Data to AI™ / Claude/Odoo
Finance teams on Odoo ask Claude about the exact models and custom fields their instance uses, from daily changes to totals per company, without writing exports.
Less tokens, better answers.
Claude
On which day last week did our invoice records change the most?
Monday 21 Sep was the busiest day, with 64 invoice records Created On and 118 Last Updated On. Updates outnumbered new records on each day shown.
| Day | Records by Created On | Records by Last Updated On |
|---|---|---|
| Mon 21 Sep Odoo Table Replication | 64 | 118 |
| Tue 22 Sep Odoo Table Replication | 41 | 87 |
| Wed 23 Sep Odoo Table Replication | 38 | 72 |
Counted on one replicated model filtered by Last Updated On, the date range field the docs recommend against Odoo pagination duplicates.
Odoo is an open-source ERP with modules for accounting, inventory, sales, CRM and HR. Dataddo replicates whole Odoo models: you pick a model from your instance and its columns, including custom fields, so the fields match your setup.
Use the dataddo server to describe every replicated Odoo entity, then write docs/odoo-fields.md listing each field with its type, custom fields marked.
A finance engineer runs Claude Code in the reporting repository, so custom fields are documented from the real schema of this instance instead of from memory.
Which records in the Odoo invoice flow have Last Updated On since yesterday, and how did their monetary totals change per customer?
Each morning the accountant asks Claude what moved overnight in the replicated invoice model, and checks the larger changes before the day's reconciliation work.
Sum the monetary total field per company for September across the companies our custom filter includes, and show the swing against August.
With a custom filter that pulls several companies, the group controller gets monthly totals per company in one answer and asks about the biggest movements.
Data to AI™ is free while in private pre-release. We onboard teams in waves and shape the product with their feedback. At launch, it will be priced like the AI assistants your team already uses. Need tailored contracts or volume pricing? Talk to us about Enterprise.
Serve governed business data to Claude, ChatGPT and other AI tools over MCP.
Free during pre-release
Flexibly deploy Dataddo in any cloud or hybrid environment.
We have a payment model that works for you
| Without Dataddo |
With Dataddo
|
Outcome for you | |
|---|---|---|---|
| Getting Odoo data into Claude | Someone exports a list view per model from Odoo and attaches the file to a Claude chat, one model and one company at a time. | Claude reads each replicated Odoo model through the Dataddo Data Access connector, with the columns you picked, custom fields included. | No list view exports |
| Instance-specific fields | Claude sees raw column names from your instance and guesses what a custom field or a many2one value means. | Claude calls describe_entity first and gets every field of your model with its type, so answers use your own field names. | Fields read as defined |
| Duplicate records | Duplicates caused by Odoo pagination slip into exported files, and totals in the chat come out too high. | Filter the source on Last Updated On, as the docs recommend, and add dataddo_hash as a natural key for upserts. | Totals without double counts |
| Claude Code on the same data | Developers request a fresh Odoo export whenever they need to check a field name while writing reporting code. | Claude Code reaches the same replicated models after claude mcp add, signing in with /mcp or claude mcp login dataddo. | One source for chat and code |
| Token consumption | Claude spends tokens exploring raw Odoo 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. | Less 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 Odoo datasets finance and RevOps teams use most, with their real field names. Claude queries them by name through the Dataddo semantic layer.
Runs on your schedule. Each run reads a relative window on the date field you choose, such as Last Updated On, and a full data re-sync loads older records.
Odoo connector →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 Odoo in Dataddo with your host, username, password and database, then create one Odoo Table Replication source per model you need, such as your invoice model. Attach the flows to a Claude destination or AI Model, add https://headless.dataddo.com/mcp-data as a custom connector in Claude, or run claude mcp add in Claude Code.
Only the models you replicate. Each Odoo source covers one model of your instance and the columns you tick, custom fields included. Claude reads them once the flow is attached to an AI destination or AI Model and has run.
Fields come from your own instance, and column labels follow the language set on the source. Ask Claude to describe the entity first so it uses your real field names and types in every answer.
Yes. Register the server with claude mcp add, sign in with /mcp or claude mcp login dataddo, and Claude Code can describe and query the replicated Odoo entities from your terminal.
A bug in Odoo's pagination can cause duplicates. Set the date range field to Last Updated On, add dataddo_hash as a natural key, or ask the Dataddo Solutions Team to help load the additional values.
Yes, if the source extracts them. By default only the selected company is extracted. Add a custom filter on the Models and columns step to include other companies, then ask Claude per company.
The Dataddo docs ask for a custom plan that includes access to the Odoo API. Without API access, Dataddo cannot extract the data Claude would query.