Questions in plain language
Revenue, orders, stock, purchasing or receivables: type your question and the AI writes SQL that fits the tables in your warehouse.
AI on ERP data
With AI on ERP data, you ask questions about revenue, orders, stock or receivables in plain language. ENABLE works with the ERP data in your own PostgreSQL data warehouse: the AI writes a query, ENABLE runs it read-only and you see the result with the query alongside it. The AI only sees table and column names, never rows.
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;
AI on ERP data means asking questions about the data in your ERP system, such as sales orders, invoices, stock and purchasing, and getting an answer that comes straight from that data. In ENABLE, this doesn't happen in the ERP itself but on a copy of your ERP data in your own PostgreSQL data warehouse.
A controller at Acme Groothandel, a fictional wholesaler, might type: "Which ten customers had the biggest drop in revenue this year compared with last year?". The AI writes SQL for it based on the table and column names in the warehouse, ENABLE checks the query with a read-only query plan and runs it. You see the result as a table or chart, with the query alongside, so you can check how the figure was arrived at.
Why go through a data warehouse? An ERP is built to record transactions, not to analyse them. In a warehouse you combine ERP data with other sources, you don't burden the ERP with heavy analyses, and the AI never comes anywhere near your accounting records.
Key points
Revenue, orders, stock, purchasing or receivables: type your question and the AI writes SQL that fits the tables in your warehouse.
The AI receives the structure of your warehouse (tables, columns, data types and relationships) and is instructed to use only names that actually exist.
Every query runs in a read-only transaction, with a 45-second timeout and a limit on the number of queries per minute.
The SQL sits next to every result. If a filter or definition is wrong, you adjust the query or ask the AI to explain it.
Save a good question. Whatever keeps coming back becomes a standard Power BI dashboard or a custom dashboard in the Dashboard Builder (pilot).
The Targets app records monthly KPI targets in your own warehouse, so you can put them next to the ERP figures in a query or dashboard.
AI on ERP data in ENABLE consists of four layers. The ERP remains your source for transactions, a loading process copies the data to a PostgreSQL data warehouse, and ENABLE puts your questions to that warehouse, never to the ERP itself.
| Layer | What happens | Who handles it |
|---|---|---|
| ERP (e.g. Business Central) | Orders, invoices, stock, purchasing | Your own ERP administrators |
| Loading process | Copying data periodically, for example with Airbyte | Outside ENABLE, for example with help from Data Analytics |
| PostgreSQL data warehouse | Modelling the data into understandable tables | Per customer; the location varies by customer |
| ENABLE | Plain-language questions, dashboards, apps, API | The portal, hosted in Frankfurt |
Acme Groothandel is a fictional wholesaler that uses Microsoft Dynamics 365 Business Central as its ERP. This is what a week of questions to its ERP data looks like:
Whether a question can be answered depends on what is in your warehouse, not on which ERP you use. If the delivery date isn't in the data, the AI can't make it up either. The clearer the table and column names, the better the AI picks the right fields; that is why it pays to model ERP data into understandable tables or views.
Some ERP vendors build AI features into their own software, such as Copilot in Business Central. Those are aimed at the work inside the ERP. ENABLE focuses on analysis: questions about the data in your warehouse, optionally combined with other sources such as your CRM or webshop.
Exact Online and AFAS Profit are business software packages widely used in the Netherlands and Belgium. The same approach applies to them: the data first goes into your warehouse.
| ERP | How the data gets into your warehouse | In ENABLE |
|---|---|---|
| Microsoft Dynamics 365 Business Central | Load with a tool such as Airbyte, then model | Usable via the warehouse |
| Exact Online | Via a loading tool or custom integration | Usable via the warehouse, no ready-made connector |
| AFAS Profit | Via a loading tool or custom integration | Usable via the warehouse, no ready-made connector |
The SQL Explorer is the place for one-off questions. What you want to see more often, you make permanent. In the Dashboard Builder (pilot), "Ask AI" suggests a chart based on the metadata of approved data sources; you apply the suggestion or discard it.
Apps work on the same warehouse too: the built-in Targets app for monthly KPI targets, or an app you have built yourself, such as a (made-up) stock planner that creates an order proposal per item. That way everyone looks at the same figures.
The technology is rarely the problem. The difference lies in definitions and access.
In ENABLE
Your ERP data is loaded into a PostgreSQL data warehouse outside ENABLE, for example with Airbyte, and modelled into understandable tables.
ENABLE connects to your organisation's warehouse over an encrypted connection and runs read-only queries only.
You type a question in the SQL Explorer. The AI receives the table and column names and writes SQL; ENABLE checks it with a read-only query plan.
View the result as a table or chart, export to CSV, save the query or turn it into a dashboard or API endpoint.
FAQ
AI on ERP data is AI that answers questions about the data in your ERP system, such as orders, invoices and stock. In ENABLE, the AI writes a SQL query on a copy of your ERP data in your own data warehouse, and ENABLE runs it read-only.
Yes, once your Business Central data is in a PostgreSQL data warehouse. Loading happens outside ENABLE, for example with Airbyte. After that, you ask questions about that data in plain language in the SQL Explorer.
Yes, if the data from Exact Online or AFAS Profit is in your data warehouse. ENABLE has no ready-made connector for those packages; loading is done with a loading tool or a custom integration.
No, not the data itself. The AI only receives metadata, such as table and column names, and the user's question. The query runs in your own warehouse and the result is not sent back to the model.
No. ENABLE has no connection to your ERP, only to your data warehouse, and queries run there in a read-only transaction. Posting, changing or deleting is not possible.
As up to date as the latest load of your data warehouse. How often it is refreshed is set in the loading process outside ENABLE.
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.