Startups (in particularly the tech kind) are renowned for their unique difficulties, for the fact that each startup faces an extraordinary set of problems that has never been seen before in this particular conjunction. This is largely borne out by the many people's definitions of "startup", which are all about uncertainty, hard technical problems, scalability, and so on.
However, in my experience, what usually kills startups is a set of much more mundane issues, like running out of cash (the ultimate killer), building the wrong product, building for the wrong market (both of these effectively equate to "building a crap product"), having a major fight between the founders, etc. The root reasons (which can be declined in a thousand different ways, but are all essentially the same) map fairly closely to Paul Graham's How to start a startup principles:
- start with good people;
- make something people want;
- spend as little money as possible.
Most startups (particularly those started by new entrepreneurs) will fail because they screw up one of those three basics, not because they face a unique set of never-seen-before circumstances unique to their market, their product, and their specific set of uncertainties.
Start with good people
The way this usually unfolds is that the founding team splits, taking the startup down with them. If you started with people who were not committed, not a good match with you, didn't have the skills, didn't have the mad perseverance to see things through, were too many or too few, and so on, then chances are the team won't survive the stress of a startup.
Even a "regular" business is ultra-stressful, putting any relationship to the test. A startup, with its pressure to grow fast, is ten times as stressful.
The way to avoid this is two-fold: proactively and protectively.
Proactively, you make sure you only start companies with determined, motivated, obsessive nutters who have the same priorities in life as you do. No "Oh heck, I'm about to start a startup, I want my chum John to be my cofounder, I've known him since we were kids!" When starting something new and scary, your impulse might be to gather friends around. That's the wrong impulse. You need to be extremely selective, and only start business with someone who is going to be as crazy about it as you will be.
Protectively, once you've picked the right person, make sure your arrangement is sustainable, and make sure the company can survive one of you changing their life priorities (it can happen to you too!). This means asking yourselves the right questions and picking the right kind of equity split approach. Vesting is not an automatic cure-all, though it helps in some cases.
Make something people want
There are a number of approaches to this. My favourite is obviously hypothesis-driven development, but all the variants of Lean Startup and Customer Development and other idea validation approaches will help you there.
The key mistake to avoid (which oh-so-many promising startups make) is building something in isolation, without any customers, with only theoretical ideas about what will drive sales of the product or user adoption. Don't fall for this, it's really tragic, it's avoidable, and it sucks when it happens to you (it happened to me twice, in variants).
An important caveat is to make sure you're fitting your product to the right market. If you're planning to eventually sell your product, don't fit it to a free-wheeling user base.
Spend as little as possible
Another way to phrase this would be "be on top of your finances". If you're bootstrapping, that'll probably come naturally (after all, it's your pennies going down the drain), but if you managed to raise funding, this can be harder than it seems.
You need to be on top of your accounting and your cash flow. This can't be delegated to someone else until you personally understand it. Similarly, even if you've raised funding, don't spend irresponsibly. It's better to be a zero-cost entrepreneur than to be an uncontrolled spender.
In short, this is about being aware of and on top of the basic life pulse of your business: its cash flow. A lot of startups stumble and fall because they've taken on way more expenses than they could afford given the stage and success of their project.
Failing in an interesting and unique way is hard. The average new entrepreneur's first startup will not face fantastic and exceptional problems - it'll die because of a combination of starting with the wrong people, building something nobody wants to pay for, and failing to be on top of the finances.
The good news is, if you can prepare yourself in those three areas, and avoid the most likely mistakes there, you'll greatly reduce the chances of a first-time flop.
If you read this far, you should follow me on twitter here.