LIBRARY

Zendesk support ticket tagging: Why NLP is so powerful

Free resources! Subscribe to our newsletter to download free templates, eBooks, and to join our exclusive Slack community.

Zendesk support ticket tagging: Why NLP is so powerful
Download Here Image

This article is written by our in-house NLP scientist, Suhan Prabhu. In this article, Suhan explains NLP support ticket tagging: why and how it works so well, how our Zendesk integration tackles the challenges faced by support teams, and the use cases made available by AI-power customer support ticket insights.

--

There's a vast quantity of customer data hidden away in the customer support department. If you unlock it, you unlock growth.


Support tickets hold rich, unbiased customer feedback that comes in rare high volumes—which makes them a goldmine for tackling customer churn and improving lifetime value.

However, a manual ticket analyses is not always feasible from both a time or objectivity perspective. The high frequency and complex nature of support tickets make this so.

At SentiSum, we've been building state of the art natural language processing (NLP) modules designed for feedback analytics. The use cases unlocked are powerful. Indeed, you can automatically analyze thousands of tickets in real-time, uncovering insight with unparalleled speed so that you can be proactive to customer friction points.


NLP makes high volumes manageable. Process automation here also removes a painful task for the contact centre: ticket tagging and tag training. Who wouldn't want to reduce agent admin time in an environment already fraught with high attrition rates?


Automating your insights also brings a new level of accuracy to tagging. It's difficult to give support agents adequate detail about tag meaning and appropriateness. This often means that tags end up being high level enough to be easy (but, therefore, fairly useless), or too detailed (requiring regular maintenance, training and resulting in error). Automation solves this dilemma.


This article talks about the different types of NLP and explain why machine learning is so powerful when applied to support tickets.


We'll also show you what we do with Zendesk, and the use cases unlocked by combining NLP with Zendesk.

What is NLP?

Natural language processing has recently emerged as a popular subfield of computer science and artificial intelligence (AI) that applies software to automatically manipulate natural language such as speech and text.

NLP systems can be one of two types:

  • Rule-based systems: These systems use handcrafted rules to detect linguistic features in text. These rules are based on part of speech (nouns, verbs and their order), stop words (and, an a, the) and keywords such as named entities.
  • Machine learning-based systems: These systems learn rules and patterns from the data based on statistical inference. They are dependent on an initial curated dataset which contains the insights/features to be extracted, along with the raw text. Using this data, an ML system learns the internal rules that determine these insights based on the co-occurrence of words in the curated text.

Let's take an example of how each would handle three customer support tickets:

Ticket one: Hi, my delivery was never recieved. Please help me find it.

Ticket two: Hi, I'm here to complain about my delivery. It hasn't arrived yet.

Ticket three: Hi, my box was due to arrive between 5-7pm yesterday. I woke up this morning and it still has not been delivered

As humans, we know these three tickets have equal meaning. The box was not delivered.

A rule-based system built for a company wanting to track how many failed deliveries they had would be built with a rule like: Tag the ticket if it includes the words 'box', 'not', and 'delivered' anywhere within it. It would only tag ticket three correctly, a 66% failure rate.

A machine learning-based system, however, would correctly tag all three. It would understand that sometimes people use 'arrived' instead of delivery, and 'received' instead of delivered. It would also understand that, in ticket two, when the customer says 'it', they are referring to the box.

ML based systems are generally more robust when it comes to analysing misspelled words or ones that haven't been seen before. A rule based system would need a lot of rules to capture all three of these tickets, and output would need explicit error handling, which is time-consuming and difficult.

Why is NLP essential to uncovering insights from support tickets?

Customer support tickets are growing to be one of the main sources of customer data in companies. Support tickets can contain tickets raised in social media chats, email conversations and web-chat based conversations, and many companies receive them in the thousands everyday. They're also a rich, qualitative form of customer feedback, and each ticket contains information related to friction between the customer and your product, operations, and partners.

Support ticket conversations are, however, very raw, unstructured and unpredictable in length. Many of them are agent-customer conversations, some are complaints and each customer has a different ways of expressing themselves. Going through such large volumes of tickets requires considerable manual effort and, more often than not, important topics are missed. Even simplistic NLP (rule-based) struggles with support ticket analytics, missing insights due to the frequent presence of misspelled words and grammatically incorrect sentences.

Zendesk, the popular help desk platform, helps companies organize ticket requests, bringing them together on a single comprehensive dashboard. It makes use of a naive rule-based system to tag and categorize tickets. However, these tags are high level and coarse such as "website_issue", "presale_enquiry" and so on. They do not mention granular reason about the reason for customer contact. Any analysis still takes significant data handling to get accurate insights, and even then, they're likely to be high level or somewhat subjectiver.

