AI Powered Analytics: Ask Questions, Get SQL + Charts

NOVA AI

Ask Your Database Anything.
Get Answers in Seconds.

AI-powered text-to-SQL analytics. Ask questions in plain English, get validated queries, live results, and instant charts — no SQL knowledge required.

Schema-Aware SQL Live Query Execution Instant Charts Zero Credential Exposure

Try Nova Free Learn More

See everything Query Streams can do beyond this guide Get Started Free

Supports

+more

Live Data

Replace exports with on-demand SQL

Excel Sheets

BI & AI Analytics

Ask Nova, get SQL + charts

Meet Nova

Database REST API

One key per partner. No credentials shared.

Build an API

+more

Automation

Scheduled sync to 6+ platforms

Explore

+more

API to SQL

Query APIs with SQL, no code

Explore

+more

AI Database MCP

Claude, Cursor, ChatGPT & Grok talk to your data

Connect AI

AI powered analytics lets anyone query a database by asking questions in plain English — no SQL required. Query Streams includes Nova AI, a natural language analytics engine that reads your actual database schema, generates validated SQL, runs the query against your live data, and builds charts — all from a conversational interface.

Whether your data lives in SQL Server, PostgreSQL, MySQL, Oracle, BigQuery, Snowflake, or any supported database, Nova AI turns English questions into accurate SQL results. No BI tool licenses, no data models, no learning curve.

How AI Powered Analytics Works

Traditional analytics forces a choice: learn SQL or wait for someone who knows it. AI powered analytics eliminates that bottleneck. Here’s the flow with Nova AI:

What Makes Nova AI Different

There’s no shortage of “AI SQL” tools. Most of them generate generic SQL that may or may not work on your database. Nova AI is different because it’s directly connected to your actual database schema through the Query Streams Agent:

Example: From Question to Results

Here’s what an actual Nova AI conversation looks like. The user asks a business question, Nova generates SQL, runs it, and presents the answer with a chart:

Nova AI

PostgreSQL — production

What were our top 5 products by revenue last quarter?

I found your sales and products tables. Here’s the SQL I generated and validated against your schema:

SQL Results

Copy Run Save

SELECT
    p.product_name,
    SUM(s.quantity) AS units_sold,
    SUM(s.line_total) AS total_revenue,
    ROUND(AVG(s.line_total / NULLIF(s.quantity, 0)), 2) AS avg_unit_price
FROM sales s
JOIN products p ON s.product_id = p.id
WHERE s.sale_date >= DATE_TRUNC('quarter', CURRENT_DATE - INTERVAL '3 months')
  AND s.sale_date < DATE_TRUNC('quarter', CURRENT_DATE)
GROUP BY p.product_name
ORDER BY total_revenue DESC
LIMIT 5

Ask Nova anything about your data…

Example: Conversational Data Analysis

Nova doesn’t just generate SQL and disappear. It reads the results, analyzes the data, and has an actual discussion with you about what it found. Here’s a multi-turn analytics conversation:

Nova AI

SQL Server — billing_prod

How are our subscription renewals trending this year compared to last year?

I queried your subscriptions and renewal_events tables. Here’s what the data shows:

Renewals are up 23% YoY — 4,812 renewals Jan–Mar 2026 vs. 3,912 in the same period last year. January and February both exceeded targets at 1,680 and 1,740 respectively (target was 1,500/month).

However, March dropped to 1,392 — that’s 7.2% below target and breaks the upward trend. The churn rate in March spiked to 4.1%, up from a 2.3% average in Jan–Feb. That looks like the primary driver.

Want me to break down the March churn by plan tier or cancellation reason?

What’s causing the March churn spike? Can you break it down by plan tier?

Good question. I ran a follow-up joining subscriptions with plan_tiers and cancellation_reasons.

82% of March churn came from the Starter tier — 94 out of 115 cancellations. Pro and Enterprise tiers stayed stable. The top cancellation reason was “price increase” — 67 of those 94 Starter cancellations cited the March 1 price change specifically.

Based on similar patterns after the Q3 2025 price adjustment, offering a 90-day grace period or loyalty discount for existing Starter customers could recover roughly 60–70% of these cancellations.

Want me to chart the churn breakdown by tier, or pull the actual revenue impact?

Ask Nova anything about your data…

Example: Auto-Generated Charts

When your question calls for a visualization, Nova generates the chart automatically based on the shape of your data. No configuration, no drag-and-drop builders — just ask and see:

Nova AI

PostgreSQL — analytics

Show me monthly revenue by product category for Q1 2026

I queried orders joined with products and grouped by category. Here’s your Q1 revenue breakdown:

Revenue by Product Category — Q1 2026 Bar

Software

$944.2K

Hardware

$578.5K

Services

$500.8K

3 categories · Regenerate

Software leads at $944K — nearly double Hardware. But Services had the highest growth rate at 33% across the quarter (Jan $143K → Mar $190K), while Software grew 22% and Hardware held steady.

Total Q1 revenue: $2,023,500. Want me to break this down by month, or chart the growth trend as a line chart?

Ask Nova anything about your data…

Who Benefits from AI Powered Analytics

AI analytics isn’t just for people who can’t write SQL. It accelerates workflows for everyone across the organization:

How to Get Started (3 Steps)

Step 1: Connect your database. Sign up at my.querystreams.com and install the Query Streams Agent on any machine that can reach your database. The agent connects outbound — no firewall changes needed. Supports SQL Server, PostgreSQL, MySQL, MariaDB, Oracle, Snowflake, BigQuery, DuckDB, SQLite, and Microsoft Access.

Step 2: Open Nova AI. Navigate to the Nova interface from your Query Streams dashboard. Select the database and data source you want to query. Nova reads the schema automatically — tables, columns, data types, and relationships.

Step 3: Ask your question. Type a question in plain English. Nova generates validated SQL for your specific database dialect, runs it against live data, and presents the results with suggested visualizations. Refine with follow-up questions or save the generated query for reuse in Excel and Google Sheets.

Supported Databases

Nova AI generates dialect-correct SQL for every database Query Streams supports. Deploy multiple Agents across your infrastructure — on-premise, AWS EC2, Azure VMs, Google Cloud, or any environment globally. Each agent auto-updates and makes its data sources available in Nova’s data source selector.

Supported databases include Microsoft SQL Server, PostgreSQL, MySQL, MariaDB, SQLite, Microsoft Access, Oracle, Snowflake, BigQuery, and DuckDB. Nova also works with API connector data (Stripe, HubSpot, Shopify, Google Analytics, Google Search Console, ShipStation) cached locally in DuckDB.

Beyond AI: Save and Share Queries

AI powered analytics is just the starting point. Once Nova generates a query you like, save it and use it across the platform:

Run in Excel or Google Sheets — Saved queries appear in the Query Streams add-in. Select, filter, run. Live data streams into your spreadsheet on-demand, with support for parallel execution across multiple tabs.

Share with teammates or external partners — Recipients run shared queries from their own Excel or Google Sheets. They can adjust filters independently without seeing your SQL code or database credentials.

Automate to other platforms — Schedule the query to sync data to Airtable, Smartsheet, Baserow, or SeaTable on a recurring basis.

Frequently Asked Questions

Get Started with AI Analytics

Sign up free at my.querystreams.com, connect your database, and ask Nova your first question. No SQL knowledge required — just ask in plain English and get validated results from your live data in seconds.

For more on Nova AI capabilities, see the Nova AI feature page. For database connection guides, see SQL Server to Excel, PostgreSQL to Excel, and Natural Language to SQL.