Really? Going to extra mile to support your customers will give you a good reputation. It's nothing to do with playing the lottery. In this case it paid off in a big way. In the general case over the lifetime of your business it will almost certainly help you out somewhere.
I've just finished several freelance jobs where I've made sure to go above what was required. That's several people who will recommend me to others. That's not playing the lottery, it's just a sensible way of operating.
But the alternative possibility, when running a retail business, is that you blow 90% of your time being extra-nice to existing customers and never get around to building the features that would have attracted new users. It depends what stage of growth your business is at; the bigger you get, the unpleasant reality is that the less important any individual customer becomes.
I definitely fell into the latter category. Of the total time I spent doing customer service, there were about ten people I knew on a first-name basis. They had my personal e-mail address and I'd always respond to them within a half hour, often with a call back.
I never did have good tracking metrics in place for time spent per customer. I was a one-man shop and this was in 2000-2003, so there were limited open source options.
Customer support ate up a little more than half my time. This took away from programming and business development. Of the customer support time, maybe five or ten percent was spent on these ten customers -- and that is out of ten thousand paying customers and a few hundred thousand free users.
I never got a four million dollar phone call, but I also don't know how much these ten users evangelized my services. Who knows -- maybe they were responsible for a thousand paying customers?
There's an old phrase: "I know half of my ad dollars are wasted, but I don't know which half!" Nowadays it's a lot easier to measure return, but I still think we're not at the point where we can quantify good will from customer service.
There are definitely times when you need to make choices; you have a finite amount of time and energy and you can't put all of it into customer service. I too would suggest that spending all your time providing perfect customer service to a few customers is probably misplaced effort.
You can always make sure that all your interactions are as pleasant as possible. That will cost you very little time and will leave a lasting impression.
Today I had my first (and hopefully only) encounter with Dreamhost customer support. I was trialling a setup for a client who was looking to host 3 sites with them. By no means a big account, maybe not worth their time. Because of the way I was treated I will have nothing further to do with them as a company if I can possibly help it. I will never use them myself and I will never recommend that my clients do either. Had they taken the time to offer me courteous support I could have gone away happy and the entire process could have been completed faster. Sometimes you don't even need to be nice, you just need to show some manners.
For a given level of service, customer support costs a certain amount C * scaling costs S your predict return on investment is R / probability P.
If C * S > R / P, then don't play.
It wouldn't make sense, for instance, for a company like Google to treat all its customers the same way that enterprise SaaS companies treat their more precious customers. S is too big for them. That level of service is manpower intensive.
On the other hand it's probably something you can afford to at the beginning, when you have few customers - an advantage of being a startup, (as I recall being told a few years back, 'You are your own slave labour.')
I have found China to be paradoxically fantastic at customer service, especially for small clients. They're willing to put in the time and effort on relatively small companies because any income is good income, and building that relationship could mean good references later on with bigger contracts. I have only had two chinese factories turn me down for projects because I'm too small, and the ones that don't I will hopefully be able to award with increasingly large projects as I grow.
The lesson could easily be: I played the lotto and won big. Therefore, everyone should play the lotto.