Chat log topic and sentiment analysis
Picture by Nicolas Nova
Creative Commons License
More and more, companies are using chat as a means of running their support function. This approach has some clear benefits over traditional phone support. It is convenient for customers, because, while they wait for a response from the agent, they can do other things at the same time, and it allows agents to serve more than one case simultaneously. In addition, customers are typically logged in to your service, allowing you to have a direct link to your CRM database. On top of this, the logged chat is available immediately in text format, as opposed to be available only as a recorded audio conversation, making further analysis easier.
While chat is proving to be an effective tool for resolving acute issues, do companies take full advantage of the insights buried in their chat logs?
As a business owner it is, of course, essential to figure out what your customers are saying about your products and services as well as how well your support staff is performing. Knowing the answers to these questions help you find ways to easily improve your business performance:
- For which topics do your customers use the support chat most? Which aspects of your products or services get the most feedback, and is that feedback predominantly positive or negative?
- Which agent is able to close the most cases resulting in a sale? What does this agent do that other agents can learn?
- What are the topics discussed in chats that lead to a sale and what are the topics discussed in chats that did not lead to a sale?
- Which agent’s chats are more positive and/or effective than other agents’ chats? Do some agents need more training, guidance or need to be reassigned to other tasks?
All these insights are buried in your chat log text data, but how are you able to analyse the data in such a manner that it leads to actionable insight?
Your agents are busy closing the incoming cases–often multiple in parallel–and at the end of the day they may not even remember the topics discussed. Because of this, they may not be best-equipped to pass on the higher level insights you, as a business owner, are looking for. On top of this, if you are running operations in different countries, your chat logs will contain multiple languages, complicating the analysis even further.
How to get a grip on the steady flow of feedback buried in your chat logs, and make sure you can put them to good use when wanting to improve your business operations?
The best way to approach this is to look for a chat log analysis service, that:
- Is able to retrieve what topics are being discussed in your chat logs
- Can accurately determine with what sentiment those topics are being discussed, in multiple languages simultaneously
- Visualizes the analysis in a way that allows you to easily drill down to what is important for you and your business
- Flexibility to build reports allowing you to track indicators such as agent performance
- Allows you to easily distribute the insights to key stakeholders in your organisation
To learn more about how Etuma can help you improve your business by using our chat log analysis services to analyse your support chat logs, please contact us for a free trial.
Loek van der Helm is Director, Sales and Business Development at Etuma Ltd (www.etuma.com). Etuma is revolutionizing customer feedback analysis with our multi-language universal analysis engine and industry-specific visualization. Requiring no lengthy or expensive configuration project, Etuma360 can be connected to any data source in real time via standard web interfaces. This allows our customers to quickly and easily obtain actionable insights on the topics and topic sentiments in their customer feedback or chat log data.