Mindex uses Amazon Q in QuickSight to democratize analytics and drive student success in education

Mindex is a leading provider of enterprise software development and cloud services. One of Mindex’s products is SchoolTool, a student management system for the New York State school district. SchoolTool’s advanced analytics platform offers various dashboards allowing administrators to unlock invaluable data insights for districts and drive systematic improvements in student outcomes.

At Mindex, we believe education technology should empower educators and students, not overwhelm them with complexity. Our SchoolTool platform aims to put data-driven insights directly into the hands of teachers, administrators, and learners through intuitive, visual analytics powered by Amazon QuickSight.

In today’s data-rich educational environments, the ability to glean actionable insights shouldn’t be limited to technical analysts. By embedding QuickSight dashboards and visualizations directly within SchoolTool, we are democratizing advanced analytics capabilities. Now, non-technical users can easily explore up-to-date data from across their academic ecosystem, spotting important trends and making informed decisions, all without needing to learn complex business intelligence (BI) tools.

Our goal is to facilitate better student outcomes by surfacing key metrics and patterns using intuitive, interactive dashboards and natural language queries. With SchoolTool’s seamless integration of QuickSight, teachers can get guidance on intervention strategies tailored to each student’s needs. Administrators gain visibility into program effectiveness across the institution. And parents can easily monitor their children’s progress through the Parent Portal App.

In this post, we’ll share how Mindex partnered with QuickSight to enhance speed and experience and offer advanced analytics to more than 410 school districts through the SchoolTool Platform. We’ll explore how Mindex used QuickSight to build efficient dashboards and used Amazon Q in QuickSight, harnessing the power of generative artificial intelligence (AI) to make dashboard developers more efficient.

Why QuickSight?

Before adopting QuickSight, Mindex faced significant technical challenges with our previous business intelligence platform. Our home-grown solution, built using a mix of .NET and various libraries, was labor-intensive and cumbersome. Developing and deploying dashboards required substantial effort, and the platform didn’t allow customers to create and share their own dashboards. Instead, customers could only access pre-created dashboards, limiting their ability to use data fully.

Mindex chose QuickSight for several compelling reasons:

1. Efficiency and enhanced capabilities: Updating datasets and visualizations on the old platform involved multiple steps and extended-release cycles. With QuickSight, updates and changes became more efficient, enabling Mindex to adapt swiftly to evolving state requirements. With QuickSight, Mindex was able to roll out modern, interactive dashboards, pixel-perfect reports, embedded analytics, and natural language queries. These advanced features significantly improved the user experience and the overall value of Mindex’s offerings.
2. Ease of integration and embedding: Using a modern cloud platform like AWS provided Mindex with the flexibility to seamlessly integrate third-party data into their datasets. The ease of embedding and integrating QuickSight with Mindex’s applications was a significant advantage. Unlike other solutions that posed challenges for programmatic control and pipeline-driven processes, QuickSight allowed Mindex to approach BI development similarly to traditional software development, streamlining Mindex’s workflow and enhancing our BI capabilities.
3. Cost optimization: Using QuickSight allowed Mindex to retire old dashboards based on SQL Server databases, resulting in a 10% annual cost reduction across all school districts. These significant savings are being reinvested into more advanced features including generative AI capabilities within the SchoolTool product.
4. Advanced data integration and customization: QuickSight facilitates seamless integration of third-party data, offering a comprehensive view for each Regional Information Center (RIC) with access to data from all districts within their purview. This setup empowers RICs to analyze and use data in new ways, such as comparing multiple districts and exploring different perspectives. Additionally, QuickSight allows users at the RIC, Board of Cooperative Educational Services (BOCES), or district levels to create customized dashboards tailored to their needs, which was not achievable with the previous custom JavaScript implementation.
5. Generative AI technology: Amazon Q in QuickSight, a generative AI assistant allows users to generate charts and insights by typing questions in natural language, simplifying data analysis and accelerating decision making. By interpreting data relationships across sources, Amazon Q empowers users to obtain quick and meaningful answers.

Accelerating data-driven education with conversational analytics

At Mindex, we’re continually striving to innovate and improve SchoolTool’s usability for the entire education community. We recognize that not everyone has data analysis expertise.

Using QuickSight, we built effective and intuitive dashboards that transformed how users interact with data. These dashboards provide a comprehensive view of key metrics and trends enabling users to make informed decisions quickly and efficiently.

