Grabbing users attention, Vol#1
Readers - or better yet consumers are very often willing to interact with anything they find engaging using relatively low threshold of participation. How can their desire to "leave the breadcrumbs of attention" be leveraged to provide instruments for serving more personal and relevant content?
As we engage with the web, we leave behind breadcrumbs of attention - even when we only read. But reading alone isn't enough to fulfill our innate desire to remix our media, consumption is active for consumers turned users (Ross Mayfield).
One of the cornerstones in modern publishing is ability to serve the best possible related content to your audience. From Wordpress to CNN, "related stories","more from this author", or aggregations such as most popular, most commented, top rated and similar variations of trending topics or stories are de-facto standard. Main problem is - most websites don't care about quality of related content - they publish something, anything at all, just to have 10-20-30 more titles on page hoping to squeeze out just a little bit more page views on the site.
Ok, related stories... but related to what?
When you think of related content, you basically try to find a correlation between two articles/stories. Correlations can occur on multiple basic levels easily computed by CMS, i.e. stories can belong to same category, have same author or share same tags. They can be even related by people/places/events(venue) paradigm. Using semantic tagging, with taxonomies in Wordpress, or topics in Vivvo, each story can be framed within those boundaries. You can even use specialized software such as OpenCalais.
The whole downfall of this concept is that two different readers get the same suggested "related" stories. The content that user reads needs to be put in the context of the individual reader, in accord with individual reading preferences and preferences of his/her social circle. That's the biggest mistake of the whole concept that arose from Digg. I'm not interested in pulse of 34,500,000 people. I'm very much interested in what people that I find authoritative and trustworthy recommend - what my social circle recommends! Of course, this is the most complexed and "expensive" thing to do. It is easy for Amazon to provide personalized suggestions, or to Google to form recommendations based on ones social circle, but what about the little people?
Goal: Using as little possible own hardware and application resources to create as powerful and accurate cross-related content based on individual affinity and recommendations of your social circle.
What if we....
For the start, Facebook with the new Social Plugin set allows publishers to leverage power low of participation using "Like" buttons and filtering it through Facebook's social graph offer your community a way to provide personalized recommendations. Ok, not fancy - but neat, automated, free, and uses no hardware resources of your own.
Huffingtonpost blog is a very good example of providing readers an tool to express their affinity using very low threshold of participation. Only problem is, I don't see that Huffington uses this to the benefit of providing better filter for serving content... But it is a good example of roadmap where you could go if you used this to cross-relate with "emotions" contributed from your social circle (even your facebook friends on the portal, if nothing else).
Grabbing attention - most underestimated engagement
One widely overlooked engagement, mostly by marketers, is simple initial challenge of grabbing someone's attention. What is the ratio of number of articles you opened following twitter or facebook links, comparing to ones you actually read (not scanned) top to bottom? Does this have a much higher value than scanning a text and routinely clicking on "share on twitter"? What's more, it is actually relatively easy and straightforward to track scroll depth to reveal content engagement in Google Analytics, and if you want SaaS solution - Chartbeat does it very neatly. Incorporating this into our, let's cal it "social circle related content graph" is not as complicated as one might think, and provides highly valuable information on actual interest of readers that could be interesting to their social circle.
Anyhow, I need to leave something for Vol #2 as well, so stay tuned :)