IRIntellrise
How-to

Chat with your Postgres database in plain English

Updated 2026-08-25

An AI data analyst is most useful when it runs on your real database instead of a file you paste in. This guide shows how to connect a PostgreSQL database so you can ask questions in plain English and get charts, dashboards and reports back — without writing SQL by hand.

The whole setup takes a few minutes. You connect with a read-only user, let the AI read your schema, and ask your first question. Your database credentials are encrypted, and the analysis runs on your own AI provider key.

  1. 1

    Gather your PostgreSQL connection details

    You need the host, port (usually 5432), database name, username and password, plus the schema to read (the form pre-fills “public”). A managed Postgres provider usually hands you one connection URL instead — you will be reading those values out of it, because Intellrise asks for them as separate fields and has nowhere to paste a URL. SSL is not something you decide here: Postgres connections always use sslmode=require.

  2. 2

    Create a read-only database user (recommended)

    Give the AI analyst its own user with SELECT-only access. This means it can read data to answer questions but can never modify it. A read-only role is the safest way to connect any production database to an analytics tool. The four PostgreSQL statements are written out at intellrise.com/help/database-requirements — we ran them on PostgreSQL 18.3 on 16 August 2026 and printed what each blocked write actually returned, and we say on the same page that Intellrise does not verify the role you give it is read-only.

  3. 3

    Create your account and add your AI key

    Sign up, then add your own AI provider key (bring-your-own-key). If you don't have one yet, a free key takes about a minute to create. Your key is encrypted and used only for your requests, so the AI bill comes from your provider at their rates rather than from us.

  4. 4

    Add the PostgreSQL connection

    In the app, add a new data source, choose PostgreSQL, and enter your host, port, database, schema, user and password — separate fields, with no box for a pasted connection string and no SSL option to switch on. Click Connect: it stores the credentials encrypted and then immediately dials your database, so you get the answer on the form itself — either “Connected — N tables found” or the exact error Postgres sent back. A source that fails is still saved, so correct the details with the pencil icon and save again. Credentials are encrypted at rest.

  5. 5

    Let the AI read and annotate your schema

    The analyst reads your tables and columns and describes what they mean. If you use dbt, point it at your models so it recognises your fact and dimension tables. This is the semantic layer — and it keeps improving as you use it.

  6. 6

    Ask your first question in plain English

    Ask something like "revenue by month this year" or "top 10 customers by orders". On a question the analyst judges complex — a join, a query spanning two sources, several steps — it shows a readable plan and waits for you to approve it before running anything; on one it judges simple it runs first and shows you afterwards. Either way you get the same two controls on every answer: Show SQL gives you the exact query that produced the number, and Show table gives you the rows it returned with their count. That is how you check a figure instead of trusting it. Refine by chatting, then pin it to a dashboard or save it as a report.

Why connect the database instead of exporting a file

Exports go stale the moment you download them, and when an assistant only ever sees that exported file it has to guess what your columns mean. A connected AI data analyst reads live data, shows the query plan in business terms before running it, and saves what your fields mean against the source itself — so the next question already knows them, and answers stay current and consistent.

Keeping it safe

  • Connect with a read-only user. That grant is what actually stops a write, and it is yours to set — and when you connect we check what the account you gave us can actually do, and tell you.
  • A read-only SQL guard refuses write statements such as DROP, DELETE, UPDATE and INSERT.
  • Connection details and your AI key are encrypted at rest.
  • Your business data is never used to train our models. Your PostgreSQL database is attached read-only and queried in place rather than mirrored — the rows an answer returns stay in your own account until you delete them.

Frequently asked questions

No, and you never have to write it. On every answer the exact SQL that ran is one click away under Show SQL, together with the rows it returned under Show table — that is there whether the question was simple or complex. On complex questions you also get a plain-English plan to approve before anything runs at all.

Related

Get your end-to-end AI business intelligence now.

Ask in plain English, pin the answer to a dashboard, export it as CSV, Excel, PDF or PowerPoint — on the free tier that whole loop runs on one database source plus Google Sheets, and the field definitions you save carry into every later session, including after a schema refresh. One exception you should hear from us rather than discover: on a Google Sheet, re-running the tab picker — even only to add a tab — clears the definitions saved for that source and reads its schema fresh. Connecting a database is the technical part: a read-only user and network access, once. If your numbers live in a spreadsheet, none of that applies: put the file in a Google Sheet and it connects on the free tier, with no read-only user and no open port. Uploading a CSV or Excel file directly is a Pro feature, and every new account gets 14 days of Pro with no card. The one thing no plan waives is your own AI provider key, and a Google Gemini key is free to create.

Get started free Talk to us

Every new account starts on a 14-day Pro trial with no card. After it lapses, the free tier stays free.