Customer service automation is based on automatic tagging as a crucial step. Tagging based on identifying keywords in the text of the conversation has been the norm for a long time. To address the limitations of the keyword-based tagging of conversations, many NLP-backed solutions offer a more robust approach based on the meaning of the ticket content rather than the actual words that compose it. Such solutions use high dimensional vectors to represent words and sentences and can easily identify similarity and proximity of meaning. In more practical terms, two cases containing the utterances “I want my money back” and “I want an immediate refund” will both get tagged as a [refund request].

This kind of processing can grasp the meaning of a ticket and do tag it according. It is naturally more reliable than the automatic tagging offered by CRMs such as Zendesk.

The limitations of the meaning-based tagging

Although it is valuable to have a tagging system that identifies more than the exact terms of the same concept, this system still fails to address the real challenges facing customer support teams.

Although it is valuable to have a tagging system that identifies more than the exact terms of the same concept, this system still fails to address the real challenges facing customer support teams.

Customer support teams deal with infinite requests and tend to divide issues into subclasses, so they can sort them and manage assignments. A very common practice is to classify cases by the employee skill sets such as technical, commercial, and billing. Such a partition seems obvious and natural while being sometimes criticized as employee-centric rather than customer-centric. But a similarity-based tagger will rapidly fail to classify tickets based only on a department label.

Besides, tickets can be classified according to the request type such as technical issues and shipping issues, or according to the customer sentiments (positive, negative, or neutral) and emotions (happy, angry, disgust, or stressed).

More importantly, it can help if we can spot the situations embedded in the ticket (urgency, request, or call to action). And why not classify tickets by the communication quality (toxicity, obscenity, awkward,…)?

Customer service automation: Tagging and Routing

Businesses need to tag tickets and conversations on multiple dimensions that represent several aspects of the same interaction. For example, in this ticket, «your app doesn’t work! Fix it as soon as possible, or I’ll have to uninstall it.»  Will be tagged as a “technical issue” from an employee skillset perspective which is correct, but ignores all other aspects of the ticket. On the other hand, the contextual approach proposed by Exacctly can detect the undertones of the conversation such as the context of urgency and frustration in this case, which provide the agent with more details to uncover the user’s feelings and expectations and help him have a better understanding of the situation.

This contextual tagging includes:

  • Situation aspects like urgency and request,
  • Emotional aspects like anger and happiness,
  • Technical aspects like installation and operating issues

The support team can add any concepts it estimates as interesting to identify inside a conversation, and Exacctly does the work. There will be no limits in identifying your unique business context by using the power of the “hypothesis detection” feature because a “technical issue” for a retail business is way different from the “technical issue” for an Edtech startup.

Exacctly combines the automatic tagging with a rule-based router to help Customer service automation through better routing for each ticket to the right agent at the right time. Agents will get more free time to focus on serving the customer in a better way.

Using Exacctly tag-based real-time routing allows the team leader to prioritize and identify when and to whom some tickets should be escalated. In some cases, the agent can’t solve the customer’s problem, especially if it is a technical issue that requires an in-depth comprehension of the product and that’s when the escalation should happen. The ticket will be routed to another agent that the team leader has already predefined to interfere in such a situation, to help the client solve his problem quickly and efficiently.

Exacctly enables effortless customer service automation through adding multi-aspect tags to all of your tickets. That will enable you to understand your customer in real-time and take appropriate actions whenever needed. By leveraging the power of AI and affective computing, Exacctly will help the support team take the customer experience to another level.

Exacctly integrates to the most used CRMs including Zendesk, Freshdesk, and the Email box. Get in touch with us for a demo on how Exacctly can redefine your customer support process!