The main financial group increases the performance of virtual sound assistant using Geneesys, Amazon Lex and Amazon Quicksight

This post was tired of Mulay Ahmed, Assistant Director of Engineering and Ruby Donald, Assistant Director of Engineering in the main financial group. The content and thoughts in this post are those of the author of the third party and AWS are not responsible for the content or accuracy of this post.

Principal Financial Group® is an integrated global financial services company with specialized solutions by helping people, businesses and institutions achieve their long -term financial goals and enter into greater financial security.

With American contact centers dealing with millions of customer calls each year, Principal® wanted to further modernize their customer call experience. With a strong AWS cloud infrastructure already in the country, they chose a first approach to the cloud to create a more personalized and quiet experience for their customers that will:

• Understand the client’s goals through the natural language (Vs the touch tone experiences)
• Help customers with self-service offers when possible
• Require customer calls based on business rules
• Help agents of the engagement center with contextual data

Initially, the director developed a virtual virtual assistant (VA) using a Amazon Lex bot to know the client’s goals. VA can perform self-service transactions or direct clients in specific rows of call centers on the Geneesys-Cloud contact center platform, based on client goals and business rules.

While clients interact with VA, it is essential to constantly monitor its health and performance. This allows the director to identify the opportunities for good adjustment, which can improve VA’s ability to understand the client’s goals. Consequently, this will reduce the goal rates of the return, improve the levels of fulfillment of the functional purpose and lead to better customer experiences.

In this post, we consider how the director seized this opportunity to build an integrated VA and analytical reporting solution using an Amazon Quicksight dashboard.

Amazon Lex is a service for the construction of conversational interfaces using sound and text. It ensures high quality word recognition and ability to understand language, enabling the addition of sophisticated chatbots, in natural languages ​​to new and existing applications.

Geneesys Cloud, an OMNI-Channel orchestration and client relations platform, offers a contact center platform in a public cloud model that enables rapid and simple intelligence intelligence of the AWS Center (AWS CCI). As part of AWS CCI, Geneesys Cloud is integrated with Amazon Lex, which enables self-service, intelligent course and data collection skills.

QuickSight is a unified business intelligence service (BI) that makes it direct within an organization to build visualizations, perform ad hoc analysis and quickly get business knowledge from their data.

Settlement

The director required a reporting and analytical solution that would monitor VA performance based on customer interactions on the scale, enabling the director to improve Amazon Lex’s World Performance.

Reporting requirements included customer interaction and VA and Amazon Lex Bot performance (target metrics and fulfillment of purpose) to identify and implement allocation and training opportunities.

The solution used a fast dashboard that derives this knowledge from the following client interaction data used to measure VA performance:

• Cloud data of Geneesys both ranks and data actions
• Business specific data such as product operations and call centers
• Data and metric specific for business APIs such as API response codes

The following diagram shows the solution architecture using genesys, Amazon Lex and Quicksight.

The workflow of the solution includes the following steps:

1. Users call and interact with Cloud Genesys.
2. Geneesys Cloud calls a function of the AWS Lambda course. This feature will return a response to Geneesys Cloud with the necessary data to direct the client’s call. To generate a response, the function receives the course data from an Amazon Dynamodb chart, and requires an Amazon Lex V2 bot to give a response to the user’s purpose.
3. Amazon Lex V2 Bot elaborates on the purpose of the client and calls it a function of fulfilling Lambda to fulfill the purpose.
4. The fulfillment function executes custom logic (the logic of the course variables and session) and calls the API needed to obtain the data required to fulfill the purpose.
5. The process of APIs and the return of the required data (such as data to perform a self-service transaction).
6. Amazon Lex V2 bot records are sent to Amazon Cloudwatch (these articles will be used for business analytics, operational monitoring and alarms).
7. Geneesys Cloud calls a third lambda function to send client interaction reports. The Generasys report posts these reports into a bucket of simple Amazon storage service (Amazon S3) (these reports will be used for business analytics).
8. A stream of data distribution of data in Amazon sends cloonwatch conversation records to a S3 bucket.
9. Fire distribution flow transforms logs into parquest or CSV format using a lambda function.
10. An AWS dragging scanning data in Amazon S3.
11. The trailing creates or updates the AWS GLUE data catalog with the scheme information.
12. We use Amazon Athena to search for data data (client interaction reports and conversations logs).
13. QuickSight relates to Athena to search for data from Amazon S3 using the data catalog.