One notable example is the SchoolTool’s Absenteeism dashboard, inspired by our early warning system but primarily focused on absenteeism. The core of this dashboard revolves around three risk levels aligned with the New York state standard for chronic absenteeism. As shown in the following figure, at the top of the dashboard are four main demographic categories, which can be filtered to focus on specific demographics of interest. Users can narrow down the dashboard to students with 20% or more absenteeism or select specific graphics to filter the information, making it more accessible for those who prefer visual navigation.

Building on this foundation, we deployed Amazon Q in QuickSight to further democratize data access within SchoolTool. Q uses generative AI capabilities to understand natural language questions and their relation to the underlying data. This allows users to ask questions and get visualized insights within seconds, instead of filtering through dashboards.

For customers, this generative Q&A capability of Amazon Q in QuickSight enables them to get answers to questions such as “What are the rates of absentees by economic status” and provide the most relevant multi-visual response along with suggested questions for the best user experience.

The following figure shows the Amazon Q in QuickSight experience in our dashboards today.

Deploying QuickSight dashboards at scale

The deployment of QuickSight has significantly accelerated our ability to bring analytics to market, offering a remarkable 300% increase in speed to market.

To automate the process of moving and deploying dashboards across our different environments (developing, staging, and production), we developed an AWS Step Functions state machine that provides a visual workflow to run QuickSight Assets as Bundle API calls. This approach allows us to create a dashboard once and efficiently manage all updates. The following figure shows a sample run of our state machine that deploys the dashboard.

When we receive customer feedback on a dashboard that is already in production, we can make the necessary updates in the development environment and seamlessly propagate them to all other accounts. This ensures efficient and consistent deployment of dashboards across all environments, minimizing manual effort and maintaining high-quality analytics for our users.

Secure access to QuickSight

Our customers access QuickSight securely through role-based data filtering and secure dashboard links. Row-Level Security (RLS) and Column-Level Security are essential methods for controlling data access within a school. RLS restricts access to specific rows of data, ensuring users only see records relevant to their roles. For example, a principal can view the grades of all students in their building, while a teacher might only see the grades of their assigned students. CLS, on the other hand, restricts access to specific columns of data, allowing users to see only the information pertinent to their responsibilities. This is used to hide columns that contain sensitive information such as if a student has an Individualized Education Program(IEP) or receives free / reduced price lunches. Together, RLS and CLS provide a comprehensive security framework that safeguards student data and ensures controlled visibility at both the record and field levels.

For accessing our entire student management system application, we adhere to NIST 800-53 r5 Moderate standards and NIST CSF under Ed-Law 2d, exceeding New York State encryption requirements.

SchoolTool supports authentication and authorization with a range of single sign-on (SSO) options, including local, Google, Microsoft 365, Microsoft Azure AD, ClassLink, and Okta, each incorporating its own multi-factor authentication (MFA) mechanism. For integrating with QuickSight, SchoolTool uses its authentication and authorization mechanisms to ensure secure access to advanced analytics. Users can access QuickSight using the same credentials they use for SchoolTool through SSO, simplifying the user experience and enhancing security. MFA adds an extra layer of security by requiring a second form of authentication. SchoolTool’s authorization system controls data access within QuickSight, ensuring users can only view data and dashboards they are authorized to see. Centralized management through SchoolTool allows efficient user provisioning and access control, enhancing overall security. This integration ensures seamless access to QuickSight’s analytics while maintaining high security standards.

Looking to the future

Our QuickSight story stands out, thanks to our collaborative approach with RICs and school districts, which has been pivotal in shaping our advanced analytics solution across New York State to enhance student outcomes.

Our partnership with AWS gave us early access to Amazon Q in QuickSight. Combined with our deep expertise in dashboard development, we are uniquely positioned to help businesses effectively solve their data challenges. We are going to leverage this experience to deliver tailored solutions that drive meaningful outcomes for organizations.

To learn more about Mindex, visit their website.

About the Authors

John Hill serves as the Cloud Data and Analytics Manager at Mindex, where he sets the strategic direction for the data and analytics practice and offers support and guidance to his team. He is involved in presales engineering to secure new business and oversees the technical aspects of all ongoing projects to ensure their success.

Gowtham Shankar is a solutions architect at Amazon Web Services (AWS). He is passionate about working with customers to design and implement cloud-native architectures to address business challenges effectively. Gowtham actively engages in various open source projects, collaborating with the community to drive innovation.

Priya Mysore is a Senior Solutions Architect at Amazon Web Services (AWS) based in San Jose, California. With over two decades of experience in data and analytics, she helps customers unlock the true value of their data leveraging the AI/ML capabilities in Amazon QuickSight. Priya’s expertise lies in empowering both business and technical users, enabling them to harness the power of self-service reporting effectively. Her passion in the business intelligence space drives her commitment to delivering innovative solutions for customers.