Machine learning can bring something special to Zendesk: hierarchy. Hierarchy means that, when tickets are tagged, the relationship between ticket tags is noted. For example, lets consider a meal kit delivery company receiving a ticket from a customer about a missing ingredient. Zendesk's tagging system just tags this ticket as "objection raised" or perhaps, if you're lucky, "missing ingredient". But, it does not contain the finer details such as which ingredient is missing.

Zendesk ticket tagging hierarchy

Looking at an individual conversation, that's not a major issue because we can read that it's about rice. However, in the thousands of tickets, going through all of those in the "missing ingredient" pile every day becomes complex and time-consuming. A properly built machine learning system would tag the ticket with Box Contents--> Missing Ingredient --> Rice, which would allow you to choose the level you'd like to analyse at and dig deeper to the root cause when you need to.

This also goes for multiple topics. Most support tickets contain 2-3 complaints and a rule-based system may only apply one topic to it.

Without this level of granularity at scale, you have no way of understand which customer issues are generating the most support tickets. And, no way of knowing their business impact.

How our Zendesk Integration works

We at SentiSum, have built a robust Zendesk chat analytics integration using the latest developments in machine learning-based NLP. Our system will analyze and categorize high volumes of support tickets in a matter of minutes. Our ML-based system accurately uncovers multi-level topic tagging, including general (Box Contents), granular (Missing Ingredients), and specific aspects (Rice). On applicable tickets, like surveys and customer review data, we also uncover the customer sentiment toward the topic raised.

Let us take you through how this works:

  • Zendesk Integration with Sentisum platform: We first integrate your Zendesk account with our platform so that we gain access to your support tickets
  • Historical Data Exploration: Following the integration, we run your data through our exploration system which extracts important text words and phrases that are being mentioned in the support tickets. These words and phrases help us to identify granular insights that needs to be tagged.
  • Custom AI model training: Based on our historic data exploration, we curate a dataset with granular topics and train it with our robust NLP module.
  • Deploy, Categorize and we are live: Once a custom AI model is trained to tailor to your data, we automatically categorize all your customer conversations with rich insights that are made highly accessible and visual through our dashboard. Not only is your historical data organized, so we can tell what's normal and what's an anomaly, but any new support tickets generated are also tagged and categorized by our system.
  • Detecting anomalies: The system monitors the daily volume of each topic event and our custom anomaly detection module notifies you if there is an abnormally high number of mentions as compared to the number forecasted by our system.
  • Heavy lifting? It's all done by us. For you, this is a seamless one-click experience.
Zendesk support ticket tagging with NLP dashboard

Zendesk NLP-powered support ticket tagging use cases

Once you have NLP support ticket insights in an actionable format, you unlock a whole host of business benefits.

The first is the most obvious, you are able to share the insight into customer problems with the team who can change the problem's cause. Whether it's a product issue or an operations issue, passing them actionable insights gives them a chance to make amends.

Integration our NLP analytics into Zendesk also unlocks these use cases:

  1. If you are a franchise business (like the Domino's pizza example in the above) you can track which locations are letting the team down. Overall NPS and customer satisfaction are often diminished by a handful of stores or restaurants, our software will help you find them.
  2. If you use third-party vendors (for example, someone who delivers an e-commerce businesses parcels) then you can identify which delivery partner is creating a pain point in your customer's journey. Monitoring them and building evidence allows you to tackle the problem head on.
  3. Save customer service agent's time. This one is also quite an obvious one, but when you partner with us you'll remove any need for manual tagging, tag cleaning, monitoring, and creation of new tags. We do the maintenance for you.
  4. Real-time anomaly detection. A simple example of this would be something like a broken discount code. A sudden spike in tickets would be alerted to you, which could be fixed before many other users face the same problem. Consumers typically leave brands after two bad experiences, so being proactive is key to loyalty.
  5. Reduce support tickets. While solve the root cause of support tickets will, over time, reduce tickets, you can also work to deflect more tickets with our solution. Using our software, you can identify the most common questions and build your knowledgebase around them.
  6. Optimise conversion rates: Inform your optimisation team's roadmap by understanding the key drivers of customer friction. VoC tools help you uncover reasons for contact at a granular level, so you can prioritise the changes that get in the way of purchases.
  7. Ticket triage: the real-time nature of Zendesk NLP support ticket tagging means we can understand inbound tickets and direct them to the right team instantly to solve.
  8. There's probably more! And we're excited to find them.

NLP support ticket tagging

Want to learn more? We're always here to give you a demo of our product. Email Harry on harry@sentisum.com and we'll show you around.

Customer journey mapping screenshot

5 ways to sell the value of the contact centre internally—see pages 4-7.
Here are 7 ways companies are evolving their contact centre so it adds to the bottom-line—see page 21.

More from SentiSum