Strata Conference and Big Data – Web Journal October 23-25th, 2012

I have been at the Strata Conference the last couple of days in NYC and got a chance to do a couple of things here including

  • Talking to most of the vendors exhibiting so as to understand this space better (I think I do, a little better now)
  • Catching some great sessions (and I spoke here earlier in the week)
  • Meeting some new people that are in an adjacent space from my own (Big Data/Strata).
  • Getting my portrait sketched and posted digitally on EventToons, which does not appear to have been uploaded yet or I would share it here.
A couple of things stood out to me so far (and I’m here for the rest of the afternoon).
  • In Human Intelligence as a Signal to Predictive Analytics yesterday afternoon there was an interesting solution that focused on social analytics and the capital markets.  Fundamentally, what I heard wasn’t much different than any of the other solutions/vendors I know of and have  looked at, but for one thing.    The difference was in how the unstructured data was decoded into signals and then compared with known performance of the market indices.   It so turns out that the same approach could be used for CPG or almost anything else, but I don’t see that particular way of looking at the data and comparing it in the vendors I’m familiar with.   So that was interesting.    What was also interesting was decoding the hype around the rest of it – many people come up with what sounds like new solutions, but in this case, the only thing I was seeing as significant was a slightly different approach (which is notable in and of itself).
  • This morning I also listened to Beyond Batchwith Doug Cutting of Cloudrea.   Doug pointed out a few things that really spoke to me including this insight (see below):
    • “People will only adopt technologies (ie: Hadoop/Big Data) when they are comfortable with it”.  That says a lot to me because it explains a lot of the problems with social media/emerging media and businesses of all sizes, particularly small and medium size business.    Clearly, nothing that can be proposed for an analytical standpoint is going to work very well until the stakeholders/corporate culture  can feel comfortable and secure with it.  That might also explain the resistance to it, esp on the Social PR side, that mouths a lot of big data, but is more about glitter and story (in a mixed sense).
  • I also listened to Storytelling with Data with Visual.ly and, much akin to my Rutgers course that focuses on storytelling and liberally uses infographics.   What Romy Misra came up with was two memorable points:
    • If your audience is Data Scientists and Analysts your analytics dashboards don’t need to tell a story.  If your audience includes anyone else that isn’t a Data Scientist or Analyst, your dashboards do need to tell a story.
    • Artful use of Infographics makes the Infographic the storyteller.
  • In Text-mining Your City, of obvious interest to me, my conclusion is:
    • Organizations, particularly governments (say, the cities of Chicago and New York for example) have decided they have a lot of great data to share, but don’t like any of the existing marketing solutions and have decided to build their own using R and The Cloud.  I see this as a continuing trend.
      • On one hand, the increased use of smart programmers who are comfortable coding and have free data has created a very decent set of quick solutions that aren’t terribly expensive to implement.  This then, undercuts the existing solutions that no longer (if they ever did) meet the market or governmental needs, that are more fluid than what these platform can provide an acceptable cost.
      • On the other hand, as more of these custom solutions get spawned, the more likely they will create increasing fragmentation (at least, it looks that way to me from where I sit, now).
I’ll be attending some sessions this afternoon and skipped lunch to write this post.  There is some more news though and it does impact some of the people / companies I know.
  • Salesforce trims Radian6 by laying off 100 people – that is the beginning of the Social Analytics bubble popping, I think.  When you think about, that’s got to be close to half of the Radian6 crew!  Wow!  Not really surprising to me though – part of it is redundancy – yes, but i wonder if the other reason is unstated.  These companies, Buddymedia and Radian6 (as an example of the social space) aren’t that profitable, just look at the numbers.   The same people can handle both – and they had extra people that weren’t needed, because Radian6 and Buddy are about to be absorbed into the Marketing Cloud, even though, in fact, they still don’t interoperate today.
  • Another piece of news is that the Social Analytics platforms are next to useless for much of the video/graphics media because it’s not keyword based.  I knew that, but maybe users are just waking up now and realizing there’s a lot of “ultraviolet” or better yet, call it “dark social” data they can’t pick up with these tools – and Big Daddy Salesforce just bought one of them last year. Ha!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>