Many companies have bought into monitoring their brand reputation via social automation tools such as Radian6, Meltwater Buzz, etc. But can they really measure the sentiment of your fans/followers? Don’t get me wrong, I’m all for social automation. I practically live out of HootSuite. But do I use any of them to measure brand sentiment? No and quite frankly speaking, it’s because I think they cannot measure the sentiment accurately. In fact, they’re about as accurate as the weather and here’s why:
- It’s a keyword algorithm. Social automation tools that claim to measure sentiment are simply filtered by keywords. But those keywords are input by a human and they are not inclusive of all positive or negative keywords.
- It can’t discern grammar very well or sarcasm. For example, if I say: Let’s eat Grandma vs. let’s eat, Grandma. It means two different things. And if you’re Grandma, it could be a scary situation.
- Context means everything. Often social automation tools have a tough time measuring context. Context can e the social network it’s coming from, for example, something scaled down on Twitter might have convey a different context than something on Facebook or Google Plus. Also the time, the day, or if something happened in the news that sparked a comment that might seem normal, but instead it was negative due to when it came out.
Or what about when someone says, “I need a Visa” do they mean they need an immigration Visa or a Visa credit card?
Bottom line, in the words of my friend Neil James, ‘social media software just doesn’t have the capability of parsing out emotion from words alone,’ and I couldn’t agree with him more.
Do you use an online social automation tool that supposedly measures sentiment? Do you think it’s accurate?