Stop. Take a moment and look in the mirror. Disabuse yourself of any bullshit you tell yourself that you have a great idea, and approach your startup from a stranger's perspective.

Are you solving a real, discernible pain point that many people have every day? Either that, or is it something that a few large companies are willing to pay you a shit-ton of money?

Are you doing it in a way that is significantly different than anybody else? Be honest here. You don't have to be the only one in your field, but you better know exactly how you're different, and how you'll be better than the competition. Don't just say "They suck". Attach metrics or some low-level comparisons to your observations. People have been doing rideshares for decades (taxis), but Uber came along and made the transaction of money painless, provided a super easy way to book, and upgraded the experience. If your answer doesn't contain any specifics, or you utter the phrase "Well nobody seems to like it" or "We think we can do better" without justification, you're officially full of shit.

Now, if you have an idea, market opportunity, and positioning set, build the damn thing. Don't worry about corporate structure or registering as an LLC unless you absolutely know you need to beforehand (i.e. if you need a business checking account that requires an EIN). Stop telling yourself that there's a million things you need to do beforehand.

Coding: If you or your co-founder can code, great. If not, learn how to code quickly. Pickup the Agile Web Development with Rails 4 book on Amazon Agile Web Development with Rails 4 book on Amazon. If you know enough to disagree with me about Rails, pickup whatever language you want but not because it does or does not scale. Try lynda.com if you like having a teacher walk you through,codecademy.com if you like hands-on lessons with real coding, and w3schools.com if you like to move quickly. Nobody gives a shit about scalability or what language you use and before you ever get to the problem of having too much traffic, you'll probably rewrite a lot of this code so do whatever will get you to Minimum Viable Product or MVP as quickly as possible.

Frontend: Use bootstrap, optional customized version of bootstrap, or if you're blessed enough with an awesome designer who can also slice their code into HTML/CSS and do it in a modular format like bootstrap, then by all means do that.

Now, build it fast. If you're taking longer than 3 months and you're not building highly complicated shit like video compression algorithms, games, or anything else that requires a patent due to the complexity, you're doing it wrong. Get an MVP out and get it in front of customers.

Figure out your acquisition channels. DO NOT SAY "All of our customers have come from word of mouth" when someone asks you how you're getting customers. This merely shows that you have zero understanding of how you're getting current customers, and you probably haven't run campaigns or gotten evidence as to what acquisition channels, keywords campaigns, content marketing, organic traffic, SEO, referrals, viral metrics, or any other ways to get customers work for you. All you need to do here is maybe spend a few hundred dollars to a thousand dollars on AdWords, test to see if a referral rewards program works, look at maximizing your viral metrics, or any other way that you can answer in great detail when an investor asks you "What specific acquisition channels are you going to use to go from X customers to 10X customers to 100X customers? What data do you have so far that shows your acquisition channels are working?"

Start getting revenue, even if it's small. Since you're spending to acquire or getting customers organically, you should have enough people to start getting revenue, whether it's money collected from sales, products, percentage of customers who buy in-app purchases, freemium upgrades, etc. Know your conversion rate, i.e. how many visitors/users out of 100 become paying customers.

Finally, know your unit economics. You might only be making $1000/month, but if you're spend $20 to get a user who ends up paying you $50 and you can demonstrate that this is repeatable and scalable, then that's precisely the sweet spot where investors want to come on board. Your only problem now is that you don't have enough capital to throw at the problem.

Once you have these down, it's just a matter of getting to the right people.

Put up a profile on AngelList. You should be able to describe the key points of everything I've mentioned above in the first 2-3 paragraphs to an investor. If you don't mention traction, you're going to look like you don't have any so you better have a good fucking story to tell.

Apply to every startup accelerator you can find, but prioritizing 500 Startups, YC, TechStars, AngelPad, and a few other top-notch programs. Look up examples of application videos online. Make sure yours is better. You don't have to accept if they give you a spot, but it's best to keep all your options open. Life is a conversion funnel, don't minimize your options at the beginning of the funnel. Setup as many opportunities as you can, then feel free to say no to an accelerator program later.

Start hitting up investors on AngelList and tell them in 3 bullet points why they should invest (with traction being close to the top). Do this enough and you'll get a bite as long as your traction is up and to the right. Try to go to startup events, or network as much as possible. Any chances you can get of warm intros will greatly help.

Top 3 things to do while you're moving through these steps:

1) Don't overload yourself. The most common ways I've seen entrepreneurs fail are long, drawn-out discussions over corporate structure, choice of programming language, scalability concerns, and other topics which will ultimately not affect your ability to create a vehicle that takes in revenue from customers as quickly as possible. You could probably do a startup written in COBOL registered as an LLC in South Dakota and still technically make some money. Once you get money, lawyers will draft up a real corporate structure for you, and you'll most likely rewrite a lot of your system from scratch when you hire more tech talent. Relevant XKCD:http://xkcd.com/1445/

2) Start listening to This Week in Startups. It's an MBA through osmosis (no apologies to those who spent $100k+ on MBA's out there). Try to listen to at least 50-100 episodes, and you'll start getting a sense of what kind of challenges other successful entrepreneurs had to overcome early on. Pay attention to recent companies, as the circumstances of how AirBnB and Uber got their first investment check is probably somewhat different than how you'll get yours.

3) Stop talking up your startup to your friends/family until you meet specific milestones. Every time you open your mouth and talk to someone about how excited you are for your startup, or that you're doing really cool shit, you are letting a bit of air out of the balloon. When you stop talking about how great your startup is, you'll start to feel an impatience to get something really special accomplished and you can use that energy to drive yourself a bit harder. If all you talk about is how your startup is going to change the world, but you still have zero revenue, then STFU until you have something to actually brag about. I promise you, it will light a much-needed fire under your ass.

That's it! Good luck, and remember to be honest with yourself. Don't repeat the same bullshit. Challenge every hypothesis. And finally, remember that a startup is simply a temporary arrangement of people coming together to see if they can figure out a product-market fit before they run out of money.

The clock is ticking, so build the damn thing, get revenue, figure out your unit economics, and go get them investors.

Don't get investors. Get revenue.