In the business world you'll often hear about 'first mover advantage'. It has an almost mythical quality to it: Gain first mover advantage, and all will be well. Of course, everyone knows it's not sufficient for success, but it is seen as a way of massively increasing your chances.
But is that really so? There is an advantage in being early, before a market becomes entrenched, but even in that respect Virgin Group have demonstrated that you can often find significant lucrative opportunities exactly in a market that has become entrenched because the competitive fervour of a growing market may have vaned and left you with a few large players that know eachother and know well enough that they have little to gain by upsetting the balance. Virgin upsets the balance, and their strength is being an unknown factor when they join a new market.
A problem with being the first mover is that you have to develop the market. You may strike lucky - Hotmail did, and managed to get a lead that made it incredibly hard for anyone to approach them in terms of size. However, for every company that gets lucky this way, there are thousands more that have "first mover advantage" in their field but never get far because the market is too hard to develop, and the cost for competitors to enter after you've spent massive amounts of resources developing the market is low.
My previous company is one example of a company fighting to grow in a market that's not yet developed to where we hoped when we started out. The company originated as a personalised webmail service called Nameplanet (the service is now owned by an unrelated company), and morphed into the registry for the .name TLD. The problem? The market for personalised domain names has so far been far smaller than everyone initially hoped. I still believed in the market for personalised addresses, but it's becoming clear that the market will require significant work before it's mature enough, and the cost of competing with alternative, but similar, products is extremely small.
I've lost faith in first mover advantage. Instead I prefer to think about second mover advantage as what you should aim for: Pick a field with one, or a small number, of underfunded companies burning cash to develop the market. Quietly prepare your product, and jump in when the pool is sufficiently full. You'll have the advantage of not having spent all your money building a market that is easily taken away.
The alternative is being a niche player or the unknown quantity. Enter an established market where the entrenched players keep doing the same boring things. Do something new. That's what Google has tried to do with Gmail.
One mistake: Make sure your main advantages aren't easily copyable. In Google's case it was storage - after their beta launch, all their competitors quickly scrambled to provide massive storage increases, and while most of them still don't match 1GB free, they offer hundreds of MB and larger premium products. Gmail does have an interface that is compelling for some, and by virtue of being Google they will gain a significant userbase, but for a smaller company a mistake like this would have been disastrous. (Btw. in case you don't know, I work for Yahoo! and while I don't speak for my employer, you might want to take that into account when assessing my biases...)
I learned that lesson the hard way with my first company - NorConnect - which I co-founded and was a director of. Our main mistake was to think that we'd make a big splash by undercutting all the main players in the Norwegian ISP market at the time. Instead we, and 2-3 other new ISP's, sparked a price war that got all the major, well funded, players to slash their prices overnight with the result that almost no ISP's made a profit on access for the next 4-5 years. In our case we sold the dial up for a pittance and went on to do consulting instead.
So what would be the best way if you, as a small player, want to enter the mail market just as Google did? Go for something that is directly at odds with the model of the large players. Something they CAN'T offer without massively changing their business models.
If I were to go back into the mail game, what I would have done was to revert to the original Nameplanet model: Offer a personalised service for a fee, and build it into a comprehensive messaging platform. This also happens to be close to the model of the company I joined after NorConnect - Telepost Inc. - a unified messaging provider.
Where I'd change things would be in offering a service that emphasise all those things you CAN'T offer in a free product: Extensive support, additional reliability and accessibility including extensive ACCESS to backup and failover facilities to increase the perceived enterprise readiness, any unified messaging services that would drive the cost of providing the service up above a level sustainable for free users, extensive access to all your data etc.
In other words: I'd aim for a service competing for a segment of the e-mail services users that they can't compete for with their free offerings, and where any attempt to go after the premium market would require a different product. At the same time, this is a market that IS developed, and where there are some players offering premium products, so competition would be fierce.
That's where second mover advantage comes in: Next time you see a Hotmail or Amazon rapidly rising, look for the segment they can't take and go for it. Let them build the market, and take the fringes. If they succeed, the market will get big and you can take a slice of it that'll give you a solid foundation. If they fail, they'll still have helped you by spending tons of cash building the market for you. If you're lucky you could corner the whole thing.
Almost the same potential, and significantly less risk.