Just start an experiment, not a damn startup
blog.fanchimp.com
blog.fanchimp.com
1.) Wait a week- if still excited about it go to 2.
2.) Discuss with my other half. As a non-techy she'll often have a very different view on things which is invaluable.
3.) Find and purchase a domain, and put up a sign up page based on a cheap theme.
4.) Buy some impressions from Adwords - this is enough to give me an idea of if anyone is already looking for what I want to build, and if they are get me some email addresses.
5.) Talk to prospects found in 4.)
6.) If still convinced its a good idea, I might start writing some code.
It is so easy to get going with writing code straight away, rather than thinking about whether you have any customers and how you're can reach them. The above approach has saved me many months of wasted effort, and found me a number of lucrative niches!
If you have picked an idea for which there is no searches yet, but people might want, don't give up. It's where you have to listen to guys like Hiten Shah and Eric Ries that say get out there and talk to customers first. Just know you'll be building the channels to reach your customers too and it's not the same as being able to look into search results that aren't being met with a product.
Can you give me more details?
What I will say- is that if a decent numberpeople are signing up with interest in your product, on the basis of a single page of html, with no release-date, screenshots or demos in sight, then its probably something they REALLY want!
Thanks for posting this. Seriously.
Test 1: are people clicking on my ad? are they at all interested in what i'm offering
Test 2: if not clicking, then how can i change the ad to gain clicks?
Test 3: if not clicking, how can i change the keywords i'm bidding on?
Test 4: if click throughs == true, how is the landing page working? am i converting people via an email sign-up or some other call to action?
Test 5: if people are coming to the LP but are not interested, how can i improve conversions? A/B test components of the landing page is a good start.
You ask any aspiring developer in Silicon Valley what they're working on and it's always "oh yea, I have a startup I'm trying to get off the ground on weekends."
In Toronto, there's a very vibrant startup community, but back there we call them "side projects". Really same idea but the word "startup" means something rather different than a few kids eating ramen. There are plenty of freelancers who band together into a consultancy and have a couple of startuppy products they're working on together. Effectively a bootstrapped team working towards a unified vision, but they'll still be shy about calling it a startup.
I suspect a lot of this has to do with the investor and acquisition community. As someone who moved down to Silicon Valley from Toronto, I was very uncomfortable calling what I did a "startup" up until I was raising funding or entertaining acquisition opportunities. It just sneaks up on you when it's an option, and these options hardly exist in Toronto.
This slow realization made me give up calling people out on calling themselves hackers. It used to be that a hacker was someone who was particularly adept at finding clever ways to misuse, change or exploit something. http://hacks.mit.edu/ and http://www.catb.org/~esr/faqs/hacker-howto.html were the canonical references. Hacks were contrasted with engineering -- both constructive creative output, but one involved detailed in-the-field knowledge while the other was a more principled approach. 'Street smarts' vs 'book smarts' if you will. Both were needed for a successful enterprise. Calling oneself a hacker was equivalent to giving yourself a nickname -- socially awkward and decreasing the likelihood the moniker would stick.
Now 'hacking' is synonymous with prototyping, or even more esoteric types of development (what the hell is 'hacking education' supposed to mean?). This isn't entirely untrue as quick prototypes often contain "awful, we shouldn't be doing this" kinds of code. However, they do tend to lack the righteous "awful, we shouldn't be doing this but damn is that awesome" code tradition calls for.
Like you said, definitions change and the only thing that matters is if your idea is communicated; ultimately you have to go with the flow...
I don't think "hacker" has yet been tarnished in Toronto the way it has been here, though.
The post says: "Just if things are going well you’ll create the company and build the structure of it."
This is the worst time to start the conversation about all the above.
I've been exposed to bitter disputes between company owners (not my company) about equity ownership and you can't possibly imagine how nasty things can really get.
These problems are not an issue when there is no pie yet, just the pan you plan to bake the pie in. Who cares about dividing that up. But pretend your business starts generating $100,000 per day in cash in the bank. Now ask yourself "who owns what?".
You may think you know your co-founders and co-owners well, but most people are unable to predict how they themselves will react in many situations, so don't assume you can.
My apologies if this post comes off as negative. Creating a company and watching it grow is one of the most exciting and rewarding things you can do, and it's so much more fun when you have co-founders and investors to share your success with. So to keep it fun, have the hard conversations when they're easy to have.
I blame this kind of mentality for much of the derivative rubbish the startup ecosystem is polluted with. We don't need any more photo/video sharing, link voting, or file sending apps.
I'm all for validating assumptions early, but some ideas (the ones that matter) are much, much bigger than 4 weeks to an MVP.
Incidentally,
I believe that if you want to create an Amazon (6+ years of under break-even) should be done when you have more experience as entrepreneur.
I think that a first time entrepreneur should resolve a real concrete problem, also not that big, but this way he will understand much more and he will have some chance of succeeding.
This is the opinion maturated for my personal experience.
Actually, that's exactly the method that me and my friendly developer are using for our own product, which should launch sometime this year. We both have a day job which would be hard to leave (we both have families to feed). And you're completely right that a single server is enough to host the MVP and we can expand further once we know we're on the right track.
However, don't bother with a team or co-founders. Just build an experiment yourself. Then use that experiment to find a co-founder if you don't have one.
I think co-founders should be people you've known for a long time, a decade or more (if you're 20 and haven't known anyone that long, then consider working for awhile and building out a network until you have someone you know well and trust and have known for 5 years.)
If you can't build the experiment yourself then you're not ready to lead a company in the space of the experiment (managers should know how to do their subordinats jobs).... so if you're an idea guy, figure out how to program. Do your experiment ins super simple code or easy to use tools, a real engineer can do it right later if the experiment succeeds.
And above all, don't build a company first. There is an unending list of chores that are involved there. All the time you see advice like "get a lawyer" and "get an accountant". You don't need an accountant when you have nothing to account.
[1] You actually say "build a team and skill the rest", but I'm pretty sure you mean "skip".
Although, a component of why I did that was because I had big dreams. I wanted to "play startup" because I didn't actually have any clue how to build one. The obvious thing that you see about every company is their public website and their logo, so it makes sense to start mimicking that first.
The irony is that once you start to understand how to actually build a company you often end up picking a name and incorporating as the LAST thing. In fact, you try to avoid it for as long as you reasonably can because it makes you lose your focus on just simply building an excellent product.
I always wanted to start writing the code first, but then started realising that was a waste of time if there was no-one to sell it to.
So now I always start with trying to find customers, via a simple sign-up page or similar, before I even think about writing code.
Not saying that either approach is right, but interesting that we've come at it from opposite angles, and both ended up changing our opinions completely!
By the way, I agree that you need a co-founder.