AWS Builder Center

Builder Center

Empowering Decision Makers: Building Automated KPI Dashboards and Natural Language Queries with Amazon QuickSight

Empowering Decision Makers: Building Automated KPI Dashboards and Natural Language Queries with Amazon QuickSight

Learn how to connect Amazon QuickSight to AWS data sources like Athena, automate KPI dashboard refreshes using SPICE, and enable non-technical stakeholders to ask natural language questions using QuickSight Q to instantly generate insights.

Mohit

BCA Graduate from Amity University | Interested in Data Analytics | Problem Solving | Quick Learner
In today’s data-driven world, getting insights into the hands of business stakeholders quickly is just as important as the data pipelines themselves. Traditional Business Intelligence (BI) often creates a bottleneck: business users need a new chart, they submit a ticket to the data team, and wait days for a response.
In this guide, we will build a modern BI solution using Amazon QuickSight that solves this problem in two ways:
  1. Building automated, executive-ready KPI dashboards.
  2. Enabling QuickSight Q so non-technical users can ask natural language questions and get instant visual answers.

Prerequisites

  • An AWS Account with Amazon QuickSight Enterprise Edition enabled (required for QuickSight Q).
  • Sample data queried via Amazon Athena or housed in Amazon Redshift.

Step 1: Connecting QuickSight to Your AWS Data Source

Whether your data is sitting in an S3 Data Lake queried by Athena or a Redshift data warehouse, connecting QuickSight is seamless.
  1. Navigate to the QuickSight console and select Datasets > New dataset.
  2. Choose your data source (e.g., Athena or Redshift).
  3. Name your data source and validate your connection.
  4. Select the database and table containing your aggregated business metrics (e.g., daily_sales_summary).
  5. Choose to Import to SPICE for quicker analytics. SPICE is QuickSight’s in-memory engine that ensures your dashboards load lightning-fast without running underlying database queries for every click.

Step 2: Automating Data Refreshes

Because we imported our data into SPICE, we need to ensure the dashboard reflects the latest metrics.
  1. Go to your new Dataset and click Schedule Refresh.
  2. Select Create schedule.
  3. Choose a frequency that matches your business needs (e.g., Daily at 6:00 AM) so the data is fresh when executives log in.
  4. Click Save. Your pipeline is now fully automated!

Step 3: Designing the Executive KPI Dashboard

Now, let's build the visual interface.
  1. Click Create Analysis from your dataset.
  2. In the visual types menu, select the KPI chart. Drag your revenue field to the Value well, and date to the Trend group well to instantly show month-over-month growth.
  3. Add a Line Chart to visualize revenue trends over time, and a Donut Chart to break down sales by product category.
  4. Once you are satisfied with the layout, click Share > Publish Dashboard to make it accessible to your stakeholders.

Step 4: Unleashing Natural Language Queries with QuickSight Q

Dashboards are great, but executives always have ad-hoc questions. Instead of building endless dashboard variations, we can use QuickSight Q, an ML-powered engine that translates natural language into data visuals.

Creating a Topic

QuickSight Q uses "Topics" to understand the context of the data it is querying.
  1. Go to the QuickSight sidebar and select Topics > New Topic.
  2. Name your topic (e.g., Q3 Sales Data) and attach the dataset we created earlier.
  3. Configure metadata: This is the most critical step. Rename your fields to natural language synonyms. For example, if your database column is rev_usd_amt, add synonyms like revenue, sales, and money.

Testing the Q Bar

Once the topic is indexed, a search bar will appear at the top of your dashboard. Try asking it questions exactly as a human would:
  • "What were the top 5 regions by sales last month?"
  • "Show me daily revenue in California for 2023."
  • "Who is our top performing sales rep?"
QuickSight Q parses the intent, automatically filters the data, and generates the appropriate chart on the fly. Users can even pin these generated charts to their own custom dashboards.

Conclusion

By combining automated SPICE refreshes with Amazon QuickSight Q, we have completely transformed the BI experience. The data team is freed from answering ad-hoc requests, and business stakeholders are empowered to explore data using the most intuitive interface possible: their own words.
# data-analytics# amazon-quicksight# amazon-athena# serverless
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article