This presentation is intended to explain the Buzzdetector attitude toward social media monitoring: the human touch is crucial to make sense of the data collected through monitoring activity.
Outsourcing is a rather common attitude for companies when talking about communication and advertising.
Most of the time, the process consists in assign the task to an agency, give a brief and wait for results.
But when it comes to monitoring and to get involved in the social web, better sit down along with the agency and stay tuned for every step taken. Not because agencies are not able to perform the tasks, but when you work on the perception of your company/brand, on the most important asset, it’s crucial to be part of the process.
Every word counts in the social web.
Self-consciousness is a real issue.
Every corporation is nurturing the highest consideration possible of itself but this may prove to turn into an issue when criticism arise: we are too good at anything we do and no one can harm us.
Reality is the no brand can ignore criticism for ever.
Ignoring criticism may prove not to impact on sales, today or tomorrow but as it work on the brand perception and brand equity, sooner or later, it’s a bill to pay.
We must be prepared to listen for everything said about us with open mind and willing to understand and to learn.
I’ve seen, lately, that most of the commenters about Social Media monitoring are moving away from tool powered only by algorithm and giving the right value at the human contribution.
What to say? I’m a huge supporter of the role of humans in the definition of key aspect of a monitoring activity, since the very beginning and that’s why Buzzdetector tool was engineered to be highly interactive with the editors.
In an interesting conversation launched by Tobi Bloomberg, in the Linkedin group Monitoring Social Media, the question was “We’ve got the tool .. now how do we make sense of the tsunami of information?”
The huge amount of data you can get from a monitoring activity should not keep you away from human interpretation but it should drive your attention to a better definition of expectation and a better keyword selection.
My take in the thread is “Toby, I believe that it’s important to define some expectations before going into the research, which is, by the way, the crucial step to get reliable research keywords. You can then, during the process, change them and change the objectives according to the first assessment.
If this first part of the process is properly performed maybe you avoid the tsunami.”
The first obstacle is the-bigger-the-better approach: it’s not necessary to present thousands of results while it’s crucial to get the relevant ones.
Being able to stay focussed in a maelstrom of feeds is the real added value.
(to be continued)
The latest upgrades are both functional and graphics.
The place where it all begin is the Marked Feeds area: in this widget, editors are placing the selected feeds split in positive and negative.
In the widget there are both the feeds found within the Buzzdetector feeders as well the ones found outside the platform. These feeds can be manually inserted through this easy to use window. The inserted feed will be included in the database and in all the graphs.
Then, to set up the graphs, there is a wide range of option:
Category chart:
through the dedicated widget editors can choose between existing category or create new one and customised them according to the specific need: categories title and definition, number of categories, colours. Editors can select the definition that fits the best with the assigned goal. E.g.: type of blogs, newspapers and magazines which covered the story, etc.
Trend chart:
the chart is built in two levels. The one below is a general overview of the selected timeframe. The one above is the magnified detail of a selection of choice.
Sentiment:
the chart comes from the manual selection of positives and negatives. It provides an accurate definition of what’s good and what’s wrong in the web about the selected keywords. As it is a manual selection, we avoid the possible mistakes coming from linguistic mismatch.
Tag Cloud:
Last the Tag cloud, still coming from a manual selection of the terms with the possibility to approve, remove or ban each single word.
What’s next?
Within April 3, there will be the option of selecting between several languages: Italian, English, Spanish, Portuguese, French, German.
Then Chinese is on its way.
If you wish to know more, just get in touch:
www.buzzdetector.com
g.facchini@buzzdetector.com
335 7465173
Follow me on Twitter