I was thinking of a model similar to the Pope commissioning Michelangelo to paint the Sistine Chapel's ceilings. It's not bad if it's something that you like to do and the Pope will pay for it. In this day, the Pope could be Google (after all, they do no evil...)
The problem with companies like Twitter is that they don't know if the Pope will like their art. My point is that if you are going to create one product to sell to the church, you should try to attract the Pope's interest as early as you can to reduce your risk.
Not a single YC company that I know of has made any remarkable amount of money without being sold. But, in fact, many YC companies have made considerable amounts of money by selling themselves.
Why is having the goal of being sold while making things people want not that great of a strategy?
But do you think any YC combinator wouldn't have eventually made a lot of money if they hadn't sold? Selling your startup doesn't cause it to have value; that's backwards. YC startups were sold for money because presumably the purchaser thought they would be worth at least that much money in the future.
If you can build something people want, it seems relatively easy to convince some fraction of those users to pay for it. Convincing a big company to buy you is a long shot; convincing a big company to do anything (even if you work there) is tough.
Look at Dinosaur Comics -- that guy can't even draw, but he's making a living at it. Now imagine trying to sell such a crazy idea to an existing company.
If you can make something people want, it almost seems irrelevant to think about big companies. They'll only consider buying you once you look like a solid business proposition already. It's a reward, not a bailout.
There are plenty of companies that get acquired before having a solid business proposition. Jaiku (along the lines of Twitter) is a recent example, but there are many others. A company can be acquired for a variety of reasons. Maybe the team is a good fit for the parent (talent acquisition). Maybe the product is hard to monetize on its own but it's a good feature that will increase the value of something else (could have been Xobni).
Perhaps it's a loss leader, a service that you plan to give away in order to attract customers to your main business (a contrived example, Godaddy could buy some game that somehow encouraged you to buy domain names).
"There are plenty of companies that get acquired before having a solid business proposition."
That doesn't make it a good idea. The answer to "Who is your target market?" should not be "Uhhh... Google?" There are many more companies that built a "cool" product that was useful and used, couldn't monetize it, and died because of lack of capital. Somehow I doubt YC has ever invested in a company that flat out had no exit strategy outside of being acquired by a big company. You've got to have something because if none of the big fish bite you're screwed otherwise.
If you're building something that might only be attractive to one or two buyers, but there's a pretty good chance those buyers would buy should you make the product attractive enough, isn't that a reasonable strategy? (Maybe not quite a sure thing.)
There's all the difference in the world between one and two. If there's just one, you're pretty much at their mercy. They can wait you out. The problem with two is that it's next to one. If one of the two has a bad year and stops buying startups, you're now down to one.
So much for microstrategy. The bigger problem is that if you're making something only one acquirer would want, it's probably not important technology.
it is like being an artist (let's say a painter). You spend 9 months day in and day out painting a "Mona Lisa". You better make sure all the notables like your work. If not you may as well have been working for the love of it because the guys who walk in the gallery will only eat your cheese, drink your wine, have a quick look at your painting and at best comment about it. But no money will be spent. Well unless you manage to sell them premium wine and cheese.
EDIT: @Diego. Had I read you comment, I would probably write mine differently. Apologies.
YC companies have a ready market of potential buyers because of pg's credibility. Building-to-sell would be significantly riskier and less strategically sound for a non-YC company.
Acquirers are looking for synergies that they can see with their current business (which they know well). Investors are looking at a range of possible exits and may be more open to possibilities. Firms like the Corum Group http://www.corumgroup.com/ specialize in software mergers and acquisition and offer a great "Selling Up, Selling Out" conference that's worth attending.
Investors are driven by optimism. M&A guys are as brutally pragmatic as sharks. Neither one is intrinsically more right than the other; they're just two different kinds of people.
Well, a "business model" to sell companies would be something like an incubator...a company that continuously created new companies and sold them off. Of course, since the talent that was able to execute on these ideas is going to be the most useful asset that these spinoff companies have, eventually your incubator will look more and more like an investment firm. If you think about it, investors are in the business of (helping to) build products to sell to other companies (or the public).
It's not a bad strategy in itself, but the lack of a business model makes it pretty hard to pitch the value proposition to any interested buyers. If the founders haven't been able to come up with a way to monetize the site, what's to think the buyers will?
I think you could count it as a business model: stay alive and thriving long enough to become an attractive acquisition target. And sure it's a good strategy -- at least, others are using it pretty well. Is justin.tv really planning to stage an IPO?
A business model that focuses on selling the company is risky. This applies even more so in twitter's case because they don't generate significant revenue at all, but they continually remain overvalued due to lots of hype. Classic web 2.0 tactics at work basically.
I think Twitter should put out a personally hosted version of their service, maybe sell it to people for a nominal fee ($10-25). That way the uptime is improved (spread over a lot more computers and decentralized) and they start to have a revenue stream. Providing both better service and making some money seems like a step in the right direction...
A "sell the company" business model is like becoming an actor with a mind of becoming rich like Tom Cruise. No doubt some people who do it will find some success, but by definition this is something that (a) works only for a small number of people (the world needs a limited number of movie stars, Google & Co acquire a limited number of companies) and (b) contains a large element of luck - many actors have comparable talent & looks, many companies can (and a few already have) build a Twitter-like product.
Selling the company: that's not a rare event. An IPO: that's like becoming an actor with the intent of being the next Tom Cruise. There are many, many buyers out there, hardly just Google. If you make something that people need (not merely what they think they want, but what they need) then you will be able to sell.
It's a valid strategy, but it's not a business model.
It's not that great a strategy either...