Everyone of my web launches has been fraught with disaster — not that my experience launching things on the web is so deep and wide, but here I am on my fourth web-based storytelling project and fifth website and I see a pattern:

Launch is hard.

For the launch of Ruby Skye P.I. on Monday, I made some mistakes.  And then there were the usual unforeseeable obstacles: inaccessible passwords, crashing servers, FTP issues.  Throw in the unpredictability of how long it take the DNS to do it’s thing and the fact that I’m tossing around terms like DNS and FTP and you’ve got a big fat mess at 3 a.m.

On Monday, there was a frantic period when we were madly directing people to watch Chapter 1 on Vimeo.

Then the site went live, I blurted out the great news to Twitter just in time for the server to crash.  Twice.

I’m not a quick study

That’s about when I decided to take my furiousity walk.

In the end, we got rubyskyepi.com online four hours later than we’d hope.  It was a hairy day.

Next morning, I made myself a list of rules for launching new web endeavours.  When next time rolls around, please remind me.

• Launch is the coming together of three forces: the site, the host and the content.  Content is all the posts, pictures, video and games.  The site is the design and layout which holds the content.  The host is the server that the site and it’s content live on.  Next time, I’m going to make sure everyone understands the process from everyone else’s perspective.
• Public launch on the same day the site goes live?  Never again.   You kind of want to because it’s all new and shiny and you just want to go ta da.   With Ruby, we’ve had our production blog sitting at rubyskyepi.com since July.  We’ve been getting great numbers.  I thought that as soon as we transferred over to the new site, we had to announce it.  Really, why?  So what if a bunch of regular readers came and played around and enjoyed the space before we yelled about it from the roof tops.  It was just lack of imagination that prevented me from seeing it that way.  It would have been better to quietly make the change and have some time to tweak and fiddle.  Some people would have found it and taken a look around, but that would have been fine.  And we could have invited the world in when we had everything fluffed and shined to gleaming.
• Launching at a specific time is a great way to drive numbers.  But you just can’t do it with a website.  You can do it with a video but it’s just stupid to even attempt to get your site up at an exact hour.  You upload the web design to the host site and then you set the DNS to the new address.  How long the DNS takes to switch over is completely unpredictable.  They say up to 24 hours but in my experience it’s usually more like a couple.  (Please excuse me if anything I said in this paragraph is wrong, I don’t know what any of those words mean and my understanding of all of this is sketchy at best.)
• Do not launch on a Monday (which I wouldn’t have done, had I spent more time talking to my developer who knows things I don’t even know that I don’t know).  That just means being up all night Sunday when you can’t reach the one person in the world who has the key piece of information that you can’t launch without.  Mondays are good for traffic, but you’ll have good Monday traffic next week and the week after.  Launch mid-week and take the weekend before to sleep.  Because once your site’s live, you’ll never want to leave Google Analytic’s side again.