Ask in plain language
Type your question the way you would ask a colleague. The AI translates it into SQL that fits your tables and columns.
AI on your own data
Ask your question in plain language. Your data stays yours.
With ENABLE, you ask questions of your own business data without knowing SQL. The AI turns your question into a query on your own data warehouse, and ENABLE runs it read-only. The AI only ever sees the structure of your data (tables and columns), never the rows themselves.
enable.dataanalytics.nl/acme
SQL Explorer
SQLDAXWhat was revenue per region this quarter, compared with last year? Generate SQL
acme_warehouse Query plan: read-only Run
1SELECT r.region,2 SUM(i.revenue) FILTER (WHERE i.year = 2026) AS revenue_2026,3 SUM(i.revenue) FILTER (WHERE i.year = 2025) AS revenue_20254FROM sales.invoices i5JOIN master.region r ON r.region_id = i.region_id6WHERE i.quarter = 37GROUP BY r.region8ORDER BY revenue_2026 DESC;
5 rows · 0.21 s CSV · JSON
regionrevenue_2026revenue_2025Δ
South1,421,3001,198,400+18.6%
North1,284,9001,121,700+14.5%
Central1,183,2001,041,000+13.7%
East962,400911,800+5.5%
West881,600852,300+3.4%
Q3 revenue by region
South
North
Central
East
West
AI on your own data means an AI model answers on the basis of your organisation's own data, rather than general knowledge from the internet. For business data such as revenue, stock, orders or hours, it works most reliably when the AI doesn't do the arithmetic itself but writes a query that lets the database do the calculating.
This is how ENABLE works: you type a question such as "What was revenue per region this quarter, compared with last year?". The AI writes the SQL based on the structure of your data warehouse, ENABLE checks the query with a read-only query plan and then runs it. You see the query, the result and a chart, and you can adjust or save any of it.
The difference from "pasting your data into ChatGPT": with ENABLE, your data rows never leave the data warehouse on their way to the AI. The model only receives table and column names and your question. The figures you see come straight from your own database.
Key points
AI on your own data at a glance
Explain or fix a query
Have the AI explain an existing query in plain language, or resolve an error message without having to hunt for the cause yourself.
Read-only execution
Queries run in a read-only transaction with a 45-second timeout and a limit on the number of queries per minute. Writing or deleting is not possible.
Results as a table or chart
View the outcome as a table or chart and export to CSV (opens straight in Excel) or JSON if you want to work with it further.
Save and reuse
Keep good queries so they can be run again. If you publish a query as an API, every publication gets its own version number.
Publish as an API
A saved query can become a secured REST endpoint, so other systems retrieve the same, verified figures.
Why AI on your own data works better through your data warehouse
Language models are strong at language, but not at exact arithmetic across thousands of rows. If you give a model an export and ask for totals, you sometimes get an answer that looks plausible but is wrong. A database doesn't make that mistake.
That is why ENABLE lets the AI do what it is good at (understanding your question and writing a query) and leaves the calculations to your data warehouse. That makes the answer verifiable: you see exactly which tables, filters and calculations were used.
ChatGPT on company data: what to watch out for
Many organisations are experimenting with ChatGPT or a similar chatbot on company data. The risks usually lie not in the model itself, but in what you share with it and who gets to see which answers.
- Which data goes to the model? With ENABLE, only metadata and the question, no rows.
- Who may query which data? In ENABLE, roles and access profiles determine who can use the SQL Explorer.
- Can you check the answer? In ENABLE, the query is always shown next to the result.
- Can you see afterwards what was asked and run? In ENABLE, every executed, blocked or failed query is recorded in the audit log.
Text-to-SQL or RAG: which form of AI on your own data suits you?
There are two common ways to make AI work on your own data. With RAG (retrieval-augmented generation), a system looks up relevant passages of text, for example from manuals or contracts, and passes them to the model. With text-to-SQL, the AI writes a query on structured data in a database.
