Data to AI™ / Claude/AFAS
Controllers on AFAS Profit ask Claude what changed in their finance data and how this month compares, reading the GetConnectors their own environment defines.
Better answers, fewer tokens.
Claude
Which article groups had the most articles modified last week, and what is their average sales price?
ArtGroup 100 changed most, with 86 articles by ModifiedDate and an average SalesPrice of €42.10. All three groups shown were edited during last week's price update.
| ArtGroup | Articles by ModifiedDate | Average SalesPrice |
|---|---|---|
| 100 Profit_Article | 86 | €42.10 |
| 200 Profit_Article | 54 | €118.40 |
| 300 Profit_Article | 21 | €9.75 |
AFAS has no date range, so each run pulls current articles up to the Limit; the week is counted on ModifiedDate, and Blocked articles are left out of the average.
AFAS Profit is Dutch business software for finance, HR, payroll and CRM. Dataddo extracts whichever GetConnector you have set up in AFAS, with the fields you pick, custom fields included. The supplied GetConnectors shown here, such as Profit_Article and Profit_Employee_Leaves, are examples from the AFAS documentation.
In the AFAS Profit_Article flow, list the articles with a ModifiedDate since the 1st of this month, grouped by ArtGroup, with their SalesPrice.
During the close, the controller asks Claude which articles were edited after the period started and checks the price changes before the books are locked.
Show the average SalesPrice per ArtGroup in Profit_Article, leaving out Blocked articles, and list the five groups with the widest price range.
Ahead of the pricing review, the finance lead gets price levels per article group in one answer and follows up on the groups with the widest spread.
Use the dataddo server to fetch Profit_Article with ItemCode and SalesPrice and compare it with pricing/price-list.csv in this repository by ItemCode.
A finance engineer runs Claude Code next to the price-list scripts and commits the differences together with the query it used.
Data to AI™ is priced per seat, like the AI assistants your team already uses, with ~25% off on annual billing. Need tailored contracts or volume pricing? Talk to us about Enterprise.
Serve governed business data to Claude, ChatGPT and other AI tools over MCP.
$10.49/month
per seat
3 seats minimum
Flexibly deploy Dataddo in any cloud or hybrid environment.
Custom
We have a payment model that works for you
| Without Dataddo |
With Dataddo
|
Outcome for you | |
|---|---|---|---|
| Getting AFAS data into Claude | Someone opens a GetConnector in AFAS, exports it to a spreadsheet and attaches the file to a Claude chat, one export at a time. | Claude reads each AFAS flow through Dataddo Data Access, with the GetConnector fields you picked, custom fields included. | No more manual exports |
| Large GetConnectors | An export stops at whatever the screen or file can hold, and nobody can tell whether older records were cut off. | Historical data is loaded in batches of up to 50,000 records with Limit and Skip, with upsert keeping the warehouse free of duplicates. | Complete history to ask about |
| Knowing what changed | Comparing two exports by hand is the only way to see which entries moved since the last check. | Where the GetConnector has a modification date, such as ModifiedDate in Profit_Article, Claude filters on it, and dataddo_extraction_timestamp shows when each row was read. | Changes found in one question |
| Access for the finance team | Each controller keeps a private export, so two people asking the same question get two different numbers. | A Claude Team or Enterprise Owner adds the connector once, and every member signs in with their own Dataddo account. | One version of the figures |
| Token consumption | Claude spends tokens exploring raw AFAS columns, sampling rows and retrying until it understands the data. | Field definitions and relations are supplied up front, so the model skips the exploration and goes straight to the answer. | Fewer tokens per answer |
| Answer quality | Prone to wrong joins, invented fields and numbers that are hard to check. | Answers grounded in defined fields. In the Dataddo benchmark, 78.5% of questions were answered correctly, against 65.5% with plain CSV files. | Better answers |
The AFAS datasets finance and RevOps teams use most, with their real field names. Claude queries them by name through the Dataddo semantic layer.
| Dataset | Metrics and attributes |
Ask Claude
|
|---|---|---|
| Profit_Article (example GetConnector) |
|
Which articles have a ModifiedDate this month, and what is their SalesPrice by ArtGroup? |
| Profit_Employee_Leaves (example GetConnector) |
|
How many leave records have a StartDate in each month of this Year? |
| ProfitGetConnectoren (example GetConnector) |
|
Which GetConnectors does our AFAS environment offer, by Naam and Omschrijving? |
| Date and key fields |
|
How many records were created on each day this month in the GetConnector we extract? |
We asked 43 pre-registered questions of three real datasets - Google Search Console, HubSpot deals and Google Ads - and delivered the same rows to the same model in four ways. Only the data layer changed.
of questions answered correctly - the best of the four delivery methods tested
fewer confidently wrong answers - 14.0% of runs against 24.0% and 26.3%
more correct answers on messy CRM data - better on 8 questions, worse on none
Same model, prompts and rows in every condition; 759 graded runs, ground truth frozen before any run. On Google Search Console alone, documented CSV files did as well as Data to AI™ - the advantage comes from messier data such as HubSpot deals.
Read the benchmark →Authorize AFAS in Dataddo with your Environment Type, Environment ID and an AFAS token from an App Connector, then create one source per GetConnector with the fields you need. Attach the flows to a Claude destination or AI Model and connect Dataddo Data Access from the Claude connectors directory.
An AFAS token, which you get by creating an App Connector in AFAS, plus your Environment ID and Environment Type (Profit, Accept or Test). Claude never sees the token; it only reads what Dataddo extracts.
Whichever GetConnector you have set up in AFAS and extract, with the fields you choose, custom fields included. Supplied GetConnectors such as Profit_Article and Profit_Employee_Leaves, named on this page, are examples.
Each run pulls records up to the Limit in the source's advanced settings. Load history in batches of up to 50,000 by raising Skip, ideally into a warehouse with upsert, and Claude can then query all of it.
Yes. Signed in with your Claude account, Claude Code loads the Dataddo connector, so you can query your AFAS flows from the terminal while you work in a repository.
No. Claude queries extracted data through the semantic layer by entity and field name. Dataddo never accepts SQL from the model and nothing is written back to AFAS.
Yes. Claude shows the connector's tool permissions, and for each tool you decide whether Claude may use it freely or has to ask you first.
