A few days ago I posed about the trend I noticed at Google of using APIs as a way to release new products — the API-as-Product-Launch I have started calling it.
As I wrote in the previous post, I think that the API-as-Product-Launch is a very promising strategy for tech companies to look into. It allows for the launch of product ideas faster than the “Agile” model and with very minimal infrastructure.
Since this seems so promising to me, I started thinking about what it would take to succeed in this new model. Here is my (incredibly) quick shot at the requirements to kick off the discussion:
- Buzz worthy service API — preferably with lots of data. Make sure that developers can play with and push data into the API at the time of launch. An invite-only beta to build buzz and get the bugs out can’t hurt.
- Launch video of the API developers talking about what it is and why they are excited about it.
- A great story and vision for the future using the new technology.
- A pretty logo, stylish diagrams and content-as-packaged-meme for people to re-blog. This is essential to make sure everyone gets the message about why your service is genius in just the right way.
- Dead simple API test console so biz dev guys can play with the new API to understand what it is all about and feel involved enough to talk about it at parties.
- Follow-up the launch video with video screencasts showing all the cool things that you can do and how to do them. Bonus points of the community does their own screencasts.
- Double geek-bonus points if you can get mentioned in the same sentence as (or are one of) the following
peoplenew geekerati:- Brad Fitzpatrick
- David Recordon
- David Heinemeier Hansson
- Chris Messina
- Chris Goffinet
- Joseph Smarr
- And dozens of other new elite geeks that the public-at-large has never heard of and for some unknown reason I am forgetting right now…
Chris Saad wants to be on the list above really bad; but needs to stop asking people how their working applications fit into his theory of Data Portability. (Just kidding man… keep herding cats.)
What (or who) would you add to the list above?
Subscribe to the feed
5 responses so far ↓
1 Jon Henshaw // Feb 7, 2008 at 11:07 pm
I like the idea of offering developers and incentive to get involved. That could be in the form of a contest, publicity or whatever.
2 Chris Saad // Feb 15, 2008 at 11:16 pm
Wow Todd - I got a whole paragraph about me - why would I want to be in the bulleted list?
Well done on your API work - loving it.
3 Todd Sampson // Feb 16, 2008 at 1:30 am
Hey Chris… Glad you liked it.
As I was reminded in a meeting with some guys at Digg today, it is easy to get cynical about the ability to get data portability working. But it is really important for users and businesses alike. And as such, it is worth the fight.
So again, keep trying to heard the cats. I really hope you can pull it off.
4 Ben Metcalfe // Feb 16, 2008 at 11:01 am
Clearly there’s a tongue-in-cheek tone to your post, but launching a product with a strong API from the beginning is very important - and I agree with most of the points you raise as being important factors into that.
Personally, something like the Google Social Graph API is not _product_ per se, but a service. Products are used by (normal) people, and the API on its own holds very little value unless it’s integrated into something that the end customer can use.
To me, launching an API with your new product not only gives it geek cred and allows new kinds of sideways usage of your product to occur, but also shows you (probably/should have) built the product on top of the API anyway…. and API driven application development is where it is at.
5 Todd Sampson // Feb 16, 2008 at 5:02 pm
Thanks for the comment Ben. There is a small amount of tongue-in-cheek in the post, but mainly just the last bullet point and following paragraph.
I do feel the API launch is a smart way to start a service or even, in some cases, a full product. I believe that we will see more of these types of releases very soon.
- Todd
Leave a Comment