GoTo Revolutionizes Contact Center Quality Management with AI

Kendra Rasmusson

About GoTo

GoTo Technologies USA, Inc. is a flexible-work provider of software as a service and cloud-based remote work tools for collaboration and IT management.

GoTo Connect Contact Center is a cloud-based solution for managing customer interactions across voice, email, chat, and social media. It offers features like call routing, queuing, and analytics, integrates with CRM systems, and provides tools for agents to improve efficiency. It’s part of GoTo’s suite for unified communications, aimed at enhancing customer service and streamlining operations.

GoTo’s Challenge

GoTo had a goal to to grow its workforce engagement management (WEM) offerings within the Contact Center product. Quality management was identified as a key opportunity to differentiate and satisfy demand from the target market of small to mid-sized businesses.

Industry best practices support the notion of reviewing and scoring contact center transcripts to ensure service quality, improve agent performance, and enhance customer satisfaction. This process helps identify training needs, ensures compliance with regulations, and provides insights into operational efficiency. By analyzing interactions, an organization can refine processes, measure performance, and maintain high standards, ultimately leading to better customer experiences and more effective contact center operations.

Historically, GoTo’s small-to-medium business customers who utilized GoTo Connect have not been able to effectively monitor and coach their contact center agents due to the time and manual effort required (an estimated 30 minutes per call transcript). In fact, on average a mere 3% of all contact center calls receive a performance evaluation.

Some of GoTo’s competitors have attempted to address this need in varying ways, but GoTo surmised that GenAI could be applied to produce more impactful outcomes for customers.

“For us, GenAI was the chance to leapfrog our competitors AND meet our customers along their maturity journey ,” said Gina Whitty, GoTo’s Director of Product Management.

Why Tribe AI?

GoTo was referred to Tribe AI by their trusted advisors at the Francisco Partners private equity firm. As a leading global investment firm specializing in partnerships with technology and tech-enabled businesses, Francisco Partners knew Tribe AI’s team was composed of GenAI experts who were both highly-skilled technicians and business-savvy advisors.

Developing the Roadmap

“Our use case was completely user-driven. Our customers have been looking for this kind of functionality and with Tribe AI’s help, we were able to understand what was possible with GenAI,” said Whitty. 

GoTo has long been a leader in the Business Communications and IT Support Software industry. Starting as a pioneer in the video conferencing world, and now innovating the contact center space, their customer-centered focus is often credited for their organizational successes.

GoTo had clearly-established goals for the final product. During the four week POC engagement, Tribe AI and GoTo worked together to identify and target the riskiest aspects of the project and aim to achieve high-accuracy results to ensure a smooth transition to production.

Final Product Goals:

  • Further Differentiate GoTo Connect Contact Center:
    Provide quality management capabilities to customers, who are eager for them, but use AI to allow for improved output over competition.
  • Improve Agent Performance:
    Lean on AI to access the metrics necessary to measure percentage change in agent performance over time, so customers can better track contact center performance. 
  • Increase Customer Satisfaction:
    Use AI to track the correlation between call score metrics and customer satisfaction metrics over time.
  • More Efficient Evaluation:
    Measure the time saved in the AI-powered evaluation process over previous manual methods. 
  • Increase Evaluation Coverage:
    Use AI to evaluate ALL calls, improving the previous call evaluation average rate (3%). 

“We know the best customer service outcomes are realized when call quality is monitored. Providing this solution – based on industry best practices – to our customers allows for smooth, easy adoption and elevates the entire customer experience,” said Whitty.

Whitty believes that launching the AI-powered quality metrics scorecard will dramatically improve present call evaluation rates and provide customers with a new sightline into contact center performance, spurring a continuous improvement loop involving opportunities for coaching and training. 

Proposed Solution

Tribe AI proposed beginning a four week engagement centered on building an AI-powered quality metrics scorecard that could be incorporated into the client’s GoTo Connect Contact Center software. The solution would analyze customer service calls and assess the performance of the customer service representative on the line.  The calls would be measured against a rubric, established based on industry best practices. 

How it works:
Call recordings are converted from audio to text using AI-based services then a custom-designed prompt and evaluation criteria are applied to score the transcripts. The output is formatted in JSON data that can be integrated into the GoTo Connect contact center software to provide the quality metrics scorecard. 

Iterations to improve output:
The team validated the feasibility of the approach by comparing the AI-generated scores to manual human review. They were able to first demonstrate a reasonable level of accuracy, and then focused next on iterating and improving the model's performance on specific criteria that were initially scored with less precision. Over several iterations, the teams made adjustments in how the data was transformed and evaluated in addition to prompt and workflow modifications to, in the end, demonstrate the possible gains in the level of accuracy. 

