Data to AI™ / Copilot/Odoo

Connect Odoo to Microsoft Copilot

Sales, operations and finance staff ask a Teams agent about Odoo invoices and orders from your own instance, while finance controls which models the agent can reach.

Better answers, fewer tokens.

  • Lower AI costs for the same work
  • More accurate, reliable answers
  • Read-only
  • Uses the Odoo data you already sync
  • SOC 2 Type II certified
Microsoft 365 Copilot
Connected to Dataddo

On which day last week did our invoice records change the most?

Copilot Answer · Odoo

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.

DayRecords by Created OnRecords by Last Updated On
Mon 21 Sep Odoo Table Replication 64118
Tue 22 Sep Odoo Table Replication 4187
Wed 23 Sep Odoo Table Replication 3872

Counted on one replicated model filtered by Last Updated On, the date range field the docs recommend against Odoo pagination duplicates.

source
Odoo · as of Mon 06:40 UTC
covers
21-27 Sep 2026
Example answer · fictional account and figures
Odoo in Copilot

Questions Copilot can answer from your Odoo data

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.

Invoice status in the sales channel

Ask Copilot

@Odoo Ledger Which invoice records for the customer I name were Last Updated On this week, and what is their monetary total?

Account managers mention the Odoo agent in Teams before a customer call, so finance stops answering the same invoice status questions one message at a time.

Monthly forecast versus actual

Ask Copilot

@Odoo Ledger Total the monetary amount of invoice records per month this year in a table I can set against our forecast.

The FP&A lead asks the agent in Microsoft 365 Copilot for invoiced totals by month and discusses the gaps to forecast with budget owners in the same thread.

Typed Odoo models in VS Code

Ask Copilot

Use the dataddo tools to describe the Odoo sales order entity and generate a TypeScript interface with each field and its type.

A developer building finance reports asks GitHub Copilot in Agent mode for the real fields of this instance, so the code covers custom fields too.

Odoo datasets and the questions they answer in Copilot →
Pricing

Priced per seat.

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.

Data to AITM

Serve governed business data to Claude, ChatGPT and other AI tools over MCP.

$10.49/month

per seat

3 seats minimum

Enterprise

Flexibly deploy Dataddo in any cloud or hybrid environment.

Custom

We have a payment model that works for you

See all plans and features →
Setup

Connect Odoo to Copilot step by step

  1. 1
    Connect Copilot to Dataddo Add Dataddo as an MCP tool to a Copilot Studio agent and sign in, or add it to mcp.json for GitHub Copilot in VS Code.
  2. 2
    Connect Odoo to Dataddo Sign in with your Odoo account in Dataddo and choose the datasets to sync, such as Odoo Table Replication (one model per source) and Date range fields.
  3. 3
    Choose what Copilot can see Select the Odoo data Copilot may query. Access is read-only, and data you don't select stays out of reach.
Copilot setup guide →
With vs. without

Odoo in Copilot, with and without Dataddo

Without Dataddo With Dataddo Outcome for you
Invoice questions in Teams Account managers ask finance in chat whether a customer was invoiced, and someone looks it up in Odoo by hand. A Copilot Studio agent published to Teams answers from the replicated invoice model when it is mentioned with @. Status answered in the channel
Which models the agent reaches Sharing Odoo exports in a Teams chat gives everyone in it every column of every model in the file. The agent only sees the models and columns finance attached, and the server description says when to call Dataddo. Finance sets the scope
Group companies Each company is exported separately after switching company in Odoo, and the agent never sees the group picture. A custom filter on the source pulls more than one company, so the agent can answer per company or for the group. One agent for the group
Developer access to fields Developers ask for sample exports to learn which datetime and many2one fields a model holds in this instance. GitHub Copilot in VS Code describes the replicated entities through .vscode/mcp.json, with every field type included. Types from the real schema
Token consumption Copilot 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. Fewer tokens per answer
Answer quality Prone to wrong joins, invented fields and numbers that are hard to check. Answers built on agreed definitions. In the Dataddo benchmark, Opus 5.5 reached 95.2% good outcomes, against 81% over raw SQL. Better answers
Datasets

Odoo datasets and the questions they answer in Copilot

The Odoo datasets finance and RevOps teams use most, with their real field names. Copilot queries them by name through the Dataddo semantic layer.

