Another example, Rupert helps started as a platform to find holistic doctors like naturopaths, chiropractors and Chinese herbal medicine doctors. However, after one year, they had very little traction and they learned that when you're trying to find a holistic doctor, it's usually done by a word of mouth because it's a high-trust situation. You also stay with the same doctor, unlike most marketplaces like Airbnb, for example, where you're coming back to the platform again and again. With holistic medicine, you're usually just looking for one special doctor. So an adjacent problem that Rupert learned about in their interviews with doctors is that holistic doctors find it very difficult to order lab tests. There are hundreds of health tests for patients and dozens of labs. Doctors are confused about where to go. So Rupert launched a lab testing platform where holistic doctors can order all their specialty tests in one place. The demand was overwhelming. Doctors wanted to use the product before it was even ready. In fact, they had to rush the launch date because doctors were depending on the platform to be ready in order to launch their clinics. Mark Andreessen, who is the founder of the web browser, talks about product market fit and market pool as when customers are literally banging down your doors demanding your product and they're happy to use a really crappy first version because the problem they're facing is so painful and there are no good alternatives. Y Combinator describes this as your hair is on fire and the best solution is a bucket of water. However, if somebody gave you a brick, then you'll use the brick, bang it against your head to put out the fire because the problem is so painful. A good resource to check out is the product market fit series by First Round Capital, where you can read 20 firsthand stories that detail the exact steps taken by unicorn companies like Airtable or like Webflow to reach product market fit and the pivots that they made along the way. The next thing to test and de-risk before you build the product is will people buy it? So at Contact Out, we couldn't actually get people to buy the product before we built it, but we did manage to cold email a lot of recruiters and have them sign up to a wait list and then we spent four weeks building a prototype and we were able to sell that. More on that story later. Here are some more examples. Ed Roller provides learning resources to high schools like recorded lectures, lesson plans, and homework assignment templates. They sold to schools using a PowerPoint presentation and closed thousands of dollars in business before they had any learning resources created. Since schools planned curriculum in advance for the following year, they then had six months to go hire a team and create the learning video. Shaving brand Harry's is another example where they had a pre-launch sign up that they promoted via press releases and influences on social media. Then after people signed up, they would encourage them to invite their friends to get free products like shaving cream or an additional razor. The budgeting and personal finance app Mint.com started as a personal finance blog where they talk about credit cards and budgeting and they managed to sign up 100,000 users in six months just off of the content. You can read about the case study on Noah Kagan's blog. In terms of validating how would I reach my customer, another good example is the makeup brand Live Tinted where they built a Instagram following of 250,000 before they launched any products, hence validating that they're able to use social media as an acquisition channel. You can read about the case study again on First Round Capital's blog. Finally, there's actually building the product which typically takes the most amount of time which is why we want to validate that there is demand first. So for our first product, we want to ask ourselves what is the smallest thing that I can build that provides the most value to users and ideally is buildable within one to four weeks. If it takes longer than four weeks, we want to break it down into smaller chunks or we want to be very very sure that we've proven demand enough to warrant the extra time investment. With Contact Out, we have a lot of features on our prospecting platform like a search model, email campaign tool, a chrome extension for LinkedIn, but the first version was simply a pop-up on top of LinkedIn that gave you an email address. We did spend a lot of time building our email database, however in hindsight that was a mistake. What we should have done is go to all our competitors, buy all their email data and aggregate it and then unsell it to customers which is something that we could have done in two to four weeks. Let's look at how some other companies develop the minimum viable product. The smallest thing that they can build in under four weeks that provides the most value to users. One quick approach is to start by manually providing the service before automating it. Pilot is the biggest startup accounting firm in the US, however they started with one co-founder manually doing the bookkeeping. From there they learned about where the inefficiencies were in the accounting process and also what the customers wanted and what providing excellent service looks like. The other co-founder was then busy studying APIs to automate QuickBooks, Xero and other accounting software and then figuring out how to use code to automate and make more efficient the accounting process. Zendit is a payment distribution service where companies can pay their suppliers and business partners in one go by uploading a spreadsheet. They also started manually by manually logging into hundreds of bank accounts to process payment. This way they could start to provide the service immediately and validate demand whilst at the same time they were building a automated payment system that would integrate with all the banks. Then there's ZoomInfo which is a prospecting platform and one of my competitors. So building a search portal for prospects where you can search by job title, company size, technologies used, lots of other filters takes a long time and a lot of data processing. Instead ZoomInfo started by manually selling lists of prospects. For example customers will provide details like I want a VP of engineering in California that has experience in big data and then ZoomInfo would manually go and compile this list and sell the data to the customer in a spreadsheet. Another strategy to build quickly is to provide one product line first. For example Amazon started with only selling books before selling everything else. This is their first website. The Rounds which is a grocery delivery service started with only delivering cereal because it was easy to deliver and everybody needed it. Zapier which is an application that connects 6,000 other applications started with only one integration which was after a person fills out a form on a website they are added to a email tool. Another approach is to white label or rebrand a competitor's product and start selling that to learn the sales channel before building your own product. For example Qubicost monitors your cloud server hosting costs. Their first product was built over a weekend using a open source software from GitHub called Grafana. Open source software gives a great head start because you can build off of it, rebrand it, charge for it and you don't have to pay any fees. Another example Airalo is