ENABLE uses text-to-SQL, because the platform revolves around figures: revenue, margins, stock, planning and KPIs. If you mainly want to ask questions about documents, RAG is the better technique; that is not part of ENABLE.
| Text-to-SQL (ENABLE) | RAG | |
|---|---|---|
| Type of data | Tables in a database | Documents and text |
| Typical question | "Revenue per customer group per month?" | "What does our sick-leave policy say?" |
| What goes to the model | Structure (metadata) and the question | Text passages from your documents |
| Verifiable | Yes, through the query | Through source citations |
Who is AI on business data in ENABLE for?
The SQL Explorer is available to customer admins by default and can be switched on for regular users per organisation. In practice, controllers, analysts and data teams use it to quickly answer a question that isn't (yet) covered by a dashboard. Whatever keeps coming back then becomes a standard dashboard or a custom dashboard in the Dashboard Builder.
In ENABLE
How AI on your own data works in ENABLE
- 01
Connect your data warehouse
ENABLE connects to your organisation's PostgreSQL data warehouse over an encrypted connection and runs read-only queries only.
- 02
Ask a question
You type a question in plain language. The AI receives the table and column names (and the data dictionary) and writes SQL.
- 03
Check and run
ENABLE checks the query with a read-only query plan and runs it within your organisation's limits.
- 04
Use or reuse
View the result as a table or chart, export it, save the query or publish it as an API.
FAQ
Frequently asked questions
What is AI on your own data?
AI on your own data is AI that answers on the basis of your own organisation's data, such as your ERP, CRM or data warehouse, rather than on general knowledge. In ENABLE, the AI does this by writing a SQL query on your own data warehouse.
Does my company data go to ChatGPT or OpenAI?
No, not the data rows. ENABLE only sends metadata (table and column names, the data dictionary) and the user's question to the AI. The query runs in your own database and the result is not sent back to the model.
Do I need to know SQL to use AI on my own data?
No. You ask your question in plain language and the AI writes the SQL. If you do know SQL, you can view, edit and save the query.
How do I know whether the AI's answer is correct?
The figures don't come from the AI but from your own database. Next to the result you always see the query that was run, so you can check which tables, filters and calculations were used.
Can the AI change or delete data?
No. Queries in ENABLE run in a read-only transaction, with a 45-second timeout.
Does this also work with documents, such as a knowledge base or SharePoint?
No. ENABLE is built for structured data in a data warehouse (text-to-SQL). For questions about documents, a RAG solution is more suitable; that is not part of ENABLE.
Further reading
- SolutionAI on ERP data: ask questions in plain languageAsk questions about your ERP data in plain language, via your own warehouse, no exports.Read more
- PlatformSQL Explorer with AI: ask your database questionsAsk your data warehouse in plain language. The AI writes the SQL, you stay in control.Read more
- SolutionSecure AI for business: private AI on your dataWhat private, secure AI really means, and how ENABLE limits what the AI sees of your data.Read more
- SolutionAI governance: control over data, AI and accessWho sees what, what the AI receives and who approves: practical AI and data governance.Read more
- ResourcesText-to-SQL explained: AI on your databaseText-to-SQL is a technique in which an AI model converts a question in plain language into an SQL query, which is then run on a database. The language model writes the query; the database supplies the figures. How well that works depends mainly on the quality of your metadata and on the safeguards around execution, such as read-only access, timeouts and row limits.Read more
- ResourcesRAG or text-to-SQL: which AI should you choose?RAG and text-to-SQL are two ways to make AI work with your own data. RAG (retrieval-augmented generation) searches your documents for relevant text passages and lets the model formulate an answer from them; text-to-SQL lets the model write a query on a database, so that the database does the calculating. Use RAG for questions about documents and knowledge, and text-to-SQL for questions about figures.Read more
Share securely, live fast, no headache
In an online demo we show you the portal: dashboards per role, row-level security, plain-language questions and how we set it up and manage it for you.