Tech Stack Details

Full-stack cloud-based application working alongside the existing GoTo environment: 

  • Cloud: Amazon Web Services (AWS) - Bedrock
  • Large language models: Anthropic Claude Models
  • Languages used: Python

GoTo’s Experience Working with Tribe

“Tribe provided more than an AI technical expertise, they also demonstrated seasoned product management skills. They paid extra attention to ensuring the output was high-quality and accurate, because they understood how important that was to our team,” said Whitty. 

Aside from the obvious task of validating the use case, Whitty applauds the Tribe AI team for their advanced knowledge and capabilities surrounding the implementation of a broad range of GenAI strategies. The internal GoTo team included many skilled people focused on software, data, product, and tech but didn’t have first-hand day-to-day experience with GenAI like Tribe AI did. Whitty believes that the success of the POC engagement was due, in part, to the teams’ abilities to collaborate. GoTo knew their own platforms, their customers, and their industry better than anyone, while Tribe AI had the knowledge of a multitude of architectures, models, and approaches to successfully applying GenAI in various use cases. Together the teams were able to validate the use case, and remove the risk associated with its future development. 

“Understanding the intricacies of the technology is where Tribe AI shines. They were able to demonstrate which models and architecture would most efficiently produce the best outcomes in our use case and that’s invaluable,” said Whitty. 

Tribe Team Members

Will: Client Lead

Nik: Product Lead

Prem: Technical Lead

Impact

By the completion of the four week POC engagement, Tribe AI was able to validate the hypothesis that GenAI could successfully produce a tool that with efficiency and accuracy could analyze and score customer calls based on a predefined rubric. The key measure of success was that, after several iterations, the tool demonstrated the feasibility and path to an accuracy level equal to or better than the level reached through manual human review.  Additionally the AI-powered solution delivered a scorecard in just 10 seconds, showcasing a 95% reduction in time over manual evaluation methods.

“The POC engagement with Tribe AI essentially de-risked the project for our team. We gained valuable learnings and guidance that informed our next steps, leading all the way to development,” said Whitty. 

GoTo’s  AI-powered quality metrics scorecard is in beta now and is expected to fully launch in early 2025.  The functionality will be incorporated into their existing platform and offered as an optional add-on that customers can purchase. 

The Future

“In the end, we are our customer’s customers. Innovations to the customer experience impact us all,” said Whitty.

Whitty believes the future of the customer experience is built for and by user-driven insights. 

Although the initial call scoring rubric utilized in the POC engagement was informed by industry best-practices, Whitty envisions a future where each client has the ability to customize the scoring process further, based on the metrics most important to their organization. 

“For the late adopters, I can see how AI might appear intimidating, but together with Tribe AI we were quickly able to uncover how we could leverage it to meet our customers’ needs in a really smart and efficient way,” said Whitty.

Related Case Studies

Case Study

How Tribe Helped Reservoir Bring Finance Infrastructure to NFT Trading

Case Study

Native Instruments Leverages Amazon Bedrock for Smarter, More Intuitive Search and Discovery for Music Creators

Case Study

How Nota Built a Roadmap for AI-enabled Journalism with Help from Tribe

Case Study

Tribe AI & Venture Labs: Accelerating Startups with Tailored AI Expertise

Case Study

How Wingspan built a machine learning roadmap with Tribe AI

Case Study

Building a GenAI Roadmap for Educational Content Creation

Case Study

Francisco Partners Accelerates Portfolio AI Efforts with Tribe AI

Case Study

How Tribe AI Built a Model on GCP That Increased Security Questionnaire Auditor Efficiency by 55%

Case Study

How Togal AI Built the World's Fastest Estimation Software on AWS

Case Study

Boomi Leverages Amazon Bedrock for Faster Help Desk Responses

Case Study

Insurance Company Uses ML to Optimize Pricing

Case Study

VitalSource Leans on GenAI to Reimagine Content Discoverability for Higher Ed Faculty

Case Study

How Tribe AI Shaped Truebit’s AI Strategy

Case Study

Building a Proprietary Investment Engine Using Public Data for a Top PE Firm

Case Study

Kettle uses machine learning to balance risk in a changing climate

Case Study

Togal.ai powers the construction industry into the age of machine learning

Case Study

GenAI Solutions: How Bright Transformed Workforce Training with Tribe AI

Case Study

How Fantasmo is using machine learning to make GPS obsolete

Case Study

Taking a Data-Driven Approach to Women's Fertility with Rita

Tribe helps organizations rapidly deploy AI solutions that have real business impact.

Close
Kendra Rasmusson