This publish was collaborated with Mulay Ahmed, assistant engineering director, and Ruby Donald, assistant engineering director at Principal Monetary Group. The content material and opinions of this publish are third occasion authors and AWS will not be answerable for the content material or accuracy of this publish.
Principal Monetary Group® is an built-in international monetary companies firm with specialised options to assist individuals, companies and establishments obtain long-term monetary targets and entry larger monetary safety.
With our Contact Middle, which handles hundreds of thousands of buyer calls annually, Principal® needed to additional modernize the decision expertise for its clients. With a strong AWS cloud infrastructure already in place, we selected the cloud-first strategy to create a extra personalised, seamless expertise for our clients.
- Understanding buyer intent via pure language (versus touchtone expertise)
- We’ll assist our clients to offer self-service every time attainable
- Precisely route buyer calls primarily based on enterprise guidelines
- Use contextual information to help engagement middle brokers
Initially, the principal developed a Voice Digital Assistant (VA) utilizing Amazon Lex bots to acknowledge the shopper’s intent. VAs can carry out self-service transactions and route clients to particular name middle queues within the Genesys Cloud Contact Middle platform primarily based on buyer intent and enterprise guidelines.
As clients work together with the VA, it’s important to repeatedly monitor their well being and efficiency. This permits principals to establish alternatives for fine-tuning. This can improve VA’s means to grasp buyer intent. Subsequently, this reduces fallback intentional charges, improves practical intentional efficiency, and improves buyer expertise.
On this publish, the principal will use this chance to discover how you can construct an built-in voice VA reporting and analytics resolution utilizing Amazon Quicksight Dashboard.
Amazon Lex is a service that permits you to construct conversational interfaces utilizing voice and textual content. It gives prime quality speech recognition and language understanding capabilities, permitting you so as to add refined pure language chatbots to new and present purposes.
Genesys CloudOmni-Channel Orchestration and Buyer Relationship Platform gives a public cloud mannequin contact middle platform that permits fast and simple integration of AWS Contact Middle Intelligence (AWS CCI). As a part of AWS CCI, Genesys Cloud is built-in with Amazon Lex. This allows self-service, clever routing, and information assortment capabilities.
QuickSight is a unified enterprise intelligence (BI) service that makes it simple to construct visualizations, carry out ad-hoc analytics and shortly get enterprise insights out of your information.
Answer overview
Principals want a reporting and analytics resolution that screens VA efficiency primarily based on massive buyer interactions, permitting principals to enhance efficiency of their Amazon Lex bots.
Reporting necessities included customer-VA interactions and Amazon Lex bot efficiency (goal metrics and intent achievement) evaluation to establish and implement tuning and coaching alternatives.
This resolution used the QuickSight dashboard, which derives these insights from the next buyer interplay information used to measure VA efficiency:
- Cloud information comparable to genesys queues and information actions
- Enterprise-specific information comparable to product and name middle operation information
- Enterprise API-specific information and metrics comparable to API response codes
The next diagram illustrates the answer structure utilizing Genesys, Amazon Lex, and Quicksight.
The answer workflow contains the next steps:
- Customers name Genesys Cloud to work together.
- Genesys Cloud calls AWS Lambda routing capabilities. This operate returns a response to the Genesys cloud utilizing the required information to route buyer calls. To generate a response, this operate retrieves the routing information from an Amazon DynamoDB desk and requests an Amazon Lex V2 bot to offer a solution concerning the consumer’s intent.
- Amazon Lex V2 bots deal with buyer intent and name the Lambda Achievement characteristic to fulfill intent.
- The achievement operate performs customized logic (routing and session variable logic) and calls the required APIs to retrieve the info wanted to fulfill the intent.
- Returns the APIS course of and requested information (comparable to information that performs a self-service transaction).
- The Amazon Lex V2 Bot dialog logs are despatched to Amazon CloudWatch (these logs are used for enterprise analytics, operational monitoring, and alerts).
- Genesys Cloud calls the third Lambda operate to ship a buyer interplay report. The Genesys report operate pushes these reviews to an Amazon Easy Storage Service (Amazon S3) Bucket (these reviews are used for enterprise evaluation).
- Amazon Knowledge Firehose Supply Stream ships dialog logs from CloudWatch to S3 buckets.