Other design considerations

Below are the other major design considerations to implement the VA solution:

• Optimization – Solution uses Amazon S3 bucket keys to optimize costs:
• Encryption – Solution encodes data on rest with AWS KMS and in transit using SSL/TLS.
• The integration of cloud genesys – Integration between Amazon Lex V2 Bot and Geneesys Cloud is done using AWS identity and access management (IAM). For more detail, see genesys cloud.
• Registration and Monitoring – The solution monitors AWS resources with cloudwatch and uses alerts to receive notification of failure events.
• The smallest privilege approach – The solution uses IAM roles and policies to give the minimum permits needed for uses and services.
• Privacy of data – The solution deals with customer sensitive data such as personally identifiable information (PII) according to compliance and data protection requirements. Implements data masking when applicable and appropriate.
• Api – API implemented in this solution are protected and designed according to compliance and safety requirements.
• Types of data – The solution determines the types of data, such as the stamps of time, in the data catalog (and Athena) in order to refresh the data (spice data) in Quicksight on a schedule.
• digress – The solution is controlled by the version, and the changes are set using pipes to enable the fastest release cycles.
• Analytics in Amazon Lex -Analytical in Amazon Lex empowers teams with knowledge directed from data to improve their world performance. The summary panel provides a single photograph of the main metrics such as the total number of conversations and levels of purpose recognition. The director does not use this skill because of the following reasons:
  • The panel cannot be integrated with external data:
    • Cloud Genesys data (such as ranks and data actions)
    • Business specific data (such as product center and call and call operations)
    • Data and metric specific for business APIs (such as response codes)
• The dashboard cannot be customized to add additional views and data.

Sample

With this reporting and analytical solution, the director can consolidate data from numerous sources and visualize VA performance to identify areas of improvement opportunities. The following appearance of the screen shows an example of their quick dashboard for illustrative purposes.

cONcluSiON

In this post, we presented how the director created an analytical report and solution for their resolution VA using Geneesys Cloud and Amazon Lex, along with Quicksight to ensure knowledge of client interaction.

The VA solution allowed the director to maintain his existing resolution of the Cloud Genesys contact center and to achieve better customer experiences. It offers other benefits such as the ability for a client to receive support in some questions without asking for a self-service agent. It also provides intelligent course skills, leading to reducing call time and increasing agents’ productivity.

With the implementation of this solution, the director can monitor and derive knowledge from his solution and adjust well, accordingly its performance.

On its 2025 map, the director will continue to strengthen the foundation of the solution described in the post. In a second post, the director will present how they automate the placement and testing of the new versions of Amazon Lex bot.

AWS and Amazon are not collaborators of any company in the main financial group®. This communication is intended to be a natural educational and is not intended to be taken as a recommendation.

Insurance products issued by the National National Life Insurance National (except at NY) and the leading life insurance company. Plan the administrative services offered by the main life. The main funds, INC are distributed by the main distributor of funds, INC securities offered through Prince Securities, Inc., SIPC Member and/or Independent Trade/Traders. Referred companies are members of the main financial group®, des Moines, IA 50392. © 2025 Main Financial Services, Inc. 4373397-042025

About

Mulay Ahmed He is an assistant Director of Engineering in the Director and capable of the architecture and implementation of complex Enterprise class solutions at AWS Cloud.

Ruby Donald He is an assistant Director of Engineering in the Director and runs the Virtual Enterprise Assistant Engineering Team. It has extensive experience in building and distributing software on the scale of enterprises.