Effectively and fairly quantifying UGC is challenging

At today’s “Isle of Man Super Third Thursday Social Media Club” (getting more of a mouthful every month!) it struck me that quantifying social media is very difficult. In order to assess the success or otherwise of a social media programme to management, you would hope to be able to point at real sales, web site visits or other conversions to justify the extra effort and time required to implement a social marketing programme well. Social media is just too fuzzy, however. While it can be gratifying to find the occasional positive comment, incoming link or your company mascot having its own facebook page, if a direct sale or income doesn’t result social media becomes just a hobby, as such less time will be spent on it and it will suffer.

Social Media covers a wide area of services, software and ideas. While we generally think of social media to be all about Twitter and Facebook, it is ultimetely about anything that encourages and supports User Generated Content. Many sites have been providing this before Facebook was a glint in Zuckerberg’s eye (or whoever claims to have written it this week). Sites which encourage users to rate their purchases, rentals, favourite films, artists, etc. are all essentially social-media services. Amazon, of course, is probably one of the oldest examples – particularly in the e-Commerce space. It has allowed users to rate their purchases (or just products if they purchased the product elsewhere), the performance of their Marketplace sellers and even generate customised lists of products and share with other users. eBay offers a similar rating system to try and reduce the risk of making purchases and selling your prize possessions. While Amazon’s scheme seems to be pretty simple in that it allows users to rate something out of 5 stars and leave a comment, it isn’t always that simple.

EBay have recently modified how their ratings system works. Whereas previously both the buyer and the seller could rate each other during their transaction either positively or negatively, this has now been reduced to only allowing the buyer to rate the seller. This obviously causes sellers to worry about no longer being able to help their fellow sellers avoid a potential difficult customer. But it has come after a number of users complained about having unfair ratings applied to them, and the system being mis-used. I for one would have loved to leave a negative mark on one seller who sold me fake Twin Peaks DVDs. After politely disputing the genuineness of the DVDs, the seller withdrew the auction and left me with no recourse against him to advise others not to take the risk or believe the “100% genuine” claim.

This highlights that allowing users to make their own judgements on service, content or products is fraught with difficulties, particularly when using a discrete rating system like a stars system or a positive/negative mark. A soon as you start to be able to quantify user-generated feedback, you enter a dangerous area. During the Work Connexions project, one of the biggest challenges was to be able to accurately and diligently identify a quality lead, individual or service but without providing the opportunity for any overtly negativity to be attached to the content. We had to be able to provide a means of rating other users, but as soon as you offer users the ability to leave negative feedback which could directly impact their performance to generate further leads by affecting their ranking in search results, visibility on home pages, etc., you run the risk of losing users who may feel that they have been slighted or unfairly rated. Equally, when you remove the possibility to leave a negative mark against content, you may be accused of being weak and a toothless “quality mark”.

The Manx Graduates web site, which was discussed at today’s Super Third Thursday is a case in point and faces similar challenges. While the site is being placed in the social media space, it is with a degree of required control, which is somewhat against the principles of social media. The site is designed to make it easier for graduates who have left the island to study to return and gain employment on their home island. There is no real target age-range or skill-set. The common theme is essentially enticing talent back to the island and avoiding the “brain drain” that inevitably occurs on an island such as The Isle of Man with its limited opportunities and education facilities. Users can contact potential employers, and vice versa. Users can also message each other, assuming they already know other users on the site, it isn’t after all, a social networking application. So the “real” social media-style tools commonly found aren’t to be found. In particular, there is no opportunity for users to be able to recommend or otherwise working for a particular employer. Again, there is the problem that as soon as you provide users the power to submit their own content, they could use it to attach negative feedback to other users, subscribers and stakeholders of the site. This could be particularly damaging if a quantifiable rating system was applied to an employer or potential employee.

There are ways to try and control how to limit the possibility of negative feedback – or even overtly positive feedback – which could impact on quantifiable scoring systems or constitute an attack on another user or service. They often come with their own difficulties, however. Moderation of content reduces the spontaneouty of content submission, causing delays before content is published. The user may not trust or agree with the decision of the moderator. Scoring maybe limited to certain ranges after certain mebership qualifications have been met, for example, disable low scores until a user has been a consistent and repeated user of the site. This reduces the opportunity for a “true and accurate” account to be provided by the user. There is no real defence against a malicious attack against a user or service, if someone wanted to create a negative response it is almost impossible to prevent without withdrawing the life-blood of the social media application. The trick comes in being able to give the illusion that user generated content is freely submissable but exercising tight control over user content, essentially a thankless and full-time job.