Posted in: Archive
Whilst there are numerous websites out there dedicated to keeping track of particular brand and competitors mentions across the social web, there isn’t , to my knowledge, any solution for watching URLs getting hotter across multiple aggregated topics and services. At least none that don’t require the social voting algorithms that are so regularly gamed.
I crave the ability to perform a search, for a phrase, or keyword, and with a custom ‘hotness’ algorithm find out what is gaining popularity across social channels such as Delicious, Twitter, Facebook, Reddit, Digg, Fark, Stumbleupon etc. all aggregated together, and broken down by keywords. Watching the ebb and flow of the web as things increase and decrease in popularity over time, is for me – one of the massively untapped areas of search.
The focus with many of the social monitoring and analytics startups out there is about what you, as a brand have achieved. What is your reach? How well did your micro-update perform? How are your social media efforts paying off?
There’a a massive list of web based startups that perform that sort of reporting already. Hootsuite, Twenty Feet, Klout, Tap11, and PeerIndex.
The list goes on. Yawn.
I can already measure outbound marketing on social media via Google Analytics, and indeed track the all important conversions by segmenting the traffic.
Personally, what I’d really find interesting, is something that listens to the heartbeat of the social web, and makes sense of the information which is gaining popularity over a period of time.
For example: To find out what is hot on the web today surrounding the topic area of say: “Golf”:
Digg and Reddit both provide searches around this. But somehow this still feels like we are examining the past. It’s not a reflection of what is getting popular, it’s more a has been popular recently. What if we take Twitter Search? It’s real time, so should provide articles which are showing signs of popularity around a keyword. Unfortunately not – there’s no way inside Twitter search to sort by the number of retweets to give any indication of popularity inside their search.
Delicious and Stumbleupon‘s tag based systems, are probably the closest tools in any bloggers toolkit to find popular content relating to a particular topic. They are still however; limited to the users of this service, and in no way reflects the web as a whole. Google News again comes close but no cigar, with the sites which are listed needing to be verified as news sources.
Google blog search works on a first come first served basis, and only shows what is recent, not that which is algorithmically popular. Where they struggle, is that recently published content has no ranking factors attributed (i.e. links) and thus, Pagerank falls flat on its face.
Topsy, is by far the best search engine I’ve found for social content in a topical area, but relies far too heavily on Twitter to be a complete data set of what is trending.
Outside the world of golf, services such as Techmeme and MediaGazer have the right direction and are useful in their own right; but seem to have a more editorial, rather than algorithm focus, which frankly will never scale.
For media companies, this type of service would be extremely useful; not only would you get a feel for what stories competitors are running with, but as the social web kicks to life, it would highlight what stories people are starting to find interesting. It could actually affect newsrooms, and the direction that editorial content should take based on live, realtime analysis of topical popularity. Not only that, but we’d also be able to examine what content the web found interesting for a period of time, and archive that information historically, and potentially restrict to a geo location.
Google are pretty good technical fit for that that sort of capability. They have the depth of information to analyse exactly what is happening on the web at any one time, although they lack the social signals that services such as Twitter and Facebook provide.
One of their more interesting acquisitions recently was Postrank. The service that is able to see on a daily basis how popular your content is on many of these social services, and indeed how engaged your audience are on them. The following is a report that I received recently via email.
All Postrank users can receive these sorts of reports for their own, and indeed competitors websites. How brilliant is that? A service that keeps its ears to the ground for you, and is listening to the web proactively, rather than watching exactly what you do in terms of social media marketing and reporting on it. Postrank scores also reflect how many comments are made on your articles, as a reflection of how engaged your audience are. The provision of a Google reader plugin also provides valuable insight into the popularity of your own, and other people’s content, making it a pretty useful service overall.
The acquisition earlier this month was no real surprise. Having this kind of data at its fingertips, as well as the signals they already have makes perfect sense for them. I’ve already expressed why I think engagement metrics are going to play a part in ranking algorithms, and this buyout is a definite move in that direction, as well as an overall bolstering of their social strategy.
My prayer to Google? To use the data they collect from Postrank, along with the other social information they have to create a living, breathing hub of social activity inside Google +1. As I’ve described earlier, it could aggregate all of the information being shared from multiple services into one place, and allow us to search and drilldown into topic categorisation to find the popular content for that day, including the +1 information.
I knocked this up (very roughly) in 5 minutes in Photoshop to try and illustrate.
1) A graph for today showing rising popularity of certain pieces of content (across the web generally).
2) Popular content within a topic on Twitter, +1, Facebook, Delicious etc.
3) Popular publishers / influencers within a certain topic of interest. (see below)
“Google announced today that it has begun indexing attribution of content to particular authors, not just to the websites they appear on. Links associated with the author of a page can now have the code rel=”author” added to them and Google will understand that to mean that the linked name is the linking page’s author. That’s a potentially significant change to the balance of power between sites and the individuals that create for them.”
4) Popular content shared by people I know inside a topic. (Friends etc.)
5) Easy access to all of the content in a topic that I’ve previously shared.
6) A way to view historically what content was popular in a topic on the social web.
For shorter tail queries in recognised topical areas, I think this would undoubtedly prove more interesting and relevant for users than the current Wikipedia entries and long established websites which don’t have to regularly update to rank. If Google +1 will be as content centric as this remains to be seen.
Feasible? Or pipe dream?