Add customer sentiment and conversation characteristics to your AWS Contact Lens analytics
The benefits offered through Contact Lens for Amazon Connect can’t be disputed. Real-time and post-call analytics provide valuable information straight from the customer’s mouth. Contact Lens performs sentiment analysis, detects issues, and enables you to automatically categorize contacts. This means you can analyze conversations between customers and agents using tools such as speech transcription, natural language processing, and intelligent search capabilities.
But that’s not all. With Amazon language artificial intelligence (AI), you can discover hidden insights from recorded conversations that help manage agent script compliance as well as find new opportunities to engage and satisfy your customers. With the information gleaned from buyers, you can modify your service offerings to address any gaps, improve areas where there are problems, and elevate the customer experience to new heights.
With Amazon Machine Learning (ML) services like Amazon Transcribe Call Analytics and Amazon Comprehend, you can transcribe and extract insights from your contact center audio recordings at any scale. You can gain valuable intelligence that includes customer sentiment and conversation characteristics. Customer sentiment analysis is essential to learn the different emotions that your customers are having while discussing your products and services. Add in conversation characteristics and call analytics can provide some insights that might surprise you. This new insight can help you offer more of what your customers say makes them happy. Let’s take a look.
Sentiment analysis and conversation characteristics in call analytics
First, let’s talk about what customer sentiment analysis is. It is the automatic detection of emotions when your customers have dealings with your products, services, or brand. The AWS Post Call Analytics (PCA) solution provides actionable insights so you can spot emerging trends, identify coaching opportunities, and assess the general sentiment and speech characteristics of calls.
Conversation characteristics include things like non-talk time, interruptions, loudness, and talk speed. All of this happens via built-in ML models that have been trained with thousands of hours of conversations. The automated call categorization capability means you can tag conversations based on specified keywords, phrases, sentiment, and non-talk time. Customer privacy can be protected by redacting information such as name, address, and credit card information.
The business benefits of call analytics using sentiment analysis
Customer sentiment analysis algorithms primarily use polarity – are the sentiments positive or negative? – and magnitude – which indicates how strong the customer’s emotions are. The insight gained just through these two parameters can provide a boost to the bottom line through:
- Improved customer service leads to more profits: When deciding whether to do business with a company, 90% of U.S. buyers use customer service as a factor, and 58% will switch who they buy from because of poor customer service. On the flip side, 93% of customers will make repeat purchases from a company with excellent customer service, and 94% will recommend the business to friends and family as well as in online reviews. It’s also worth noting that increasing customer retention rates by a mere 5% can lift profits 25% to 95%.
- Your marketing campaigns have better ROI: Selling your products or services profitably depends largely on giving the customer what they want. Sentiment analysis provides the call analytics you need to know how customers feel about your products and see how you stack up to competitors. This is invaluable insight for your marketing team, who can use it to improve campaign results.
- You’ll make better products and provide better services: With sentiment analysis, you have the power to anticipate industry trends as well as gain new market insights. If a product release didn’t go well, you’ll have the information you need to understand why. These are all essential insights for new product development and existing product upgrades.
There’s a goldmine in call analytics using recorded customer calls. With Contact Lens and post call analytics, you can excavate the details that improve the customer experience, which is essential to improving your bottom line.
Detailed call analytics made easy
This post call analytics solution has a home page showing all of your calls. To see the details of a call, you simply choose a particular record and scroll down to see annotated call details, turn-by-turn. Search call transcripts based on sentiment characteristics, dates, or entities. You can even query detailed call analytics from your business intelligence tool.
PCA currently has the following features:
Call transcription
- Batch turn-by-turn transcription with support for Amazon Transcribe custom vocabulary
- You can redact personally identifiable information from transcripts and audio files via Amazon Comprehend and use vocabulary filtering to mask custom words and phrases.
- Multiple languages and automatic language detection
- Standard audio file formats
- Caller and agent speaker labels using channel identification or speaker diarization, which labels each speaker utterance
Call analytics
- Get caller and agent sentiment details and trends
- Talk and non-talk time for both caller and agent
- You can configure Transcribe Call Analytics categories based on the presence or absence of keywords or phrases, sentiments, and non-talk time
- Use built-in call summarization models in Transcribe Call Analytics to detect a caller’s main issues, action items, and outcomes
- Use Amazon Comprehend standard or custom entity detection models or simple configurable string matching to discover entities referenced in the call
- Detect when the caller and agent interrupt each other
- Gauge speaker loudness
Call search
- You can search call attributes such as the time range, sentiment, or entities
Other valuable features
PCA also detects metadata from audio file names, such as call GUID, agent’s name, and call date time. It also scales automatically to handle variable call volumes and bulk loads large archives of older recordings while keeping the capacity to process new recordings.
PCA is deployed using AWS CloudFormation.
The importance of robust post-call analytics can’t be overstated. You might think you know what the customer wants, but needs and wants change over time. Today’s completely connected experience should not overlook the human aspect. Hyper-personalization is in overdrive, and that means more than just using the customer’s name in an email or chatbot. The sentiments your customers convey, as well as how they convey them, will be the fuel that powers your business future.
Get the sentiment data you need to power profitability
It’s time to make the most of your valuable business intelligence. Give your customers what they want, and they’ll keep coming back for more.
CloudHesive is a cloud solutions consulting and managed service provider with expertise in all things Amazon Web Services. We have eight AWS Competencies, more than 50 AWS Certifications, membership in nine Partner Programs, and the experience and knowledge to help your business realize all the benefits of AWS cloud. We’ve helped more than 100 companies reduce their operating costs and increase productivity with our focus on security, reliability, availability, and scalability. With over 30 years of experience, we leverage cloud-based technology to its full potential. Contact the CloudHesive team today.
