ChangeLog

Rails Hosting: The Times They Are A-Changin'

Rails Hosting

Hosting Options Are Plentiful

It used to be difficult to find Rails hosting, oh how the times have changed. With the wide spread adoption of Rails came wide spread hosting. You've got cheap hosting at Godaddy mid-tier at Slicehost and high-end at Engine Yard with lots of options in-between.

For internal development we host on one of our dedicated servers using the standard Mongrel Clusters + Apache. This is an easy to configure and manage solution which has been very stable and headache free.

Clients who are being cost conscious and do not anticipate large spikes of traffic migrate over to a provider such as Slicehost. Slicehost has been pretty good to work with, however setting up a functional slice is a bit of work.

Engine Yard

Engine Yard is our choice for clients that are concerned with performance and scalability more than costs. So far my time with Engine Yard has been unlike any hosting I've experienced. I'm impressed with the "high-touch" level of service. They want to know your application, they write your deployments, even those tricky little custom tasks needed to put everything in the right spot. You tell them a bit about your app, give them svn read access a few bits of information and in return you get a ready to run Capistrano recipe.

I think my best EY story was when I was out in the middle of the woods, a client called and was concerned about their application, within minutes I was on the phone with an EY rep who hopped right on the matter, checking logs and load. Things cleared up and I received a follow-up email. I'd never experienced this before, heck I've never been able to get a live person on the phone before!

Not only is their support excellent but the infrastructure is quite impressive, have a look, if your new Facebook application becomes wildly successful overnight then EY can easily adjust your slice to handle the load.

My only two concerns with Engine Yard are pricing and can they scale their support? The pricing is kind of steep however it can essentially save you from hiring a systems admin. They offer such great support I wonder if they can continue that level of support as they add more clients?

Hosting Considerations

When choosing Rails hosting consider:

• What is the Budget?
• What amount of resources are required by the application?
• What is the expected traffic and growth potential?
• How are deployments are managed?
• Is there an upgrade path?
• How is the support?
• Are there backups?
• What is their uptime?

(NOTE: Another interesting hosting option that is still in beta is Heroku no setup required, hosted in the cloud, edit code in your browser.)

Social Marketing with GotVMail: Viral Video

This week we launched a Social Marketing campaign for GotVmail, a great group of folks that we've been working with through our agency partner RampDigital.  The goal of the campaign is to build awareness for GotVMail's "Virtual Phone System" among entrepreneurs online.

It centers on a set of video clips where Gary Busey gives his off-the-wall perspective on useful business advice, excuses, cliches (and pretty much anything else that pops into his head). 

The campaign itself includes a bunch of Social Marketing components, all aimed at maxmizing exposure and making it easy to share the videos:

• A shareable video widget that we customized with SplashCast (these guys are great by the way)
• A Facebook Page we set up to allow people to become fans, discuss and share the video
• A YouTube Page with all the videos for easy favoriting and external linking
• A Landing Page that links the campaign to the GotVMail site and brand

The campaign just launched yesterday, and already we've gotten thousands of views across Facebook, MySpace, YouTube and the video widget.

Congratulations to GotVMail, and to our partners RampDigital and SplashCast... and thanks for giving us the opportunity to put together the technology behind such a fun campaign!

Picking Your Widgets Wisely

Owing largely to the degree to which "widget" has become an over-used term in the industry, I talk to a lot of companies and potential clients who aren't sure what kind of "widget" they need (or if they need one at all).

As with all online marketing and application development efforts, realistic goals and success metrics are key... which are hard to come by if you're not sure what you're what/why you're building in the first place.

So, based on the work we've done over the past year at StepChange, I offer the following (slightly over-simplified) chart of the main types of widgets out there today, and what they do:

Having seen each of these types through their lifecycles of design, development, launch and adoption - I think the most important guidelines to follow are these:

• Match your widget investment to your strategy and budget. No one in their right might would spend $10,000 putting an RSS feed in a square box.  But $10,000 probably won't get you an entire affiliate network solution.  Make sure your business ROI lines up.


• Expect surprises, but not magic.  Distributing your content and engaging your audience will undoubted tell you a lot about your market and likely open up new business opportunities.  However... if you put your weekly 10%-off coupon in a widget, it's not likely that a million people will send it to their friends


• Think outside the mothership.  A shocking number of clients come to us for widgets that, ultimately, are focused on dragging users out of wherever they are and back to their current destination site. It works (sometimes), but the power and promise of widgets lies in taking the content/service/value to them... wherever they are.