a provider of virtual SIM cards that allow you to buy data for over 200 countries in the world and you just download the SIM card to your phone. You can start a similar service by using a white label provider like Keepgo where you can wholesale buy mobile data and then rebrand it and resell it. And as I mentioned we could have gotten started with contact out a lot faster by buying competitor contact data, aggregating it and reselling it but we didn't do that. When you're working on a moonshot project and you can't build the first version in under four weeks you still want to pick the smallest thing that you can build. For SpaceX Elon went to Russia to try to buy and repurpose a intercontinental ballistic missile and that was the first thing he tried before deciding to build his own rockets. It's the same thing for Tesla, building a car takes a really long time. For the first version of Tesla instead of designing it from scratch what they did was they used the body from the Lotus Elise and then they used an electric drive train which was already developed by a separate company and then they just put the two together. The point of all this is to avoid building stuff that nobody wants so we build something really fast and get it into the hands of users to prove that people will buy it. Going back to our startup process after we tested our business ideas we'll have new learnings about what worked and what didn't work and then we'll form new ideas and test those and we'll just keep repeating this cycle of learnings, ideas and fast tests. Tests should not take longer than one week to complete and we should break things down into one week experiments as much as possible. The goal of this is to learn something every week and improve our business ideas instead of being stuck for six months building a product only to find out afterwards that nobody wants it. Here's what this looks like at contact out. So every Monday everybody at our company posts on a slack channel hashtag goals and they write about the goals that they accomplished last week, the tests that they ran, what did they learn, what can be improved as well as the tests and goals that are scheduled for next week. Some of my recent goals included releasing a AI email writer feature that uses chat gtp to automatically write personalized emails that are based on a person's LinkedIn profile, improving our sales operations, our lead assignment and lead scoring processes and studying the AI courses from Stanford. So what I learned was for sales operations we needed to focus on things like reply rate, book meetings, sales pipeline and we're building a new dashboard based on this. We also need a better product roadmap overall that includes AI features with the new chat gtp stuff that's coming out. So my goals for next week include working on a two-month product roadmap, studying all the other AI applications out there and validating if they are useful to our users recruited. Our goals and tests are a bit more developed because we've been running for eight years, we have a 60 person team and we found some product market fit already so we're focusing on discovering new features that add more value and improving our sales process. Here are some more examples from the team. Our account executive Ursyn is focused on testing out different email campaign messaging in order to get a higher response rate and book more meetings. One of our product teams is working on improving the speed of our search portal, SEO and growth metrics and optimizing our mobile site experience. We aim to push code every week but realistically we ship perhaps every two to three weeks. Our product designer Siren was working on usability testing for some of our new AI features. We tested the product with eight users, we got the feedback on what parts were hard to use and now her goal is to implement some of these changes in new designs to give to the engineers. Finally, our head of content Joseph is working on creating courses for recruiting, sales and code outreach. What we learned was that we could use chatGTP to rewrite competitor articles as a shortcut to create content, however we found that the output wasn't actionable enough so our goal next week is to make the content more actionable by including step-by-step guidance and adding screenshots. So we just keep repeating this process every week, learnings, ideas, tests and this is what would lead our startup to grow. Y Combinator advises for five percent to seven percent growth per week, however for a contact out and for most startups you won't hit this. Instead your growth might look something like this where maybe for months or even years on end you won't grow and then suddenly you will grow by 10 times. It is helpful to set yourself a goal of growing seven percent per week even if you don't achieve it. What's more important is that when you don't achieve the seven percent growth you ask yourself what can I do differently next week to grow, what did I learn and you aim to learn something new every week. So every week I'd ask myself what's the highest impacting that I can do to grow my business with the least amount of effort and time. So you basically just keep repeating this lift process, learnings, ideas, tests every week and you do this for a very long time and eventually you'll Perseverance is what Steve Jobs said is the most important factor in determining startup success. Most people give up trying after two to three years and not getting any results. This is really strange. Let's say you're trying to be a doctor. Three years into medical school you drop out because you haven't made any money yet. Everybody will laugh at you. Of course you haven't made any money yet, you haven't even graduated from medical school. It takes 10 years to be a doctor, 10 years to be a trained lawyer, 10 years to be good in any profession. Yet for entrepreneurship people think it's different. People think that hey I can give up after two three years. No it takes 10 years to be good at anything including entrepreneurship. In fact everyone that I personally know that has been doing startups full-time for 10 years is making at least 1 million dollars in profit every year because they just kept at it. There's a common saying that 9 out of 10 businesses fail. This is completely wrong but let's assume that it is true. So what we'll do is we'll try a thousand business approaches over 10 years and then we'll be guaranteed success. Now oftentimes it's faster than this. For example I heard a story about a guy who at a weekend hackathon he managed to close $10,000 in sales because he spent the whole weekend just doing pre-sales. For a contact out it took us six months to make our first dollars of revenue and it took us about two years to get to 1 million in revenue. But you shouldn't plan to give up after two or three years and you can expect that it would likely take up to 10 years. That's it guys, thanks for watching and good luck with your businesses. Keep lifting, do you even lift bro? Persevere and I'll leave you with some wise words from Steve Jobs. I'm convinced that about half of what separates the successful entrepreneurs from the non-successful ones is pure perseverance. It is so hard, you pour so much your life into this thing so you got to have an idea or a problem or a wrong that you want to write that you're passionate about otherwise you're not going to have the perseverance to stick it through and I think that's half the battle right there.