twssgOne of the single most important decisions a startup’s founders can make is deciding which platform is going be their first. The era when Bill (Gates) and Steve (Jobs) defined a duopoly of two desktop platforms on which startups could build is well and truly over.

Today, SaaS, PaaS, Social, Mobile, Hybrid, Open Source CMS platforms have joined Desktop. How does a startup weigh and consider these platforms, what do they need to know about SaaS, what do industry moves like Google Android mean are the main topics of Chapter 3 of my new book, The Web Startup Success Guide seeks to explore.

<shameless plug>The Web Startup Success Guide will start shipping from Amazon July 22nd. If you’re planning to get my book, please pre-order now! Other booksellers – from the big chains to the independent booksellers – look closely at these pre-order numbers when deciding how many copies to stock. Your pre-order today means this book will be on their shelves.</shameless plug>

After beating up Google Maps yesterday, I’d like to share with you a pro-Google except:

“Loving Google App Engine”:

Dave Westwood is a cofounder of BuddyPoke (see Figure 3-3), a 3D avatar-poking app available on a variety OpenSocial networking sites * and to say he’s head over heels in love with Google App Engine is an understatement. Dave made this YouTube video (Buddypoke on Google App Engine) and bent my ear when I met him at one of Dave McCure’s Startup 2 Startup dinners (see Chapter 4).

Figure 3-3. BuddyPoke

Figure 3-3. BuddyPoke

When I asked David via e-mail about Google App Engine, here’s what he had to say.

Reasons we love App Engine:

  1. Scaling that “just works.” Many other web services provide a means to scale, but do not handle all of the scaling for you. With App Engine it just works, allowing us to concentrate on innovation and features for BuddyPoke, rather than the problems of scaling, backing up databases, etc.
  2. Scaling quickly. OpenSocial, the iPhone, Android devices, Facebook, etc., give developers access to over 800 million users around the world. These days, more so than in the past, if an app goes viral, traffic can increase by orders of magnitude beyond what you may have hoped for. On one day BuddyPoke experienced an eightfold increase in traffic, doing 260k new users in a single day. There’s simply no way we could have planned for that kind of rapid growth. With App Engine we didn’t have to.
  3. Quality and global reach of Google’s infrastructure. Ninety-five percent of BuddyPoke users live outside of the United States. Google, more than any other company, has created an infrastructure that can deliver apps to users on a global scale, with minimal latency.
  4. Pay-as-we-go model. Many social apps never take off. Many have instant overnight success and then fade away. Others have slower growth but manage to sustain a high active user count. It’s difficult to predict usage for any social apps. So the pay-as-you-go model is perfect for us.
  5. Cost-effective. We have a very good idea of what our hosting on App Engine will cost us, and it’s very fairly priced. The way I think about it is I just hired some of the most talented engineers in the world for free, and now I’m just paying them for the bandwidth and CPU usage.

A few years back one of the most popular phrases any startup would say is “We’ve solved scalability!” Quite frankly it’s a problem we don’t want to have to solve. We’re two guys with a background in 3D. The last thing we want to learn is how to scale. We’re in the business of making social apps. We’re happy to let Google take care of the scaling for us.

Google has been incredibly helpful in making sure we had enough App Engine quota to grow while App Engine was still in preview. And I’ve met many engineers from the Google OpenSocial and App Engine teams at local meetups, hackathons, etc. Google is really reaching out to developers even more so than in the past, with new developer relations groups, opening up their code labs, etc. They’re trying to make themselves accessible for support and help developers succeed. For example, we just had a “weekend apps” hackathon at Google two weekends ago. Google hosted the event and had some hands-on expertise from the OpenSocial and App Engine teams. So we did receive firsthand support, but I think that support is also available to anyone that asks.

I asked Dave my standard “Any advice for startups?” question: I think my response is much like any other startup before us,” Dave replied. “Just never give up. We released a product in 2007 and it was a failure. We learned from our mistakes, adapted, and tried again. Most entrepreneurs that have succeeded have looked failure in the eye at least once. Learn from it.”
===
* If you have no idea what an avatar-poking Open Social network app is, see the later section “Social Networks as Platforms.”
** One other quick note: these excerpts are from my last version pre-pdf and before Apress’ crack team of proofreaders caught the last of my typos. If you know a clean way of moving pdf to html on a Mac, please let me know!