Online Retailing - Product Promotions Using Facebook Open Graph API
It has been just over a year since Facebook launched its Open Graph API. It was touted as the next wave that would enable online retailers to target potential customers with personalized product promotions. The Open Graph API is yet another step from Facebook to challenge the stranglehold that Google has on online search.
The Open Graph API made its debut just when Google's search results started losing its sheen. Organized spam is already taking a toll on the quality of results one gets from Google. Added to that is the issue of syndicated content farms that forced Google to even declare a war against it. Given these struggles faced by Facebook's biggest and currently the dominant competitor, it is no surprise that the eCommerce community welcomed the Open Graph API with arms wide open. It was considered to be the messiah of creating personalized content on the web, the tool that everyone was waiting for eagerly.
One year hence, the question naturally arises. What has been the adoption of the Facebook Open Graph by online retailers? Has it become ubiquitous across all retailers? Though there is no official publication of such data from Facebook, there is enough anecdotal evidence to suggest that this is not happening. The Blind Five Year Old claims that only 27% of the Top-100 retailers have adopted the Open Graph API.
The Facebook 'Like' button - the social widget that is almost synonymous with the Open Graph API - aids product marketing by providing quick feedback on how many people liked the product. The first thing any campaign would be interested to know is whether the campaign was indeed successful. The following are some of the questions that they will need answers to:
- How can one ensure that the 'likes' are converted into product 'buys'?
- What are the ways in which an online retailer can make use of this data?
- What analytical tools are needed to make the above happen?
Likes to Buys
If you are familiar with online retailing, you may be aware of a user's tendency to like many products that he/she happens to see on the internet. Amazon was probably one of the pioneers to use this concept by creating their Wishlist feature. While the shopping cart aided in immediate sales, the Wishlist was used by Amazon in two ways:
- To remind the user during a later visit that they may be interested in buying a product that they were interested in before. Very often, users tend to forget what they had liked earlier. Unless reminded, this is as good as a missed sale opportunity.
- To recommend more products to the user based on what they bought and wished to buy during their previous visit. If you had noticed, the more you browse and buy at Amazon, the better they become at recommending products that you may be interested in.
Online retailers can use these 'likes' to build similar recommendations. Over a period of time, they will notice that they are getting better at predicting what a user is likely to like and actually likely to buy. As Darren Vengroff says on his Mashable.com blog, a teenager may like a Porsche-911 but that may most probably not translate into a buy. Therefore, organizations need to build a strong data models that can aid filter such noise and retain only data that may be most relevant to them. A simplistic approach to this will be to associate various attributes with the product being launched that helps in targeting the customer segment. Track only 'likes' of people who fall into this customer segment and focus on promoting the product or related products to only this segment.
Uses of the 'Likes' Data
The Facebook Open Graph API aids getting more data from a customer's 'like'. When a customer likes a product, it is quite likely that the person's friends and people in his/her network are likely to have similar preferences or be influenced by the customer's preferences. The Open Graph API can be used to post on the customer's Facebook Wall about his/her 'like'. This is likely to be picked up in the news feeds of the customer's friends and in turn their networks. There is a likelihood of this re-directing traffic towards the retailer's website. Additionally, when a customer visits the page where the product is promoted, the Open Graph API can be used to display information on how many of the customer's friends like the product being viewed. It is also possible to list and highlight their comments that were added when they visited the page last. These aid in building a common mindshare across a group of people who are already known to each other, thus raising the possibility of a buying decision.
The Facebook Insights is a platform that provides Facebook platform developers with metrics around their content. It provides a dashboard that displays these metrics. Alternatively, organizations can pull the metrics relevant to their business using the Open Graph API. By persisting with this data, the eCommerce platform can be designed to use this data to provide more refined and personalized content.
The other advantage of using the Facebook Open Graph API is the fact that you are likely to trace users liking the product outside of your website. The likes against links to the product posted on users' walls, the number of visits recorded from Facebook and other external websites are also indications of the product popularity.
Amazon, the pioneer in recommending related products, is tying up with Facebook now to leverage the 500+ million users on Facebook and their networks. It is only a matter of time when more and more online retailers jump on to this bandwagon. What is lacking today though is the availability of tools that aid in easy implementation of the Open Graph API as well as the limited analytics capability around the data that gets generated. With the increasing popularity of social community based shopping and with increasing adoption from leading online retailers, it is only a matter of time when such tools start mushrooming everywhere.
[Thanks to Saurangshu for providing me with the following tips]
1. The 'Likes' actually provide another opportunity to the retailers. They can use this data to push a sale or any discount scheme that they may offer for one of the items that is on a customer's Like-list. This is similar to what Amazon does to customers who have added one or more products to their Wishlist.
3. If a customer visits the eCommerce sites of two competing retailers who implement the Open Graph API, Facebook may be privy to a customer's visits on both sites opening up the possibility of data being shared with the competitor. This is definitely likely to be a huge contention before a retailer decides to implement this. Facebook may need address this issue to increase adoption across retailers.