Note: Cross-posted from our new Extend and Engage blog.

SightSpeed: Launching Video Chat in MySpace and Hi5

Over the past few months, we've had the pleasure of working with SightSpeed, a great team whose Video Chat service has recently been awarded a CODiE for "Best Communication Solution" and selected to power Dell's Video Chat.

SightSpeed has focused on Social Networks as a way to quickly build awareness and acquire new customers, and we have been working with them to build their "SightSpeed Light" application in MySpace and Hi5.

SightSpeed is a very high-performance platform, and this was some of the most complex work we've done in MySpace and Hi5 (Open Social) platforms.

Here you can see the two apps running in the MySpace and Hi5 platforms...

...same codebase, different wrappers and social implementations.

Some Learnings from working in MySpace & Hi5 vs. Facebook

• Neither platform is as "evolved" as Facebook.  I realize this is not news to anyone following the space... but it took really implementing a complex app to see the maturity difference in the platforms vs. Facebook.  Not to say that Facebook doesn't have growing pains, they're just more adolescent.


• ...Which has its benefits, too.  Especially in Hi5, we've seen the kind of viral application growth that made Facebook famous - before they changed all the rules.  For an app that's delivering a real, integrated service (not just throwing sheep), this is very encouraging to see.

Social Marketing - What's Worth Paying for NOW?

Whenever marketing technology is changing quickly - as it is now with Social Media and Social Networks - it becomes hard for a marketing department to know:

1. How much should XYZ cost now?
2. What's worth paying more for now, based on the differentiated business value it offers?

As with Websites in 1998, and E-commerce platforms in 2001/2002, "Social Networks" as marketing platforms are undergoing very rapid evolution in 2008... and these are the questions we are getting asked daily.

Thus... a chart!

Now, put all the predictable caveats around this (nothing's black-and-white, actual cost depends on specific features, some things are easier in Facebook than iGoogle, etc, etc.)  But in general, I find it useful to think about an emerging technology landscape in terms of the things that are:

• Commodity:  Table stakes.  Still valuable, but not differentiated.


• Standardized: These are things that have moved from "custom" into more "packaged" offerings.  Vendors can offer them more repeatably (so they're cheaper)...but that also means they're less differentiated and tend to be less integrated with your core business.


• Custom: This is where innovation is happens.  Companies leading a given market tend to run experiments and spend money wisely here.  Some Custom things eventually become "Standardized", but not all.

With that, I offer my take on business value and types of Social Marketing initiatives as of June 2008:

PS - our client work spans all of the above, so... you know where to find us!

Putting User-Generated Content to Work - Outside Your Site

Traditional "Destination Sites" are building up their User-Generated Content, now it's time to put that content to work driving awareness, acquisition and engagement - outside the main site.

User-Generated Content is Great on Your Site...

New tools are making it easier and easier for commerce, publishing and services sites to incorporate user-generated content - typically in the form of Rating, Reviews, Comments and Comparison data.

The rapid growth of the companies that provide these on-site tools, like our friends over at BazaarVoice, is testament to a simple fact: When it comes to driving conversion, user-generated content works very well.

...but it's Even Better off your Site.

But here's another important fact: User-Generated Content also works incredibly well off-site- by extending your reach and engaging customers in-context.

For our clients starting with a Destination Site and looking to extend their reach by leveraging their User-Generated Content, the first step is create a virtuous circle between their content and their audience:

One key getting this wheel "turning" is to embrace the various roles and participation levels of your target audience (numbers are different for every segment and context, but Comscore threw a 90/9/1% estimate for Watchers/Commenters/Creators that seems about right in my experience).

Another key is putting your content, especially your User-Generated Content (the "sizzle"), to work across your many Points of Engagement:

• If someone wants an RSS feed of your highest-rated products every week -- great
• If they want that in a Google Gadget they can stick on their home page, allowing you to push targeted offers to them- - even better
• And if you can package up some content that they're willing to share and discuss with their audience (creating even more content for you!) -- that's a home run

When it comes to leveraging User-Generated Content - there are lots of ways to win, but we've found that getting the big off-site "Extend & Engage" benefits begins with the wheel.

What's Next for the Big Online Players: Extend, Engage, Optimize.

Most of our clients aren't starting from scratch.

