Several speaking events this week and next, I posted about them recently and it starts happening in about 24 hours.  Added to it is the Internet Evolution (an IBM Sponsored Community) on Analytics Costs and Benefits, Friday, November 4th, at 2PM EST that wrote about last night (the presentation for that is pretty much done).

Keyword Thoughts

One of the things that struck me recently is how iterative social analytics is, when keywords are being used (which is 99% of the time).  On Thursday, at #scrmnyc (glad Luke finally got the twitter hashtag right for this conf, I was confused with his #scrmync that he had, before) I will be on a panel focusing on customer listening strategies.  On that panel, among other things, I’ll be asked how online listening is being used and how it moves an organization forward.

If you start with a simple keyword list, the most obvious words and perhaps some misspellings (along with Boolean logic so you don’t pull in a lot of mentions you don’t want), you’ll probably be missing most of the interesting conversations that are happening – because as conversations begin, they change and morph and other keywords are added to the list (queries).   Also, as the same people chime in over and over, people who are familiar with the subject, they drop out many of the likely suspects, because everyone in that conversation already knows many of the details.

A friend of mine recently has to do this kind of “Iterative Keyword Building” and I asked them how long it would typically take.  First, you need to know something to begin, then as you find your first stories or conversations, you build out, and out, and out, the keyword lists – the process might take several hours, but never really is done – as long as the conversations are going, they’re changing and growing.

There’s also a segmentation that could be done, I suppose, for those who are “familiar” and stop using the common keyword phrases, that might be very interesting to have.   I suspect, however, we’ll run into the very familiar roadblocks – money (no one seems to want seriously fund this kind of work) and the all too frequent PR Gatekeepers (who are antagonistic, since they can’t easily monetize and scale this kind of work, and don’t know how to set it and keep it running in the first place).

That leads me to the Web Journal part of this post. Gary Angel posted on PR & Social Media Measurement yesterday and said a few things to add to the above picture that I just painted (and left vague, deliberately).

1. Most organizations (not accounting for COMMs agencies) have several functions, but tend to put Social Media Analytics in the PR organization, which cripples it’s use for other functions such as Market Research, or even Marketing Campaigns.  Gary explains why that is, but you have to read the post a few times to understand it).
2. Just as he put it “Some of the leading listening tools started life as PR clipping services and have simply expanded their scope over time to embrace a wider range of channels. As a consequence, PR functions tend to have a larger voice and a clearer set of imperatives when it comes to Social Media and Social Media measurement“.
3. In Gary’s post, he suggests that platforms which include more of the mainstream media (traditional clipping services) are better for PR (that would include Cision and Dow Jones Insights/Factiva) because many of the monitoring platforms such as Radian6, Alterian, Brandwatch, Sysomos, at el, don’t capture TV / Radio Transcripts, don’t capture Commercials, don’t capture Daily Motion, Consumer Magazines, Trade Press or Opinion or Review Sites (such as Amazon, TripAdvisor, Yelp, etc).

Gary Angel ends the post by saying that it’s better to just abandon the other functions (like market research, social campaigns, community outreac,  etc) if PR/Comms is running the social media analytics (which it is, too often, sometimes simply by choosing the platforms) because working with “bad data” is worse than just not doing the reporting at all.  It’s hard to argue with that position.

ReadWriteWeb has a post on the new Facebook Lifestreaming – that is the first actual example I’ve seen; but it’s also an example of how visualization of data can fundamentally change data, making it far more useful, as Richard MacManus  suggests.  I suppose that would include mapping technology – such as Breakthrough Technology that Maps Rainforests In 3D, Including Carbon Storage as reported in TreeHugger. The data was already there, it was a new technology to map it that showed new relationships that you could not see as clearly before.

According to the post:

“…The mapping includes not just the 3D structure of the forest, but even the chemical diversity from photosynthetic pigment concentrations to water content in leaves to micronutrient contents. This is what allows them to know which species they’re looking at on the map, and how the forest is doing in the face of a previous drought and possible deforestation. “

Getting back to online monitoring – not all of it is considered to be positive – such as  Do we need a line between big data and big brother? There are pros and cons here – but it’s interesting what direction this is going in, just read this quote from the post..

… consider today’s distributed workforces in which employees are spread across multiple offices or even are working from home. In the past, Charnock said, managers could see their employees on a daily basis and might be able to put two and two together if, say, the guy who came in looking hungover every morning was also an unproductive employee. That’s not always possible today, but Cataphora can help replace those in-person interactions if that same guy is emailing co-workers about his wild nights or tweeting about them from his office desktop.

So .. not only is Facebook coming at this in one way via Lifestreaming, but Cataphora is going at it from another angle .

I guess there’s more here (after all, I had hoped to cover Oct 19th till today) but in reality, I had posted web journals for most of that period already, and most of what I found interesting from the last two weeks, I posted about already.