Business data should be as easy to question as it is to collect.
That’s what we’re building with Insights.
An AI-native analyst that helps teams ask questions, get answers, surface patterns, and understand what to do next from their own data.
Built to be useful, grounded, proactive, and private from day one.
Would love for people who work with business data to try it and tell us what we should improve.
Introducing Insights, an AI-native analyst for business data.
Your business data should not need a translator.
With Insights, anyone can ask questions in plain English and get clear answers, charts, and next steps from their actual data.
It also proactively surfaces patterns, risks, and opportunities your team might miss.
Why did revenue fall?
Which product is hurting margin?
Where are customers slowing down?
What should we focus on next?
No complex tools.
No waiting on reports.
No need to be technical.
No made-up numbers.
Insights is grounded in the data you provide, so answers come from your numbers, not guesses.
Start with spreadsheets today.
Integrations are coming soon.
Your raw data stays private.
The model only sees what it needs to help find the pattern.
Early access is now open.
Get started here: https://t.co/nHXOkT2w7m
Built by Quiver.
Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for general use.
Its capabilities exceed those of any model we’ve ever made generally available.
Data tools have always assumed you know what you're doing.
Know SQL.
Know which chart type.
know how to structure a query.
know how to interpret the output.
Insights assumes none of that, you just ask what you want to know, it figures out the rest and hands you back an answer with a visual and the reasoning behind it.
The question is the only thing you need to bring.
The spreadsheet was never the product.
The dashboard was never the product.
Even the insight was never the product.
The product is better decisions.
For decades we’ve been forcing people to learn SQL, build reports, hire analysts, and navigate dashboards just to get there.
AI changes that.
The future is not software that shows you your business.
It’s software that understands it.
Makes me appreciate how lucky we are to be building right now.
A few years ago, building something like @insightsbyQ would’ve required a much larger team and significantly more resources. Today, a small team can take an idea from a conversation to a working product in a matter of months.
The barrier is increasingly becoming creativity, not capability.
Check out the dashboard feature in Insights
you don't set it up. you just ask questions, pin the visuals that matter, and it becomes yours over time.
new video showing exactly how it works, in the new dark mode
Honestly this took us so long to get right.
We kept asking ourselves why the AI should see your actual data just to give you a simple insight.
The answer was clear. It shouldn’t.
So we made it sure it never does.
The AI never sees your raw data. Ever.
Every other analytics tool asks you to just trust them. That was never good enough.
There's a hard wall between your data and the AI. It only ever sees aggregates. Totals, counts, averages. Never a single raw row. Never a customer name. Never anything sensitive.
Your data stays yours.
Most businesses drown in data but starve for answers.
Exports, SQL errors, analyst queues. All just to answer a question your data already knows.
Insights cuts through all of it. Ask in plain English, get your answer in seconds.
The bottleneck wasn't your data. It was the process. We fixed it.
Try it now at https://t.co/BqcXljL92l.
The first wave of AI was reactive.
You opened a chat window, asked a question, and got a response.
The next wave is proactive.
AI systems connected to live environments, continuously executing, monitoring, analyzing, and acting without waiting for prompts.
That shift is going to be much bigger than most people realize.
A lot of AI products stop at generating responses after being prompted.
Insights is built around live execution instead.
The system connects directly to live data sources, proactively monitors and analyzes incoming data, and generates analytical workflows automatically without requiring users to constantly query dashboards or manually prompt the system.
The goal is not just AI-generated text. It is AI-generated computation operating continuously on live business data.
Try it now at https://t.co/BqcXljLGRT
A lot of AI products stop at generating responses after being prompted.
Insights is built around live execution instead.
The system connects directly to live data sources, proactively monitors and analyzes incoming data, and generates analytical workflows automatically without requiring users to constantly query dashboards or manually prompt the system.
The goal is not just AI-generated text. It is AI-generated computation operating continuously on live business data.
Try it now at https://t.co/BqcXljLGRT
Most AI data tools have two problems: they hallucinate, and they expose your raw data.
Insights has neither.
Zero hallucination. Your data stays yours.
Query anything in plain English and get answers you can actually trust.
Try it now at https://t.co/ey1znqmbVM
The interface for data has barely changed in decades.
Dashboards reduced data to charts.
SQL reduced it to queries.
AI changes the interface entirely.
People no longer need to understand databases to work with data. They only need to understand the problem they’re trying to solve.
Most data workflows today still depend on SQL, dashboards, or analyst intervention for basic analysis.
Insight introduces a different execution model.
Users interact with their data using natural language, and Insight generates structured analysis programs that execute in a controlled environment. The model operates on schema-level context rather than direct access to raw data.
We’re sharing an early look at the system below. Still actively refining the product and execution flow, so feedback from people working with real data workflows would be especially valuable.
First demo below.