Facebook starting to look at understanding sentiment

So, according to Robert, Facebook is now looking at understanding sentiment. Yay!

To quote Robert:

**Facebook is, he told me, studying “sentiment” behavior. It hasn’t yet used that research in its public service yet, but is looking to figure out if people are having a good day or bad day. He said that already his teams are able to sense when nasty news, like stock prices are headed down, is underway. He also told me that the sentiment engine notices a lot of “going out” kinds of messages on Friday afternoon and then notices a lot of “hungover” messages on Saturday morning. He’s not sure where that research will lead. We talked about how sentiment analysis might lead to a new kind of news display in Facebook. Knowing whether a story is positive or negative would let Facebook pick a good selection of both kinds of news, or maybe even let you choose whether you want to see only “happy” news.

I have been working with sentiment analysis as the cornerstone for my social media strategies for over two years and have found it to be one of THE most useful aspects of measurement on the social web…so much so that it has helped me build a very basic and very assumptive ROI calculator (more on that in another post perhaps!)

Measuring volume is only half the story

“Traditional” web marketers find it fairly easy to see how we can count conversations when using our wonderful tools. It is not a difficult challenge to understand how we can aggregate conversations into the largest or newest topics, or even the most popular locations of those conversations.

But, thinking that the aggregation or watching of the volume of responses is sufficient to understand the true picture of online chatter is to fall into the “no press is bad press” trap.

Sure, understanding how many people are talking about out product launch is one indicator of success but how is it possible to understand the meaning behind these conversations using volume alone?

This is where the measurement of sentiment make the difference and IMHO sets the good social media strategies out from the bad.

Sentiment is actually much more than half the story in fact…

The above is a screenshot from a wonderful tool I use which tracks the volume and sentiment of conversations. Looking at the graph, I can easily see precisely when sentiment was especially good (high) or bad (low) and dig deeper to understand what was driving the score to be so good or bad.

Looking at the table below the graph, I can see what the main topics of conversation are, the sentiment of those conversations but perhaps more importantly, the specific score of that conversation – based on the authorities of the people discussing it.

Frequently, marketers see sentiment as being simply good or bad. BUT by attaching a sentiment score to a comment or topic based on the “authority or influence” of that person, we can get a more precise idea of whether someone is simply satisfied or a real fan of our product.

Likewise, if we are comparing our performance against our competitors, would/should we be satisfied with a positive score of 1.2 if our competitors are averaging 2.4?

Volume of comments doesn’t tell us how we compare. Sentiment does.

Sentiment is all about the feeling

Sentiment monitoring provides us with the ability to measure the emotion or feeling behind chatter, giving us insight into what people really think about our product.

It tells us (amongst other things):

• Has our product been well received?
• If yes, who are the people expressing positive sentiment about it?
• If not, why and who are those people?
• If sentiment is being expressed, what are the key issues that this sentiment is being expressed on?
• If positive, who are the people we can begin to engage with to help us spread the word further?
• If negative, who are the people whose problems we need to be solving or to show them the light?

Only through understanding sentiment can we really get to the heart of social media conversations and if Facebook are looking at this seriously, they have a great opportunity to take Market Sentinel’s lead in using the measurement of feeling to drive more targeted and therefore more relevant content to users (I use the word content rather than ads, because i genuinely feel that if advertising is to get anywhere on social networks it has got to be through interactive content with compels people to respond rather than simply being a passive piece of colour on a web page within a social network).

Social ads were/are great, polls are the great untapped gem of a resource on Facebook and go some way to delivering a targeted message, but both are only targeted to a fairly static personal profile – it’s like advertising convertible cars on a billboard all year round because someone who lives nearby once said they liked convertible cars – targeted but not relevant at a given point in time.

Imagine ads that could be delivered according to the feeling of someone’s status updates, or infamous relationship status’…and we are getting somewhere near where sentiment can truly take us.

Sentiment moves us closer to Social Media ROI

I’ve been thinking and tinkering with the idea that if we know the numbers and identities of the negative commenters, that we can use outreach to try and “turn” them positive. (i’m being blunt with the term “turn” for brevity) In turning negative commenters into positive ones, then the chances are we are going to increase the number of buyers.

If we know that form a certain site or referrer, the conversion rate is a certain % and that the ABV is worth £XX then theoretically (and with some generous assumptions!) we arrive at some sort of ROI for social media activity.

I hope to explore the thinking behind this a little later, but for now…just imagine what you could do with your social media activity if you knew what people were FEELING.