≡ Menu
Pawel Brodzinski on Software Project Management

Release Later

Release Later post image

Release early, release often they said. What a great idea I thought. You get an early feedback, you check whether the direction you’re heading to is good and you can quickly adjust your course. And then you repeat. At least that’s the theory.

An advice I have for many of you today is slightly different. Release later.

Yes, you’ve heard me. Later. Actually quite the opposite to release early. Haven’t I just mentioned I thought it was a great idea? Um… Yes, I have. Am I trying to confuse you? No, I’m not. So… what the hell am I talking about?

I’m talking about a few apps I have an occasion to work with recently. All of them have few common things:

  • They all are web-based. No surprise, eh?
  • They all follow release early, release often principle.
  • If I had to write a review for each it’ll look like “A piece of crap. Keep away from that shit at all cost.” It would even suit Twitter when it comes to review length.

The main problem is they were all released too early. Somewhere in pre-alpha stage. You know, barely anything works and half of crucial features are not even there.

What’s the problem you ask? Well, when eager crowd sees someone (author) is all “can’t wait to show you my app (My App actually), you should feel bad you haven’t checked it yet” and then they see something which makes them cry in a minute they’re walking away and they aren’t coming back.

Or, if they’re like me, they dig deeper just to find the app is not only barely tested and far from completeness but it also have strange foundation architecture-wise and/or functional-wise. I guess “strange” isn’t an epithet I’d like to hear related to the base of my software. If your earliest users are up to this point they just won’t come back for anything else but writing a rant about this crappy app they’ve seen the other day.

Hold your horses then. Find someone who will give you their honest opinion about the app being ready or not ready to ship. If it isn’t don’t take release early principle as an excuse. It won’t do any good. Usually releasing too early is even worse than releasing too late.

http://www.limitedwipsociety.org/2009/10/12/kanban-at-lonely-planet/
in: project management, software business, user experience

4 comments… add one

  • Phil Ruse January 7, 2010, 11:05 am

    You weren’t thinking of Google Wave by any chance ;) I have a feeling they’re trying something completely different – a new way of developing if you like – but I have to confess if it’s too buggy the first time I look at something then it takes a herculean effort to take a look at a ‘next’ version. First impressions count – something my Grandmother use to tell me!

  • Pawel Brodzinski January 7, 2010, 11:17 am

    Actually I was thinking about Google Wave too. And Wave being shipped too early in not only my impression. However I agree Google case is different. They can allow themselves to ship too early because the buzz will be so big people will come anyway.

    But you can’t do the same if you don’t have big name. And this was the case of the rest of examples. Startups trying to get the word out about their product. They won’t get the word they want for some time I guess.

  • Josh Milane January 10, 2010, 8:22 pm

    Iterations? Anyone?

  • Pawel Brodzinski January 11, 2010, 1:16 am

    Josh,

    Iterations – yes. But the question here is when you should go live for the first time with your app, namely after n iterations or maybe after n+3 iterations etc.

Leave a Comment