Amazon Connect Wisdom brings disparate data together to quickly provide the information your service agents need
Key Takeaways:
- Consumers expect speedy and personalized customer service
- Wisdom uses ML to search integrated databases using plain language
- Real-time analytics are available via contact lens
- Wisdom learns over time
- Wisdom is part of Amazon Connect cloud services and makes it even smarter
Today’s consumers expect speedy and personalized customer service, and they don’t want to be left hanging on the line or waiting on chat while an agent searches for an answer or missing info. Currently in preview, Amazon Connect Wisdom (Wisdom) leverages machine learning to significantly reduce the time spent by customer service agents searching for answers in siloed databases. This allows agents to use AI to search an integrated database with queries based on precisely how a customer would phrase the question.
A recent report reveals that 63% of buyers are loyal to brands with superlative customer service and 62% will recommend that brand to friends and family. Two bad customer service experiences will see 45% of respondents abandon that brand forever. This makes it imperative to arm agents with everything they need to provide the best experience.
Wisdom plugs into Contact Lens for Amazon Connect and uses real-time speech analytics to auto-detect any issues during phone calls and recommend relevant information in real-time. This not only helps resolve the issue quickly, but also eliminates the need for lengthy manual searches.
The machine learning improves its content recommendations over time as agents rate search and recommendation results. Wisdom also has built-in connectors for Salesforce and ServiceNow, among other third-party applications.
Real-time insights into the customer experience
By using Contact Lens for Amazon Connect, a set of machine learning capabilities, supervisors can better understand customer interactions to train agents, replicate successful experiences, and gather and analyze product and company feedback.
Supervisors can easily identify when they should listen in and provide guidance to the agent. Insights are gathered during live calls and delivered to the Amazon Connect real-time metrics dashboard, so supervisors know when agents need some guidance or they should speak with the customer directly.
Rules can be defined with specific terms, such as “unhappy,” “horrible quality,” and “cancel,” using natural language processing (NLP), which uses intelligent matching to automatically detect variations in the words spoken, even when there are limited example phrases.
Wisdom is designed to help your agents provide the ultimate in customer satisfaction via:
- Machine-learning-powered knowledge search: When an agent enters a question or phrase such as, “how long is the warranty on this mp3 player?” Wisdom will search all information repositories and give the agent the most pertinent information and best answer to resolve the customer’s issue.
- Customer recommendations in real-time: By using real-time call transcripts from Contact Lens for Amazon Connect, Wisdom can auto-detect customer issues during calls and give recommendations to relevant content in all connected knowledge bases.
- Continual improvement: Because agents can rate the relevance and effectiveness of recommendations, Wisdom learns over time. Knowledge articles can be improved and updated using agent-input comments, as well.
Wisdom makes Amazon Connect even smarter
Wisdom is one of the latest tools in the Amazon Connect cloud arsenal that includes Amazon Connect Voice ID, Amazon Connect Customer Profiles, and Amazon Connect Tasks.
Amazon Connect Voice ID
Amazon Connect Voice ID, also in preview, provides real-time caller authentication to make customer service interactions efficient and more secure.
After the agent gets the consent of the customer, Voice ID first authenticates their voice by capturing 30 seconds of speech and creating a voiceprint. This voiceprint captures rhythm, pitch, loudness, and intonation as a mathematical representation and provides an API.
On the next call, Voice ID uses 10 seconds of the customer’s voice to authenticate them and sends an authentication result in real-time to the agent, who can then decide to keep going with the call or ask for more authentication credentials.
Amazon Connect Customer Profiles
Amazon Connect Customer Profiles brings together customer information from different sources to provide a unified customer profile to agents and IVR systems, which boosts your ability to deliver better, personalized service to customers and achieve faster call resolution.
Integration with third-party systems happens with just a few clicks using pre-built connectors enabled by Amazon AppFlow to capture data stored in applications like Salesforce, ServiceNow, Zendesk, Marketo, and others.
When Customer Profiles is enabled, it automatically detects customer records from the third-party application, then matches and duplicates them. This means customer information is always accurate and up-to-date.
Amazon Connect Tasks
Customer service center managers can automate, track, and manage agent tasks by prioritizing, assigning, and tracking them across all of the applications used by agents to ensure they are focusing on the highest priority work.
Tasks can be created programmatically in third-party applications, like Salesforce, business-specific software. Agents no longer have to switch from application to application with Amazon Connect Tasks, because they will see all their assigned tasks on the control panel, which is the same one they use for call and chat interactions. Agents can also create their own tasks to make sure things like customer follow-up calls are not forgotten.
You can submit a request to preview Amazon Connect Wisdom to provide agents the content they need to personalize and satisfy customer experience expectations.
From cloud consulting to managed services and beyond, contact the CloudHesive team today to learn how we can help you build a robust cloud strategy that increases operational efficiencies.
