Justin Hayward`s SI Project

Random Timberlake...

Another post by way of Seamus McCauley for no particular reason except for the fact that he has summarised some of the postings regarding the NYT article by Duncan Watts on cumulative advantage.  What is measured, matters.  What matters is measured.  Just because there's not enough money in the world nor the ability to get an honest answer as to why people believe what they do consciously or subconsciously does not mean that it can't be measured and that there wasn't a reason why people changed their minds, either singly or collectively. 

Is life art on the underground...?

Seamus McCauley makes an excellent point of example about context and content and the value attached to the optimum offering causing people to pay a premium for it, but it could equally be taken in a different way.  It reinforces what is fashionable amongst different groups and what is out of fashion in one environment may well be vogue within another.  It's about picking the right contextual background in which to make the most valid contribution and at the same time understanding that people within that environment will take a different opinion or experience away.  It's about picking the right networks for the audience you are looking to connect with.  It's not about money, or commoditisation, it's about art and relevance.

X marks the spot...

Aha!  John Bell, an ex-colleague at Ogilvy Public Relations and author of Digital Influence Mapping Project blog, focuses on measurement here this month.  I was wondering when he would get back to measurement.  He flags up Compete's Attention 200 sites and rightly states that all they're trying to do is own the issue rather than define something.  I'll be keeping an eye on John's site more closely this month - if anyone's smart enough to work with the right people to define things, it's John.

Resources...

There's a very matter of fact list of social influence theories here as well as a number of network mapping technologies

Compete-ly fabulous...

I hadn't heard of Compete before this post on Mashable!, an excellent site to keep up to date with the crazy pace of social networking sites.  Anyway, Compete is moving away from traditional page view metrics and delivering two new free measures.  In the words of Mashable!:

"Much like Nielsen Ratings, Compete’s Attention metric will now measure the amount of time a visitor spends on a site as a percentage of the total amount of time spent online by all U.S. internet users.

Their second metric will show a site’s velocity, and is similar to AttentionMeter as a comparison tool. This is their relative change in daily Attention, and will show growth in comparison to other U.S. sites. It’s important to note that Compete’s new metrics are relevant regardless of technologies such as Flash and AJAX, curtailing some issues raised with simply counting the number of times a page has been viewed."

I hope it's not just U.S. or that someone will help them expand to other territories.  All of this kind of stuff is useful to drive information into the pipe for defining social influene and that relies on knowing what can be measured.  These two look like a dramatic source of quality information if they work but there are a whole tonne of sites outside of the US which seem to be influential.

Mashable! also notes that ComScore has filed for an IPO today.

Making a list, checking it twice...

Kami Huyse has an interesting list of measures for working out the social influence of a site, but also makes the point that different measures are required for different delivery formats - one set for text, another for audio/podcasts, etc.  Thinking about this also leads to the conclusion that a weighting should perhaps be considered for different vehicles of delivery across media platforms.  We know that many more people read blogs than download podcasts at present if the industry is to be believed, but do many more people watch video than read blogs?  There's been recent findings showing that video adverts online are more sticky than on television. 

The definition of social influence is not a simple one, but it can be worked out.  What matters should be measured and all that...

Network Sense and sensibility...

It's good to see that Spannerworks has announced a mapping tool for social media that provides an understanding of cross-media penetration of brand recognition - more can be found here.

"Despite the surge in interest in social media, and potential impact to connect brands with their customers, it remains one of the least understood areas of digital marketing. Network Sense provides unrivalled insight into social media networks for marketers by mining and mapping social network links to illustrate a brand's total online reach as well as benchmark its level of presence against competitors."

I haven't seen this mapping in operation - but would love to Antony :-) - in order to see what the results are but I still have a question about the following:

"Spannerworks' advanced social network analytics are then applied to help marketers build models of relationships and influence within their identified networks."

I understand the reach and ability to track conversations, but it's the final leap that makes me uncomfortable, not just from Spannerworks but from a number of other suppliers that claim to be able to map influence - even at a model level. 

Again, I'd be interested to see the definition of influence behind this process - I'm not calling for transparency in the algorythms that might weight things in differing ways and thus provide SpannerWorks with the edge - but I'd truly like to see a standard definition of social influence developed in an open way for the industry to agree upon from which proprietary systems can then be delivered.  If we can't be open about the definition we can't convincingly say for sure that the recommendations are based on anything more than assumptions.

Accessibility to current opinion...

Stowe Boyd reminds me of the fact that social influence online is still dependent on design and ease of use from a technology perspective.  Accessibility of connection to the content needs to be addressed, in the same way that our very own PR blogger Stephen Davies flags up the need for a site to be designed well to be discovered by search in the first place.  Connection to the content can't become recommended unless a number of different types of people can find the content on the site in an easy way. 

And yet, as the design-led sites of the internet change into fast-moving and simple meme-driven conversations, this will pass.  Whilst Twitter is scorned by most in the western world, looking at it from the perspective of the developing worlds which use the mobile phone as the office and have no intention of buying a desktop PC, discoverability of conversations and the resulting change in the flow of social influence will change also. 

The permanence of the data on the Internet as a search resource will move towards a conversational and consultative process where opinion can be gathered in a peer-to-peer manner, thus getting over the historical inaccuracies of the web as database.  This will inevitably lead to a move from the information economy to the opinion economy, and is the reason why traditional media needs to take a long hard look at its business model.

He's got the power...

It looks like Todd Andrlik has automated the Power 150 list which highlights amalgamated rankings between a number of different tracking solutions - now that's more like it.

Of course, as the case study highlights, this ranking is based on "a proprietary ranking algorithm that uses Google PageRank, Technorati Ranking, Bloglines Subscriber Count, and his own personal rating on content and delivery"

So, Todd, care to share?  Is the Power about influence, reach, popularity or footprint?

Map of Science...

This is a wonderful representation of location-based science papers:

"Data for this map is based on Thomson’s 2003 citation databases. The analytical procedure was hiearchical clustering using co-citation analysis. Visualization uses VxInsight, a proprietary software package from Sandia National Labs."