This post was co-authored with Herb Brittner of NetSertive.
NetSertive is a leading digital marketing solution provider for multi-location brands and franchises, helping businesses maximize local ads, improve engagement and gain deep customer insights.
With increasing demand in providing more actionable insights from customer call tracking data, NetSertive needs a solution that can unlock business intelligence from all calls, making it easier for franchises to improve customer service and increase conversion rates. The team was looking for a single flexible system that could do several things.
- Understand the phone – Automatically create a summary of what has been discussed
- Measure customer sentiment – Determine if the caller is happy, upset or neutral
- Identify important topics – Exploit keywords related to frequent services, questions, issues, and competitor mentions
- Improve agent performance – Provide coaching advice and suggestions
- Track performance over time – Generate reports on trends across individual locations, regions, and countries
Importantly, this new system had to run smoothly on the existing Multilocation Experience (MLX) platform. The MLX platform is specifically designed for businesses with many locations and helps you manage both national and local marketing. You can run campaigns on a variety of online channels, including search engines, social media, display ads, videos, connected TVs, online reviews, and manage SEO, business lists, reviews, social media posts, and individual location web pages.
In this post, we show how NetSertive can use Amazon Bedrock and Amazon Nova to deploy MLX-generated AI-powered assistants to MLX to enable the next generation of platforms.
Solution overview
Running a comprehensive digital marketing solution, NetSertive delivers key success metrics through Insights Manager products while handling campaign execution. The platform features location-specific content management and robust lead capture capabilities, collecting data from multiple sources, including paid campaigns, organic website traffic, and attribution proforms. With CRM integration and call tracking capabilities, MLX creates a seamless flow of customer data and marketing insights. This combination of managed services, automated tools, and analytics makes MLX a single source of truth for businesses looking to optimize their digital marketing efforts while leveraging NetSertive's expertise in campaign management. To address their desire to provide more actionable insights on the platform from customer call tracking data, NetSertive has looked at a variety of solutions. After evaluating various tools and models, I decided to use Amazon Bedrock and Amazon Nova Micro models. This choice was driven by Amazon Bedrock's API-driven approach, a wide selection of large-scale language models (LLMS), and performance of the Amazon Nova Micro model. They choose Amazon Nova Micro based on their ability to provide fast response times at a low cost, providing consistent and intelligent insights. With over 200 tokens per second generation speeds and extremely high-performing language comprehension skills, this text-only model has proven ideal for Netshathi. The following diagram shows how the MLX platform receives real-time calls, using Amazon Bedrock on Amazon Nova Micro to handle real-time calls.
The real-time call processing flow consists of the following steps:
- Once a call arrives, it will be routed to the lead API immediately. This process captures both the live call transcript and important metadata about the caller. The system continuously processes new calls upon arrival, facilitating real-time processing of incoming communications.
- The captured transcripts are transferred to Amazon Bedrock for analysis. The system currently uses standardized base prompts for all customers, and the architecture is designed to allow for quick customer-specific customizations as an additional layer of context.
- Amazon Nova Micro processes transcripts and returns a structured JSON response. This response includes multiple analytical components. Conversation sentiment analysis, a brief call summary, identified key terms, overall call theme classification, and specific coaching suggestions for improvement.
- All analysis results are systematically stored in an Amazon Aurora database with relevant key metrics. This ensures that the processed data is properly indexed and readily available for both immediate access and future analysis.
The Aggregated Report Schedule Flow consists of the following steps:
- The aggregation analysis process starts automatically on both weekly and monthly schedules. During each run, the system will invoke the appropriate data within a specified period of time.
- This aggregate analysis uses both Amazon Bedrock and Amazon Nova Micro to apply special prompts specifically designed for trend analysis. This prompt differs from real-time analysis, which focuses on identifying patterns and insights across multiple calls.
Processed aggregated data from both workflows is converted into a comprehensive report that displays trend analysis and comparison metrics via the UI. This allows stakeholders to provide valuable insight into performance patterns and trends over time, allowing users to dig deeper into specific metrics.
result
Implementing generation AI to create real-time call data analytics solutions has been a transformational journey for NetSertive. Their new Call Call Insights AI feature using Amazon Nova Micro with Amazon Bedrock takes only a few minutes to create actionable insights compared to the previous manual call review process. NetSertive chose Amazon Bedrock and Amazon Nova Micro for its solution after a quick evaluation period of about a week of testing different tools and models. Their development approach was systematic and customer-focused. Call Insights AI capabilities have been added to the platform's roadmap based on direct customer feedback and internal marketing expertise. From creating and testing the Amazon Nova Micro prompt to integrating Amazon Bedrock's MLX platform, the entire development process was completed within about 30 days and launched in beta. Transforming real-time call data analytics is not just about handling more calls, but about creating a more comprehensive understanding of customer interactions. By implementing Amazon Bedrock and Amazon Nova Micro, NetSertive can better understand the purpose and value of calls, enhance measurement capabilities, and advance towards more automated and efficient analytics systems. This evolution not only streamlines operations, but also provides customers with more actionable insights into digital marketing performance.
Conclusion
In this post, we shared how NetSertive introduced an AI-powered assistant with MLX generated AI using Amazon Bedrock and Amazon Nova. This solution helped us expand our MLX platform to provide instant, actionable insights to our customers, creating a more engaging and beneficial user experience. With Amazon Bedrock's advanced natural language processing power and high performance low latency Amazon Nova Micro Model, NetSertive has been able to build a comprehensive call intelligence system that goes beyond just transcription and sentiment analysis.
The success of this project demonstrates the potential for transformation of generated AI in promoting business intelligence and operational efficiency. For more information about building powerful, generic AI assistants and applications using Amazon Bedrock and Amazon Nova, see Generate AI in AWS.
About the author
Nicholas Switzer I am an AI/ML Specialist Solution Architect at Amazon Web Services. He joined AWS in 2022 and specializes in AI/ML, Generator AI, IoT, and Edge AI. He is based in the US and enjoys building intelligent products that improve daily life.
Jane Ridge Senior Solutions Architects at Amazon Web Services have over 20 years of technology experience. She joined AWS in 2020 and is based in the US. She is passionate about enabling customers to grow through innovative solutions combined with the deep technical expertise of the AWS ecosystem. She is known for her ability to guide customers through every stage of her cloud journey and provide impactful solutions.
Herbs and Britners He is Vice President of Product & Engineering at NetSertive and leads the development of AI-driven digital marketing solutions for multi-location brands and franchises. With a strong background in product innovation and scalable engineering, he specializes in using machine learning and cloud technology to drive business insights and customer engagement. Herb is passionate about building data-driven platforms that increase marketing performance and operational efficiency.
