Ask Your Data a Question. Get an Answer You Can Check.

01 What is text-to-SQL?

Text-to-SQL turns a question in plain English, such as which products sell best in a region, into a SQL query on your database. The query runs with read-only access, and the answer comes back with the query that produced it, so anyone who knows SQL can check it.

02 What is conversational BI?

Conversational BI puts text-to-SQL in a chat. Your team asks about your existing data in Claude, ChatGPT or a chat inside your own app, and gets numbers back without waiting for a report. It answers questions about what has already happened. Forecasting what comes next is a separate service.

03 How is this different from our dashboards?

A dashboard answers the questions someone planned for. Conversational BI answers the ones nobody built a chart for, from the same data. Most teams keep their dashboards for the numbers they watch every day and use the data assistant for everything else.

04 How accurate are the answers?

As accurate as the definitions and tests behind them. We write down what each business term means, test the data assistant on a test set of real questions with answers you already know, and show the query with every answer. When a question is unclear, it asks what you mean.

05 Can it change or delete our data?

No. It uses a read-only database account, so it cannot write, change or delete anything, whatever it is asked. We can also point it at a copy of your database kept in sync, so queries never slow the live system.

06 How do you keep private data out?

The account it uses can read only the tables and columns you allow, so fields such as phone numbers and emails can stay out entirely. Each person signs in and sees only what their role allows. AI providers such as OpenAI or Anthropic process what the query returns, so we keep results to what the answer needs.

07 Can our team ask from Claude or ChatGPT?

Yes, through an MCP server, a standard connector that gives the AI app read-only tools on your data. At TrueWholesale, the team asks Claude in plain English and it reads the orders, stock and app activity the platform already records, through six read-only tools.

08 Which databases can it work with?

Any SQL database you can give a read-only account on, such as PostgreSQL. We look at how your data is laid out in the discovery call, because clear table and column names, or a short data dictionary, matter more to answer quality than which database you run.

09 Can it forecast sales or demand?

No, that is predictive analytics, a separate service. Conversational BI answers questions about the data you already have, such as what sold, who ordered and what is overdue. If you need forecasts, our predictive analytics page covers them.

10 How much does conversational BI cost?

Three things drive the cost: how many databases and tables it reads, how many business terms need defining, and where your team will ask. Send us the questions, and within two working days we come back with an approach, a rough timeline and a rough number.

11 How long does it take to build?

It depends on the data and the questions. We start with a pilot on one database and the questions your team asks most, so you test real answers before we build more. We agree the timeline at the discovery call.

12 Who owns the code, and what happens after launch?

You own the code. It sits in your own code repository and runs in your own cloud account. After launch we monitor the answers, keep it updated and add tables as your questions change.

13 Do you work with UK and US clients from India?

Yes. We sign an NDA before we start, on request, and work only in the accounts you invite us into. The code repository and cloud account stay in your company's name.