Social Shopping and The Game of Life

Most of the current social shopping ideas are lame. They are mostly derivations of either virally allowing users to put links, or very simple widgets, on social sites, or they are copycat social networking sites with some basic ecommerce built in. These ideas are laudable as at least they are attempting new ecommerce solutions, but they feel like mashups of ecommerce and social networking, hoping that 1 and 1 will be 2. The social shopping roadmap can start with these things if you wish, they will not take long to implement. Actually, just use addthis, I'll give you the link right here. It should only take 20 minutes to implement and then you will be caught up. Go do this and then come back and finish the rest of the blog. I will insert a smiley face so you remember where you left off. :-)

Now you have 90% of most of the current social ecommerce capabilities - let's move on.

The information architects are experts at understanding how an individual interacts with a computer. The information architect worries about how a user uses a program, how easily information is found, how the user feels when using a program. Information architecture is a mature discipline. People can get a degree in the discipline from the University of Baltimore or Kent State University or the University of Michigan or Bentley college or perhaps a masters from the Illinois Institute of Technology. There are dozens of options. Information Architecture is mature, and it is this discipline should hold the answers to social shopping. If the ecommerce information architect 1.0 understands how a customer interacts with a site, then the information architect 2.0 should understand how a group of customers interact with a social site. But the current social shopping ideas are best guesses at functionality, undisciplined. An information architect will tell you that all features of a web site should be directly related to the desired user behavior - form follows function. This same perspective should be applied to social shopping. First what is the desired behavior of the group, and then all social features should be matched and measured against this desired behavior. Retailers are masters at understanding and guiding group behavior for in-store shopping, but their on-line counterparts are not so in-tune.

Done right, social shopping has tremendous monetization. Think about the wisdom of crowds. Amazon built an empire on this social personalization technique. What is the behavior of crowds in a social shopping setting? Some of the principals of social networks seem to be:

• You are left alone and are anonymous to most people in the same social site, but you know you are not alone. This creates buzz, but not annoyance
• A small percentage of people author social content, most of the rest are consumers. For more on this topic take a look at Forrester’s Social Technographics report
• If the social network is not transparent and open, people will reject it
• Any feeling of belonging or exclusivity is a good thing
• On-line social activity is a generational phenomenon. People who grew up with the internet value their an on-line identity, people who have not grown up with the Internet fear an on-line identity - what is a better definition of a generation then enthusiasm from one generation and fear, uncertainty and ambivalence from the previous generation

Once these and other core social shopping behaviors are understood, then all site features should map directly to them. Today a retailer is not going to add functionality that does not directly drive conversion, decrease abandonment, increase traffic or define brand. So why add social capabilities that does not monetize or directly map to pre-defined social behavior metrics? The value of the network must be derived to appropriately prioritize features. Social metrics should be defined and measured such as the:

• Velocity of social network growth
• Value of customers measured by how they help grow the network
• Network conversion rate defined as how many people invited ultimately participate and buy.

These values are measurable, and all A/B testing should drive defined value such as these metrics. The existing analytics engines do not measure this social dimension, so features are being added and measured in an old and socially inaccurate way. Similarly usability testing tests how an individual interacts with a computer, it does not simulate a social experience. Using these 1.0 filters blur 2.0 capabilities. You can't evaluate 2.0 features using 1.0 tools. Social shopping is a new dimension - a dimension 1.0 tools do not recognize, measure or appreciate. Using 1.0 ecommerce information architecture and product management tools for a 2.0 social shopping site will create a site that is suboptimal in both ecommerce and social capabilities. 1 plus 1 will equal .5.

Measuring social success and predicting social experiences is not science fiction. Social behavior is predictable. Do you remember the game of life? Not the board game, the algorithm created by John Conway in 1970 that describes how dynamic systems interact. It is a very simple model, the rules are:

• Each cell with one or no neighbors dies, as if by loneliness.
• Each cell with four or more neighbors dies, as if by overpopulation.
• Each cell with two or three neighbors survives.
• Each cell with three neighbors becomes populated.

Here is an example of the game.

The game is a simple model that describes how dynamic systems interact, like a population, or a virus, or a social graph. A social graph is a term that describes the network of connections that exist through which people communicate and share information.

A small social graph that feels lonely to its participants will wither. A social graph that is over-populated will also wither because it is not exclusive or personal enough for its participants. Healthy social graphs grow. But to a retailer, this is still not enough - a retailer needs a healthy network that will also buy.

Imagine an analytics engine that looks like the game of life. It would visualize the value of the network by showing the network graph and highlighting ‘valuable' that are either helping to grow the network or are buying. For example, just because the network grows does not mean that is valuable because new members may not purchase, like virtual loitering, or registered members may not log in. A truly vuable customer is not only loyal for their purchases, but they also are evangelists that help to grow the network. We need analytical tools to show this social network value. For example image an analytics tool that visualizes social graphs analytics that looks like the following diagram:

This sample social network is very simple. Here adjacent cells represent user relationships known by the retailer. The highlighted cells are ‘active' customers, where active is either being social, which the red cells represent by inviting or communicating with others or have done some other type of social activity such as ‘digg'ing or 'del.icio.us'ing a product. Green cells represent users that are buying. The graph can be filtered by time, or average order size, or most growth, or most withering, or whatever. Most ecommerce sites already have the data, we just need the analytics engines to show the social activity in a valuable way and understand the rules that social behavior follows. The graph could also be played forward a few steps to predict future value, and actual results measured against it. Then any A/B test can be predicted and measured against 2.0 metrics. The information architects and product managers could then predict and measure the success of their designs not only against traffic, AOS and conversion, but also against defined social metrics. Without this perspective, we are just social guessing, and function is following form, and what could be more backwards than this? And as an aside, we should stop all the talk about Web 3.0 until the 2.0 ideas are well understood, commoditized, and "Social Information Architecture" has become standard information architecture curriculum - which is obviously still a ways away.