I think it's a nice trick, and I will definitely apply it myself once we're ready to roll out. I do have one comment though (similar to redguava's).
Emailing the first 1000 users, fine. But doing it personally and manually, why? Aren't you going to end up using exactly the same text for all your users (either that or risk typos that make you look less professional, not more human)? Don't forget the risk of developing CTS while you could be doing much more constructive stuff.
I would say, have it automated but sent a couple of hours after signup, or have them sent in batches at a time where you would normally sit down and do it. I'm thinking that people either assume it's automated or they assume it's manual, one way or the other.
This automated approach is fine, but only if you don't try to pretend it's personal. SlideRocket does this really well - sign up for SlideRocket and watch the helpful emails roll in. Not too many, and always with a good tone and good content. But none of them pretend to be produced by hand in Gmail.
Sending personal emails to the first 1,000 customers is great, and switching to automated lead nurturing after that is great. Partly for scale - there are a hundred high-value things a CEO can do once a business gets going. But also because customers don't think it's appropriate to get a personal email once you grow to a certain size.
Anecdote: I've been heavily involved in our sales process at Zencoder since the beginning (as CEO), including writing personal emails to 500-1000 signups early on. (We've stopped that by now, but it was the right thing to do early on.) A few months ago, our sales director and I went to meet with a high-value prospect. At that point, their financial value wasn't huge (hundreds per month), but we knew they would grow, and that they'd be a good reference customer. So we pitched them and everything went well.
I ended up meeting with them a second time, and they asked me point-blank: "How is your business viable, if we're only paying you $X00/month?"
The subtext was this: "If we're only paying you $X00/month, and we get the CEO in for two in-person meetings, then you must be small, and we must be a critical customer for you to win." Customers should feel important, but never critical.
This came to a head when we sent them a final proposal. They asked for a price break and concessions, and said explicitly that because they were an important customer for us, we'd give them these things concessions. In reality, they were no more important than 20 other prospects we were working with. But because we brought our CEO into sales meetings on a small deal, they thought we were smaller than we were.
Its not an engineering problem, its a customer service, business and relationship problem. I contend that the more you humanize these types of interactions, the better your results will be, the more you will learn. Look at it this way, 30 seconds to send each email works out to a bit more than 8 hours work to get some amazing customer feedback. Some organizations will spend this much time in planning meetings just to setup a discussion to strategize customer sat approaches. You can get the feedback for almost free. For me, its an easy choice! ;)
But, again, what do you say in an email that is personalized? Do you switch sentences around at random, change "hi" for "hey", forget a comma, replace a dot by an exclamation mark?
Responding to "it's a waste of my time to write the same email a thousand times" with "you should reply to each email personally" is moving the goalpost. The OP is about sending a first welcome email to every user that signs up, not about replying to an email sent for customer service or technical support.
I would never reply to a client using a canned response, I hate those with a passion. But the first email being sent, the client didn't send anything, doesn't know you or your writing style, didn't offer you any chance of customizing because it's just a "welcome to our service, thanks for signing up, I'm there if you need me" email. My point is, write it as if you would send it personally, but automate the actual sending.
In theory, I think your approach is fine, but in practice, there will always be opportunity to personalize. If you are really motivated, you can google the person and see if something interesting comes up, or you might already know the person tangentially from a conference, or there was an issue with the order that your team had to fix (a great opportunity to reconnect with a potentially unhappy customer). etc. Sure, a number of them might be exactly the same, but many of them won't be if you commit to the exercise and look for opportunities to form a relationship with these clients.
I think I'm starting to see the value, especially reading other comments here. Seeing an email that I recognize may be a great opportunity for a personal hello. I think perhaps one proposed solution would work best - having a list of new users that haven't received the welcome email, see their information, and then decide to send the regular message or edit it before sending. I think that's the way I may be heading, it sounds balanced and right.
Emailing the first 1000 users, fine. But doing it personally and manually, why? Aren't you going to end up using exactly the same text for all your users (either that or risk typos that make you look less professional, not more human)? Don't forget the risk of developing CTS while you could be doing much more constructive stuff.
I would say, have it automated but sent a couple of hours after signup, or have them sent in batches at a time where you would normally sit down and do it. I'm thinking that people either assume it's automated or they assume it's manual, one way or the other.