Sure, we enjoy working with startups on some 0-60, disrupt-the-world-in-4-weeks, rapid-fire development projects... but that's not how a lot of our projects work.

In fact, most of our clients have a substantial online presence already - with all the associated infrastructure, teams, development cycles and constraints.  Often, they're working with us because they want to extend their current systems to reach out and engage new audiences effectively... and don't want to re-invent the wheel to do it.

Which brings me to another picture...

This is the way our larger clients, who already have "Destination Websites", are looking at the opportunity (need!) to reach out and engage their customers more effectively, "on their turf".

Some clients are focused on a Facebook Application to start - or maybe some Widgets and Landing Pages - but ultimately those are part of a bigger Extend, Engage, Optimize vision.

Turning that vision into a marketing & technology roadmap for our clients has become a big focus of our work, and is guiding the evolution of our Distributed Marketing Application Platform. 

It's not an overnight process, and we're betting that for a lot of existing web companies, this evolution - and the technologies that enable it - will be a primary focus for at least the next couple of years.

You Know You Want It

We've got a couple of very, very cool Facebook applications going live this week, and I'm excited to introduce the first of those -- I Want, built on behalf of one of our clients, ThisNext. Check it out here: apps.facebook.com/wantitnow. I Want is a social shopping application. Users tell the world what they're looking for ("a non-dorky snowboarding helmet"; "sporty sunscreen"; etc.), and other users recommend products that fit the bill.

I Want is something relatively new for Facebook -- a true application with a purpose, not just another communication device or game. For ThisNext, Facebook is about much more than games. Like many successful destination web sites, ThisNext hopes to expand its footprint on the web and its touchpoints with customers into places where those touchpoints can be more contextual, relevant and personalized.

While it's fun to tell people we "build applications on Facebook," it's serious business, too. Facebook, and its compatriots in the OpenSocial Network, are building a parallel web that's based as much on community as content. Even after people get tired of "super-poking" each other, there will be money to be made and businesses to grow.

Exploring Silverlight Around the World

We are pleased to announce the launch of our latest Silverlight application on Silverlight.net. This app was created in conjunction with our partners over at Stimulant.

The application is integrated with the current Silverlight.net showcase and plots 181 applications across 39 countries on an interactive world map using geocoding from the Microsoft Live mapping API. Users can explore Silverlight applications by country and drill into a city level to view thumbnails and description of the applications. In addition, we've added in a data layer that uses Silverlight to visualize global internet metrics that show growth by country over time.

Check it out and let us know what you think...

More than Widgets: Distributed Marketing Applications

We've been creating Social Media Apps for quite a while here at StepChange, and have built all kinds of widgets for agencies and clients.

While we've done some basic Flash/Feed widgets, most of our design and development work has been on Social Media Apps that function more like true "applications" - with our clients requiring a relatively high degree of administration, content management, targeting reporting and integration.

I think these kind of 'super-widgets-turned-applications' need a better name, so I'm going to start calling them Distributed Marketing Applications.

For companies that have existing Websites or Online Applications (which is most of our clients), extending their current business into Social Media often requires a Hub and Spoke strategy (kudos to SexyWidget for nailing this one) and this type of Distributed Marketing Application.

A Platform for Distributed Marketing Applications

The system-level requirements coming from an agency or client who wants to extend their current site/service into Facebook/MySpace/Blogs/etc. are getting more and more consistent:

• An engaging User Front-end - which packages up some subset of their existing value and capabilities and presents in a way that's most useful/engaging and (hopefully) viral - based on the Social Media environment(s) they're targeting.
• A set of APIs to the "Mothership" - there has to be way to get data in and out of existing site... typically at least content & customer data, and often targeting/campaign information as well.  If the application is focused on new customer acquisition, it has to be very good getting that new info into the system of record.
• A set of Administration Tools - that allow the client or agency to manage content, users and campaigns - as well as see reports about activity and engagement.

...there are other pieces, but those are the main ones, and form the basis of the StepChange Platform for Distributed Marketing Applications - which we now use with nearly all of our clients:

Most of this is code we've created over the past year, plus some hooks into Widget distribution measurement partners.

The net result is that by leveraging this platform (starting with all the stuff in green, instead of from scratch) is that we can often put together relatively complex Distributed Marketing Apps in only a few weeks.

I'm also willing to bet that, as more and more companies look to extend their existing sites and services into Social Media, we're going to see a significant market need for these types of platforms.