Social media monitoring is all about the people who read
There are essentially two ways of measuring how much discussion exists online (the 'volume of buzz') for any topic. The first option is to measure mentions of a particular search term: “x posts mentioned Barack Obama yesterday”. This seems like a very intuitive way of doing things – indeed, it’s what we used to do until early 2005.
The other way of measuring buzz is to focus on what is being read rather than what is posted: “there were x page views of Barack Obama discussion yesterday”.
Measuring readership rather than posts may sound a bit cumbersome. It’s certainly harder to do. You need to make a bunch of estimates to get the page views - and that clearly introduces some room for error. However, since early 2005 WaveMetrix has run all projects with buzz measurements based on the readership method. Why? Because not all posts are born equal. Treating them as such makes buzz data a nonsense.
Imagine two posts saying exactly the same thing - for example, “My experience of Sky was made worse when they swapped my existing box for an Amstrad version”. Let’s imagine the first post sparks a large thread about Sky on a very popular forum with millions of page views. The other comment is posted on the homepage of an occasional internet blogger with fewer than 10 online friends. If you are Sky and you are trying to understand buzz around your various Sky+ boxes, would you allocate equal weight to both posts? Of course not.
At WaveMetrix we did a lot of work looking at our quantitative data and how it correlated to traditional research and sales figures. We found that measuring readership rather than just mentions brought about an order of magnitude improvement in data quality. So we changed our methodology - and now all buzz volume is measured as readership.
Measuring buzz volume: impressions
At WaveMetrix we measure volume of buzz in 'impressions'. These are essentially page views, just as you would count for any web page, or impressions as you would count them for a TV or print ad. We call them 'impressions' because this is our in-house measure.
We use impressions for measuring not only the volume of buzz, but also its sentiment. Within all categories each opinion is weighted according to its number of impressions, so that those opinions which are read most and are most important also receive the greatest weighting when calculating sentiment.
Measuring engagement in impressions is also very useful in determining which sites or which posters are most important for a particular brand. This insight is particularly valuable for marketing and brand managers, as well as media planners. We can identify specific sites or show where buzz is located according to site type. For instance, we can differentiate fan site buzz from that located on more mainstream sites, allowing us to understand how engagement and attitudes vary by community.
Harry Potter Buzz by type of site