Dataset Metrics and attributes Ask Copilot
Odoo Table Replication (one model per source)
  • model
  • custom_filter
  • columns
  • language
  • date_range_field
  • + others
Which replicated Odoo models use a custom_filter, and what does each filter restrict?
Date range fields Which records were Created On over a year ago but still have a recent Last Updated On?
Replicated field types How did the monthly sum of a monetary field change this quarter, grouped by its date field?
Dataddo metadata columns
  • dataddo_extraction_timestamp
  • dataddo_hash
  • + others
Did any scheduled extraction fail to run this week, judging by gaps in dataddo_extraction_timestamp?
See all Odoo datasets, metrics and attributes on the connector page →
Benchmark

Fewer confident mistakes. No leaked personal data.

We asked three Claude models the same 21 business questions about a NetSuite company built from real-world data: once over a raw SQL connection, once as CSV exports and once through Data to AI™. Only the connection changed, and the answer key was fixed before the first run.

95.2%

good outcomes with Opus 5.5 on Data to AI™, against 81% on a raw SQL connection

58%

fewer confidently wrong answers with Sonnet 5 - 8 of 63 runs, against 19 on raw SQL

24x

cheaper per question: Haiku 5.5 on Data to AI™ (87.3%) beat Opus 5.5 on raw SQL (81%)

0 of 9

attempts disclosed a sales rep's email address, against 9 of 9 on raw SQL

Good outcomes Correct answer or correct decline, 63 runs per setup - higher is better
Wrong, but stated as fact Confidently wrong answers out of 63 runs - lower is better

21 questions, 3 runs each. 17 carry a real-world caveat (transaction types, voided invoices, foreign currencies, group companies billed as customers, a partial year) and 3 can't be answered from the data. Ground truth computed directly in SQL; CSV exports were run with Sonnet 5 only. Run date: 26 October 2026.

Read the benchmark →
FAQ

Odoo and Copilot: FAQ

How do I connect Odoo to Copilot?

Replicate the Odoo models you need, for example invoices and sales orders, as Dataddo sources and attach the flows to a Copilot destination or AI Model. In Copilot Studio, add a Model Context Protocol tool with https://headless.dataddo.com/mcp-data, choose OAuth 2.0 with Dynamic discovery and publish the agent to Teams and Microsoft 365 Copilot.

How does a Teams user get answers from Odoo?

Through a Copilot Studio agent with Dataddo Data Access as its MCP tool, published via Channels > Teams and Microsoft Copilot. The agent reads replicated Odoo data, never Odoo itself.

What server description works for an Odoo agent?

Name the Odoo models you replicate, for example invoices and sales orders from Odoo. The agent uses the description to decide when to call Dataddo instead of answering from its own knowledge.

Can GitHub Copilot read our Odoo models?

Yes. Put the Dataddo URL in .vscode/mcp.json with type http, start the server and sign in, then use Agent mode. On Copilot Business and Enterprise an admin must enable the MCP servers in Copilot policy.

Why does the agent not know our Odoo field names in advance?

Odoo fields depend on your instance and the columns you pick. The agent can call describe_entity to learn them, so custom fields need no extra setup in Copilot Studio.

The agent reports duplicate Odoo invoices. What now?

Odoo's pagination can cause duplicates. Set the date range field on the source to Last Updated On and add dataddo_hash as a natural key, or ask the Dataddo Solutions Team for help.

How far back does Odoo data in Copilot go?

Each run reads the relative window you set on the date range field. For older records, run a full data re-sync with a wider range, then ask the agent again.

New from Dataddo · Data to AI™

Ask Claude or ChatGPT about your ads, analytics and CRM

“What did a new lead cost us last week, by paid channel?”

Connect 400+ sources once and get the answer right in your AI chat, without building a report. Better answers: your metrics are defined once, so the AI doesn't have to guess.