Google announced last Wednesday that they would discontinue Google Wave:
But despite these wins, and numerous loyal fans, Wave has not seen the user adoption we would have liked. We don’t plan to continue developing Wave as a standalone product, but we will maintain the site at least through the end of the year and extend the technology for use in other Google projects.
Google Wave is a web application that enables rich communication between Google Wave users. Rich meant a lot of things, such as sending as sending specific widgets directly (instead of sending a link to Google Maps or Doodle by e-mail, you could embed them), or moving away from the «every message is standalone» approach of e-mail and keeping track of conversations (sequences of messages) instead.
I remember Wave for the astronomical amounts of hype that surrounded it in the early days. It was said to be revolutionary. You had to be invited to use it, and there were only so many invitations around. And regardless of the apparent lack of interest, Google Wave is an impressive and exciting technical achievement:
Last year at Google I/O, when we launched our developer preview of Google Wave, a web app for real time communication and collaboration, it set a high bar for what was possible in a web browser. We showed character-by-character live typing, and the ability to drag-and-drop files from the desktop, even “playback” the history of changes—all within a browser. Developers in the audience stood and cheered. Some even waved their laptops.
Yet, Google Wave was designed as an application for communication and collaboration. Such applications, by design, need several people to be using it together. It’s a classic chicken-and-egg situation: I won’t use the system if I have no one to talk to, and I won’t have anyone to talk to if other people don’t use the system. And Google Wave made two mistakes here which prevented it from reaching the expected critical mass.
This morning, I turned my computer on. Within seconds, Skype and MSN Windows Live Messenger launched as well. These bloody parasites increase my boot time. But they’re doing the Right Thing: this means every time my computer is on, they will be running. Let me rephrase that: every time my friends’ computers are on, they will be running. This increases the probability that my friends will appear online for me to talk to them, which in turn increases the usefulness of the software.
It means that if Alice asks Bob to install Skype so she can send him a large file (or view his screen for some tech support, or whatever), then Bob will be connected to Skype all the time, even though he does not believe it to be useful.
On the internet, there’s no special “Launch when Windows starts” flag you can use to keep your users connected to your web site all the time. You have to rely on your users to either come back to the web site of their own accord every day, or set up your web site as their home page. Bob will only be connected on Google Wave if he genuinely believes it to be useful. This is an extremely important difference, and it was Google’s first mistake.
Rule 1. To achieve a critical mass for an application that revolves around user collaboration, you need to attach it to another application that is used on an extremely regular basis by the users.
Skype attaches itself to Windows. Facebook chat attaches itself to Facebook. Gmail chat attaches itself to Gmail. Google Wave attaches itself to… nothing.
Then, there was the other issue. Google decided that Wave would be initially available only by direct invitation, and every Wave user would only get a limited number of invitations. I think there were only eight. How could this have gone wrong? The reduced availability combined with the persistent Buzz made everyone run around in circles looking for invitations! Contests were held to win Wave invitations [fr]! Instead of thinking «oh, I’ll try it someday» everyone thought «I have to get my hands on an invitation now» !
I managed to get my hands on a Google Wave account when someone on my alumni network newsgroup offered them to anyone interested. I signed up, and was faced with a mostly empty screen. My contact list was empty, save for the person who had invited me (whom I knew in a «keep in touch» way, but not enough to warrant using Google Wave). I toyed around a bit with the interface, failed to get the point of it all, and never came back.
For someone in the web analytics business, Google made a pretty surprising error: they went for a high quantity of eyeballs, instead of a high quality. They attracted people through curiosity and the illusion of a shortage, instead of attracting people that wanted to communicate with existing users.
Rule 2. If your software relies on people using it together, then you should target non-users that might want to use your software with existing users.
The invitation model worked for Gmail because you can use it to send mail to any person on earth with an e-mail address. It worked for the early Facebook because of the privacy implications (you don’t want to invite someone unless you are comfortable with sharing your personal information with them) and because the friends of your friends are usually your friends as well (so you fill your contact list very fast). But with Google Wave, you could get yourself invited by anyone, and you would end up with an empty contact list and an useless application.
To quote Seth Godin, Ideas that spread, win. And here is a Godin-ly good hour-long video of his talk at Business of Software 2008:
Related Posts
- Understanding Social Media – how users interact online.
- 1. Find Crowd – 2. ??? – 3. Profit! – on leaving a good first impression.
- Eternal Sunshine on the Spotless Reputation – how to handle a bad first impression.
Hi. I'm Victor Nicollet,
“Oh my God! They killed Wavey!”
I agree with your points and want to add that when launched Wave was slow and sometimes buggy, so even if you did have friends online it wasn’t a good experience to chat there. Most of my geeky friends had invitations (me too…) but we ended up just giving Wave a short try. As you mentioned first experience matters greatly.
I recently I worked on a project with 3 other guys, and we were geographically far so I figured its a good time to give Wave a second chance and try the collaboration aspect of Wave.
Sure, we could group chat in skype or gtalk or other, but Wave offered a bit more:
- One could see what others write before they press enter.
- Wave extensions such as mindmap and a Diagram thingy allowed brain storming together.
- The forum style (threads) of replies meant conversations can split and yet maintain order.
- You could play back the conversation if you want.
But honestly seeing what others type before they press enter isn’t so great. I for one stop to think as I write, and often fix some wording before submitting.
The extensions are a super idea but when I tried them they weren’t very stable so that wasn’t such a good experience.
Now in Wave there is reply, and indented reply and inline reply and continue thread. Using the mouse to pick your reply style meant slow and uncomfortable handling. Yes there are keyboard shortcuts and yes I guess if you spend time you can learn to reply in orderly way, but it was too easy messing up and indenting needlessly or not indenting when it should have.
Again this may be our inexperience with Wave, but we mostly just acted like a regular group chat. And when group chatting the best GUI is just wide flat vertical list of replies, compact but readable like good old IRC. Wave has beautiful GUI, but its just not very comfortable to group chat. The rows are too narrow (because actual “Wave” takes just part of the screen) and the padding and margin between replies are too big so even a few replies become very high stack. It makes me think that oddly enough the Wave GUI is more suitable for asynchronous chatting like email or forum.
I am at mixed feelings about the outcome of the experiment. It wasn’t so awful that we stopped using it. In fact we had several Wave sessions and managed to talk and “collaborate”. But it wasn’t a great experience either, far from it.
The really surprising thing about Google’s blunders with Wave is that they were all set up to get both of your rules right. All they needed to do was integrate Wave in with Gmail.
People already check Gmail every day. If Wave was built in, they would be checking that by default too. I know a lot of people like you that never came back to Wave after the first experience, because they were never *reminded* too.
Similarly, Wave could have acted as an email client. If I want to send an email to someone, I just make a wave and send it to their email address. To those people, it’s an email. To me, it’s a Wave. This seems like it would have avoided the whole “what the hell do I do with this?” question that everyone was asking about Wave, and it would have allowed new users to slowly join.
I personally love Wave. It has a lot of problems, but there’s no better way to quickly share website designs with other people and collect feedback. I’ve only had a few conversations on Wave that really worked well, but when it did work, it was a thing of beauty. I’m sad to see Google giving up on it.