Revenue and margin analysis
Ask about revenue, cost of sales and margin per customer, product group, branch or period. The calculation is in the query, so you can see which definition was used.
AI for finance
Financial questions in plain language. Every figure traceable.
AI on financial data helps controllers and finance teams get faster answers to questions about revenue, margin, receivables and budget versus actuals. In ENABLE, the AI writes a query on your own data warehouse; the figures come from the database, not from the model. You always see how a figure was calculated, and the AI only sees metadata.
enable.dataanalytics.nl/acme/eigen-dashboards
Custom dashboards /Margin by product groupDraft Saved PreviewPublish
Visual
Gross margin31.6%+1.2 pt vs last year
Margin in €€1.37m+12.1%
Best product groupPackaging38.2% margin
SelectedMargin by product group · this year vs last year
Packaging
Office
Hospitality
Cleaning
Technical
Revenue share
- Packaging
- Office
- Hospitality
- Cleaning
- Technical
AI on financial data means you ask questions about your financial figures in plain language and an AI model translates each question into a calculation on your own data. For finance, one requirement is non-negotiable: every figure must be traceable. That is why ENABLE doesn't let the AI invent or summarise figures, but has it write a SQL query that your database runs.
A controller at Acme Groothandel, a fictional wholesaler, might ask: "What was the gross margin per product group in the third quarter, compared with the same quarter last year?". ENABLE shows the result as a table or chart, with the query alongside. If the margin definition differs from yours, you adjust the query and save it as a standard analysis.
The AI only sees the structure of your data: table and column names, no amounts, customer details or ledger entries. The query runs read-only, so nothing can be posted, changed or deleted.
Key points
AI on financial data at a glance
Comparison with last year
Have the AI put a period next to the same period last year. In the Dashboard Builder (pilot), prior-year comparison and margin calculations are built into the charts as standard.
Receivables at a glance
Open items, ageing and the largest debtors, if that data is in your warehouse. Export to CSV to work with it further.
Budget versus actuals
The Targets app records monthly targets in your own warehouse, in euros, percentages or quantities, with history. Put them next to actuals in a query or dashboard.
Traceable and reusable
Save good analyses so everyone runs the same calculation. If you publish one as an API, every version gets a number.
Restricting access per entity
With row-level security on Power BI dashboards, a branch manager only sees their own figures. Without an assigned tag, there is no access.
"Ask AI" asks follow-up questions
In the Dashboard Builder (pilot), "Ask AI" asks for clarification when it isn't defined what margin or revenue means in your data, instead of guessing.
Why AI on financial data only works if every figure is traceable
A language model can give a plausible answer that is slightly off. In a piece of text that is annoying; in a margin figure in a management report it is a problem. That is why in ENABLE the database does the calculating, not the model. The AI only writes the query, and that query is the audit trail: which tables, which filters, which calculation.
| Pasting an export into a chatbot | AI on financial data in ENABLE | |
|---|---|---|
| Who calculates | The language model | Your own database (SQL) |
| What goes to the model | The rows with amounts | Table and column names and the question |
| Traceable | Barely: the answer is text | Yes: the query sits next to the result |
| Access | Whoever has the export sees everything | Roles, access profiles and row-level security |
Financial analysis with AI: example questions
This is how the finance team at the fictional Acme Groothandel puts questions to its data warehouse:
- "What was revenue per month this year, and what percentage change is that compared with last year?"
- "Which product groups had the largest drop in margin last quarter?"
- "Which ten debtors have the highest outstanding amount older than 60 days?"
- "What is the average payment term per customer group?"
- "How does actual revenue per branch compare with the monthly target?"
Margin analysis: definitions first, then the AI
Most discussions about figures are not about the calculation but about the definition. An AI can't make that choice for you. So record what you mean, for example in a view in the warehouse or in a saved query that everyone uses.
In the Dashboard Builder (pilot), a margin chart calculates (revenue minus costs) divided by revenue, after summing per period or group. Which columns count as revenue and costs is set in the definition of the approved data source.
- Revenue including or excluding VAT?
- Do credit notes, discounts and rebates count?
- Cost price: standard or actual?
- Is intercompany revenue eliminated?
Who sees which financial figures?
Financial data is sensitive, even within your own organisation. In ENABLE you control access at two levels. You restrict Power BI dashboards with row-level security: a profile or user tag is linked to a Power BI role, so someone only sees the figures for their own entity or branch.
The SQL Explorer queries the entire warehouse. It is available to customer admins by default, and to regular users only if you switch it on for your organisation. So give it to people who are allowed to see all the figures. Executed, blocked and failed queries are recorded in the audit log.
In ENABLE
How AI on financial data works in ENABLE
- 01
Financial data in the warehouse
General ledger, invoices and receivables from your ERP are loaded into your PostgreSQL data warehouse outside ENABLE and modelled.
- 02
Record the definitions
You record revenue, margin and other key figures in views or saved queries, so everyone uses the same calculation.
- 03
Ask and check
You ask your question in the SQL Explorer, the AI writes SQL and ENABLE checks it with a read-only query plan. You review the query and the result.
- 04
Make it permanent and share
Save the analysis, export to CSV or put it in a dashboard with row-level security. You record targets in the Targets app.
FAQ
Frequently asked questions
What is AI on financial data?
AI on financial data is AI that answers questions about your financial figures, such as revenue, margin and receivables, based on your own accounts. In ENABLE, the AI writes a SQL query on your data warehouse, and the figures come from the database, not from the model.
Is AI reliable enough for controllers?
Only if you can check every answer. In ENABLE your database does the calculating and the query sits next to the result, so a controller can see which tables, filters and calculations were used. You correct a wrong definition yourself before saving the analysis.
Does OpenAI see our financial figures?
No. ENABLE only sends metadata, such as table and column names, and the user's question to the AI. Amounts and ledger entries stay in your own data warehouse and the query result is not sent back to the model.
Can I compare budget and actuals with AI?
Yes, if budget and actuals are both in your data warehouse. You can record monthly targets in ENABLE's Targets app; it stores them in your own warehouse, so a query or dashboard can put them next to actuals.
Can a branch manager see only their own figures?
Yes, in Power BI dashboards through row-level security: a profile or user tag is linked to a Power BI role. That restriction does not apply within the SQL Explorer, so only give it to people who are allowed to see the entire warehouse.
Can I export results to Excel?
Yes, via CSV. You can export every result in the SQL Explorer to CSV, which opens straight in Excel, or to JSON. Save any query you need more often.
Further reading
- SolutionAI on your own data: ask questions of business dataQuestion your data warehouse in plain language. The AI sees the structure, not your rows.Read more
- 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
- PlatformDashboard builder: build your own dashboards (pilot)Build your own dashboards on approved data, alongside your Power BI reports. In pilot.Read more
- PlatformPower BI row-level security and dashboard accessAccess profiles, Power BI row-level security, two-factor authentication and an audit log.Read more
- ResourcesRow-level security in Power BI explainedRow-level security (RLS) in Power BI is a security layer in the semantic model that determines which rows of data a user gets to see. Everyone opens the same report, but a regional manager only sees their own region and a customer only their own orders. You define RLS in roles with DAX filters: static with fixed values, or dynamic based on who is viewing.Read more
- ResourcesHow to build a KPI dashboard: steps and examplesA KPI dashboard is an overview of the key figures you use to steer towards your goals, such as revenue, margin, on-time delivery or absenteeism. A good KPI dashboard does not start with the software, but with the decisions it needs to support. You then choose a limited number of KPIs, define them, sketch the dashboard and only then build it.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.