AWS Contact Center Intelligence: Post-Call Analytics
What is the AWS CCI Post-Call Analytics solution?
The AWS Contact Center Intelligence (CCI) solutions allow you to quickly and easily add intelligence to your existing contact center solution. The ultimate goal? Improve service while reducing cost.
As announced by AWS on the 18th of August 2020, AWS CCI solutions are focused on three stages of the contact center workflow: Self-Service, Live Call Analytics & Agent Assist, and Post-Call Analytics. At Lucy in the Cloud, we’ll be focusing on the latter. We’re proud to offer you this new solution as a trusted AWS Advanced Consulting Partner.
AWS Call-Center Intelligence: How it works
The AWS Contact-Center Intelligence Post-Call Analytics solution focuses on providing you with customer insights from previously recorded calls or chats. This will help your agents and supervisors to better understand the conversations they have with customer. Doing so, will allow you to find patterns and quality issues much faster than before. Ultimately you’ll be improving the overall customer experience. Which is a crucial element in the new digital age, being customer-centric.
Post-call speech analytics dashboards provide you with agent an operational performance statistics. Next to that it offers you with insights for managers, quality assurance personnel and other leadership groups.
Which AWS services are a part of our AWS CCI Post-Call Analytics solution?
Our experts use Amazon Transcribe for automatic speech recognition (ASR) to create high quality transcripts. Amazon Comprehend then applies Natural Language Processing (NLP) to said transcripts, which then allows you to analyze these interactions.
All of this enables you to obtain deeper quality management insights. It also allows you to generate actionable insights such as product and service feedback loops or the best performing interactions. These best performing interactions can, for example, be those ending with a positive sentiment score.
Amazon Transcribe is a speech to text service by AWS. It makes it easier for developers to add voice AI to their applications. Amazon Transcribe is designed to process audio input from a variety of sources (microphones, audio or video files, etc.). It provides you with high quality transcriptions that can be used for search and analysis.
Let’s go over the most interesting features for your call-center.
You can add new words to the base vocabulary. This allows you to generate much more accurate transcriptions for domain-specific words or phrases containing product names, technical terminology or names of individuals.
Custom Language models
You can build and train your very own custom language model (CLM). First you submit a corpus of text data to Amazon Transcribe. Second, that data the underlying speech recognition models will be used by Amazon Transcribe to generate a CLM tailored to your specific use case and domain.
Why would I want a CLM? It can be a desirable feature for enhancing speech recognition accuracy. Especially if you have large amounts of text data in a certain domain that matches your audio data. This can include archived transcribed logs of call center interactions, as well as subtitles videos, customers’ websites, and many other data sources.
Want to learn more about this special feature? Click here.
Amazon Comprehend is a natural language processing (NLP) service. It uses machine learning to discover insights from text. Amazon Comprehend can also provide you with: keyphrase extraction, sentiment analysis, entity recognition and topic modeling.
Let’s dive a little deeper into those features!
The Keyphrase Extraction API returns the key phrases or talking points and a confidence score to support that this is a key phrase.
The Sentiment Analysis API returns the overall sentiment of a text (Positive, Negative, Neutral, or Mixed). For example: a customer is posting his feedback on a pair of shoes. The API identifies the sentiment expressed by the customer along with a confidence score.
The Entity Recognition API returns the named entities (“People,” “Places,” “Locations,” etc.) that are automatically categorized based on the provided text.
Custom Entities allows you to customize Amazon Comprehend to identify terms that are specific to your domain. Using AutoML, Comprehend will learn from a small private index of examples (for example, a list of policy numbers and text in which they are used), and then train a private, custom model to recognize these terms in any other block of text. There are no servers to manage, and no algorithms to master.
The Custom Classification API enables you to easily build custom text classification models using your business-specific labels without learning ML. For example, your customer support organization can use Custom Classification to automatically categorize inbound requests by problem type. This will be done based on how the customer has described the issue.
Creating a custom model is simple. You provide examples of text for each of the labels you want to use, and Comprehend trains on those to create your custom model. No machine learning experience required, you can build your custom model without using a single line of code. An SDK is available for you to integrate your customer classifier into your current applications. With your custom model, it is easy to moderate website comments, triage customer feedback, and organize work-group documents.
What does this new AWS CCI Post Call Analytics solution mean for you?
Nos experts utilisent Amazon Transcribe pour la reconnaissance automatique de la parole (ASR) afin de créer des transcriptions de haute qualité. Amazon Comprehend applique ensuite le traitement du langage naturel (NLP) auxdites transcriptions, ce qui vous permet ensuite d’analyser ces interactions.
Tout cela vous permet d’obtenir des informations plus approfondies sur la gestion de la qualité. Il vous permet également de générer des informations exploitables, telles que les boucles de rétroaction sur les produits et services ou les interactions les plus performantes. Ces interactions les plus performantes peuvent, par exemple, être celles qui se terminent par un score de sentiment positif.
A Great Customer Experience
Providing a great customer experience boosts repurchase odds and long-term loyalty, while poor customer service experiences increase costs and leads to customer defection. Understanding your customer, is a key part of that. To make sure you’re getting all out of the data your call-center generates, we’re now offering the AWS Contact Center Intelligence Post-Call Analytics solution*. Main goal? It helps you keep track of each customer’s experience. Thus, allowing you to not only tap into real-time analytics but truly capture the customer’s sentiment. Next step? Play into it and improve the customer”s experience.
*This solution is not a one fits all approach. It will be tailored to your business’ specific needs.
By automating the repetitive and time-consuming tasks, you free up your agent’s time to allow them to do what they do best: help a customer in need. The use of AWS dedicated services such as Amazon Comprehend and Amazon Transcribe, allow you to lower operational costs and agent’s turnover. After all, providing excellent customer service at reduced cost is what the AWS Contact Center Intelligence Post-Call Analytics solution is all about.
Discover how Lucy’s experts implemented this new solution for one of Belgium’s biggest Healthcare Insurance provider’s call-center ‘Genesis’ in our most recent success story.