- Firehose supply streams use Lambda capabilities to transform logs in Parquet or CSV format.
- AWS adhesive crawler scans Amazon S3 information.
- The crawler makes use of schema info to create or replace the AWS Glue information catalog.
- Use Amazon Athena to question your datasets (buyer interplay reviews and dialog logs).
- Quicksight connects to Athena and queries Amazon S3 information utilizing the info catalog.
Different design issues
Beneath are different vital design issues for implementing a VA resolution:
- Value Optimization – The answer makes use of Amazon S3 bucket keys to optimize prices.
- Encryption – The answer makes use of SSL/TLS to encrypt information at AWS KMS and relaxation throughout transportation.
- Genesys Cloud Integration – The mixing between Amazon Lex V2 Bot and Genesys Cloud is finished utilizing AWS ID and Entry Administration (IAM). For extra info, see Genesys Cloud.
- Logging and monitoring – The answer makes use of CloudWatch to watch AWS assets and makes use of alerts to obtain notifications when failing occasions.
- Least Privileged Entry – The answer makes use of IAM roles and insurance policies to grant the minimal privileges required to be used and repair.
- Knowledge Privateness – The answer processes customer-sensitive information, comparable to Personally Identifiable Data (PII) in accordance with compliance and information safety necessities. Implement information masking when applicable and applicable.
- A safe API – The APIs applied on this resolution are protected and designed in keeping with compliance and safety necessities.
- Knowledge Varieties – The answer defines information sorts comparable to timestamps within the information catalog (and Athena) to replace information (SPICE information) in QuickSight on schedule.
- DevOps – The answer is model managed, and modifications are deployed utilizing pipelines, leading to sooner launch cycles.
- Amazon Lex Evaluation – Amazon Lex analytics empowers groups with data-driven insights to enhance bot efficiency. The Overview Dashboard gives a single snapshot of key metrics comparable to the entire variety of conversations and intentional recognition charges. Principals don’t use this characteristic for the next causes:
- Dashboards can’t be built-in with exterior information.
- Genesys Cloud Knowledge (Queues, Knowledge Actions, and many others.)
- Enterprise-specific information (comparable to product or name middle operation information)
- Enterprise API-specific information and metrics (comparable to response codes)
- Dashboards can’t be built-in with exterior information.
- You can not customise your dashboard so as to add views or information.
Pattern dashboard
This reporting and analytics resolution permits principals to consolidate information from a number of sources, visualize VA efficiency and establish areas of alternative for enchancment. The next screenshot reveals an instance of a pace dashboard for illustrative functions.
Conclusion
On this publish, we introduced how the principal created a reporting and analytics resolution for a VA resolution utilizing Genesys Cloud and Amazon Lex, and the way Quicksight can present perception into interactions with clients.
VA options allowed principals to make use of Genesys Cloud to keep up their present contact centre options and supply a greater buyer expertise. It presents different advantages, comparable to the power for patrons to obtain assist on a number of inquiries with out requesting an agent for a name (self-service). It additionally gives clever routing capabilities, decreasing name occasions and growing agent productiveness.
This resolution implementation permits principals to watch and derive insights from the VA resolution and fine-tune efficiency accordingly.
On the 2025 roadmap, principals will proceed to strengthen the fundamentals of the options mentioned on this publish. The second publish reveals how principals can automate deployment and testing of recent Amazon Lex Bot variations.
AWS and Amazon should not associates of the Principal Monetary Group® firm. This communication is meant to be academic in nature and isn’t supposed to be thought-about a suggestion.
Insurance coverage merchandise issued by Principal Nationwide Life Insurance coverage Co (excluding NY) and Principal Life Insurance coverage Firm®. Plan the administration companies offered by Principal Life. Principal Fund, Inc. is a Principal Fund Distributor, Inc., Member SIPC, and/or Impartial Dealer/Supplier. It’s distributed by securities. The corporate being referenced is a member of the Principal Monetary Group® of IA 50392.
In regards to the writer
Mulay Ahmed He’s an assistant engineering director for Principals and is properly versed in architecting and implementing complicated enterprise-grade options within the AWS Cloud.
Ruby Donald He’s the Principal’s Engineering Assistant Director and leads the Enterprise Digital Assistant Engineering Group. She has in depth expertise in constructing and delivering software program on the enterprise scale.
