Organizations now have a goldmine of social data about about their customers available at their fingertips (and for free). This is a blessing for a customer service organization that truly wants to understand their customers and optimize their products and service. Unfortunately, it can also be pretty daunting. It is not easy to structure this wealth of information, and more importantly, interpret and analyze it meaningfully. The growing volumes of social data isn’t helping to speed up this process.
That’s where data classification and identification comes in. The process of organizing (i.e. tagging) your data makes it easier to find what you need.
It’s a proven concept for social customer service to tag incoming mentions. This allows you to easily capture context about an incoming mention, filter on it, pour a specific data set into dashboards, and report on your social media efforts. You can do this manually, but in order to save time and cut costs, tagging mentions automatically is commonplace.
How companies are categorizing data (i.e. whether or not they’re automating it, the percentage of mentions they’re tagging) differs greatly from one organization to the next. It depends on the different types and volumes of data they’re pulling in, and the overall social objectives they want to achieve. In recent times, “tagging fatigue” is becoming more common. Especially for contact centers and marketing departments, etc. that have attempted to categorize every mention, tagging each mention has become a pain.
Bringing Data Classification to the Next Level
Beyond tagging automatically, other aspects of data classification have taken a huge leap forward with the introduction of time-tested category models. These models help to filter, route, and analyze data even more quickly and effectively.
Category models come in two flavors:
- Vertical models that focus on data categorization of major industries (such as banking, retail, telecom, and transportation).
- Horizontal models that help with the analysis of different feedback approaches. For example, these models help you understand the website or customer support experience, or determine basic emotions in customer feedback (such as anger, happiness, surprise, etc.), and many more.
There are many different use cases across different industries for category models. Below are a few cases to help illustrate the purpose of category models.
- Social usually is the first channel to pick up on a crisis. Imagine if an automotive brand has a separate folder of data collecting all mentions about the breakdown of cars. By automatically filtering on it, you can instantly prioritize and troubleshoot this top priority.
- By collecting all the data that expresses anger in one folder, your customer service agents can quickly counter these cases.
If you’re interested in learning more about category models for data categorization, read more details about the horizontal and vertical category models in our support article guiding you through a few use cases for the different types of classification. For a higher-level look at getting your social customer service program started, download the eBook Building a Social Customer Care Team.