A major benefit of social media is that you can see the good and bad things people say about your brand.
The bigger your company gets, however, the harder it becomes to keep a handle on how everyone feels about your brand. For large companies with thousands of daily mentions on social media, news sites and blogs, it’s extremely difficult to do manually.
That’s where sentiment analysis software comes in. It monitors and evaluates your online mentions to show you how the whole Web is reacting to your news in real time. Goodbye ‘overwhelmed’, hello ‘big picture’.
To find out how it works and what the benefits are, we asked Natalia Dykyj, director of product management for Vocus software (including the sentiment-analysis enabled Vocus PR Suite.)
Vocus Blog: Why is sentiment analysis so important for marketers?
ND: We’ve heard for years about how important it is for brands to listen to what people say about them online whether it’s customers, prospects industry influencers or the media.
Not only do these represent real opportunities for you to engage with customers, but they’re also a great way to keep your finger on the pulse of consumer satisfaction with your brand.
So think of sentiment as a natural extension of your monitoring. It’s about being able to categorize all those conversations based on the perceived tone so you can react accordingly.
How do you measure sentiment?
Many marketers use themselves as their own barometers of sentiment. Meaning, if you’re already monitoring conversations online, you’re probably a pretty good judge as to whether a particular mention is positive, negative or neither.
But to measure sentiment on a larger scale, brands typically use software that applies Natural Language Processing (NLP). Although the exact methods used by each software may differ slightly, the ultimate goal of NLP technology is to process and understand human language and automatically derive meaning. Or, in this case, sentiment.
NLP makes quick work of identifying the sentiment for a large amount of conversations, helping you as a marketer home in on messages that require more immediate attention or to better understand the broader trends.
Although the technology behind NLP is very strong and considered the industry norm, it is important for marketers to understand that sentiment analysis is never perfect. There is a lot of subtlety in language – like idioms, sarcasm and metaphors.
Even the best automated sentiment tools can’t gauge the intent of a message with 100 percent accuracy, just like two people reading the same message may disagree on its sentiment. Especially since so many conversations happen on social media, it can be hard to derive meaning or sentiment out of something as short and informal as a 140-character tweet.
Where should marketers track sentiment?
The natural tendency is to think first about social media because that’s where your customers and prospects so often go to voice their opinions. But you can’t ignore blogs, the media and other digital outlets and their ability to influence customer buying behavior.
Research has shown that 65% of people who visit blogs say that brand mentions within the blog’s content influence their purchasing decisions.
Customers also still place a lot of trust in news outlets. And remember that regardless of where the conversations happen online, they’re likely to end up in search. Search is still the first place that customers go to research and find information on products or services. They don’t have to go directly to that blog or news site to get the review about your business. So don’t forget to look across channels in your monitoring and sentiment analysis.
Once you measure sentiment, what should you do with that data?
The first and most important way to react to sentiment is to engage directly with customers. There are many, many ways for you to engage with consumers and your fans/followers online. But that’s definitely a topic for its own blog post!
For today, I’ll focus specifically on conversations where there is a distinct positive or negative sentiment. I’ll start by saying that different brands have varying policies for how to react to mentions online. For example, some choose not to engage with anyone posting negative comments. So you should start by figuring out what’s the right approach for your brand.
But being able to identify those positive and negative mentions is what will help you define your next steps. If someone says something positive, we see brands have success when they reach out with something as simple as a thanks! It’s an easy way to show your appreciation and maybe even turn that consumer into an evangelist.
Equally, if someone is saying something negative, you might want to reach out and extend a helping hand. Help resolve the problem before it becomes something bigger or affects how others perceive your brand.
The second thing to do with that data is to look at broader trends. How is perception of your brand shifting over time? This can help you not only pinpoint potential causes, but also head off any negative trends or continue to bolster the positive ones. It’s taking a macro view of sentiment.
But regardless of whether you’re looking at individual mentions or broader trends, that sentiment analysis will help you think through the best way for your brand to react and engage.
How can businesses monitor everything said about them online?
Depending on the scope of your organization and the resources you have for brand monitoring, it can be tough. Brands must ultimately rely on software to do the heavy lifting for them. There are a lot of benefits to these types of automated tools, but perhaps the biggest is their ability to search across all digital channels.
Having access to all your social, blog and media mentions in one central place makes it easier for you to manage, act quickly, and really have the broadest possible view into what people are saying about your brand online.
Is it important to monitor everything said about your brand?
This really depends on the goals of your organization. If your intention is to respond to all positive and negative mentions of your brand, then individual messages are important. But when you’re looking at broader trends, you don’t need to worry about every last message.
Remember that there is a lot being said out there. My favorite analogy of Twitter is that it’s like drinking out of a fire hose – it just keeps coming and you’re not going to be able to catch everything.
So as long as you’re able to catch a consistent cross-section of messages across the different digital channels, that’s really all you need for trend analysis. That’s more than sufficient in helping you understand the spikes and dips – where any individual message in itself is not important, it’s the trend as a whole.
Image: roboM8 (Creative Commons)