Can Power BI connect directly to an Odoo Online database?

No. Odoo Online and Odoo.sh do not expose PostgreSQL to the internet; Odoo's own staff state that direct access conflicts with their security, performance, and uptime commitments. For those hostings we use Odoo's external API (XML-RPC / JSON-RPC) through a connector or on-premises data gateway, or scheduled exports into a warehouse. Direct PostgreSQL connections are only an option for on-premise or self-hosted Odoo.

Will Power BI slow down our Odoo?

Not if the data path is designed for it. On self-hosted Odoo we point Power BI at a read replica, never the transactional database. On the API path we use incremental refresh keyed on write_date, batch requests, and run pulls off-peak, so the load is a fraction of what a naive full-refresh connector would generate.

How fresh is the data?

Read replica: near real time with DirectQuery, or hourly with import mode. API path: typically hourly to every-4-hours incremental refresh; Power BI Pro workspaces allow up to 8 scheduled refreshes a day, Premium / Fabric capacity more. Warehouse path: nightly is the norm. Truly real-time operational screens belong in Odoo, not Power BI.

What Power BI licences do we need?

Report authors and most viewers need Power BI Pro; Premium Per User adds larger models and more frequent refresh; Fabric capacity lets viewers consume without per-user Pro licences at scale. Microsoft's pricing page (checked 13 August 2026) lists Power BI Pro at USD 14 per user per month and Premium Per User at USD 24 per user per month, billed yearly, with regional variation. Many Microsoft 365 E5 tenants already include Pro — we check before you buy anything.

How do you handle multi-company Odoo and data security?

One dataset, with a company dimension and row-level security roles mapped to Microsoft 365 users or groups. A country manager sees their company; group finance sees all. Credentials into Odoo or the replica are read-only, and nothing personal beyond what the report needs (e.g. no payroll fields) enters the model unless explicitly scoped.

Should we use Power BI or TechUltra's AI BI Dashboard app?

Different jobs. The AI BI Dashboard app runs inside Odoo 19 and lets a user type an English question and get KPI cards, charts, and a backing record list — ideal for quick, in-Odoo answers by Odoo users. Power BI is for governed cross-module models, history, DAX measures, and sharing with people who don't have Odoo logins. Several clients run both.

What does it cost and what is the first step?

Scoped fixed-price after a 30-minute call. Bring: Odoo hosting and version, the KPI list (even rough), number of companies, and whether you have Microsoft 365 / Power BI already. We reply with a data-path recommendation and a week-by-week plan.

BI & Reporting Integration

Power BI ↔ Odoo Integration

Put Odoo data into Microsoft Power BI without breaking your Odoo instance or your API quota. TechUltra builds the data path (read replica, API gateway, or scheduled warehouse loads), models the dataset as a star schema over sales, accounting, and stock, sets refresh and row-level security, and hands finance and operations dashboards they can share with people who never log in to Odoo.

Last reviewed: August 13, 2026

What it is

A reporting layer that reads Odoo data into Power BI datasets and dashboards. Which data path we use depends on your hosting: on-premise or self-hosted Odoo can expose a PostgreSQL read replica that Power BI queries directly; Odoo Online and Odoo.sh do not expose PostgreSQL to the internet (Odoo states this conflicts with its security and uptime commitments), so those instances go through Odoo's external API (XML-RPC / JSON-RPC) behind a connector or on-premises data gateway, or through scheduled exports into a warehouse such as Azure SQL, Fabric Lakehouse, or PostgreSQL. On top of the data path we build the semantic model: fact tables from sale_order_line, account_move_line, stock_move, and purchase_order_line; dimensions for date, product, partner, company, warehouse, salesperson, and analytic account. Power BI reports are then authored, published to a workspace, and secured with row-level security so a regional manager sees only their region.

Why it matters

Odoo's built-in reporting is good inside a module and weak across modules. Finance wants gross margin by product family next to stock ageing; operations wants on-time delivery next to open purchase commitments; leadership wants one page, refreshed nightly, that they can open on a phone. Odoo can be bent to do some of that with custom SQL views, but every view is developer work, and Odoo users need a licence to see it. Power BI solves the cross-module and the sharing problem: one model over all the tables, distributed to people who are not Odoo users, with Microsoft 365 identity. The risk is doing it badly — hammering the production database with ad-hoc queries, or exhausting API rate limits with a connector that pulls every record every hour. The integration exists to get the value without those failure modes.

Features

Three data-path options, chosen by hosting

PostgreSQL read replica for on-premise / self-hosted (fastest, DirectQuery-capable, no load on the transactional DB). Odoo external API via connector or Power BI on-premises data gateway for Odoo Online / Odoo.sh. Scheduled export to a data warehouse when history retention or multi-source joins matter.
Star-schema dataset modelling

Fact tables built from sale, account (journal items), stock moves, and purchase lines. Conformed date, product, partner, company, warehouse, and salesperson dimensions. Measures written in DAX for revenue, COGS, gross margin, DSO, stock turns, fill rate, and on-time delivery.
Refresh cadence that fits the source

Read replica: DirectQuery or hourly import. API path: incremental refresh keyed on write_date, so only changed records are pulled and Odoo's rate limits are respected. Warehouse path: nightly or every-4-hours loads with full history retained.
Row-level security (RLS)

RLS roles mapped from Odoo companies, sales teams, or warehouses to Microsoft 365 users or groups. Multi-company Odoo tenants get one model with per-company visibility rather than one report per company.
Finance and operations report packs

Starter reports: P&L and balance sheet trend, AR / AP ageing, cash position, sales by product / customer / region, margin analysis, stock valuation and ageing, purchase commitments, manufacturing WIP. Extended per client.
Sharing beyond Odoo users

Board members, franchisees, investors, and field staff view Power BI reports through their Microsoft account without an Odoo licence. Embedding in Teams or SharePoint is supported.
Odoo-side complement: AI BI Dashboard

For users who want answers inside Odoo, TechUltra's AI BI Dashboard app (Odoo 19) turns plain-English questions into KPI cards, charts, and a drill-down record list. It complements Power BI — quick in-Odoo questions there, governed cross-module reporting and external sharing in Power BI. See /odoo-apps/ai-bi-dashboard.
Monitoring and hand-over

Refresh-failure alerts, dataset documentation (table lineage, measure definitions), and a training session for the finance or BI owner so the model can be extended in-house.

How it works

Discovery and KPI definition

Agree the 15–30 KPIs that matter, who consumes them, and refresh expectations. Confirm hosting (on-premise, Odoo.sh, Odoo Online), Odoo version, multi-company structure, and Microsoft licensing already in place.
Data-path selection and setup

Provision the read replica, deploy the API connector / gateway, or configure the warehouse pipeline. Credentials are read-only and scoped; API pulls are incremental from day one.
Model and measures

Build the star schema and DAX measures in Power BI Desktop. Reconcile totals against Odoo's own reports (trial balance, sales analysis, stock valuation) before any visual is designed.
Reports, RLS, and publishing

Author the report pages, configure row-level security roles, publish to the Power BI workspace, and set the refresh schedule.
UAT and go-live

Finance and operations sign off on numbers, not on visuals. Refresh runs unattended for one week; alerts and documentation are handed over.

Deployment timeline

Read-replica or API-connector setup with the starter finance + sales report pack: 3–4 weeks. Adding operations (stock, purchase, manufacturing) and row-level security across multiple companies: 5–6 weeks. Warehouse-based builds with history back-fill or joins to non-Odoo sources (payroll, e-commerce, marketing): 6–10 weeks. Ongoing model extension is usually delivered as monthly support hours.

Best for

Companies on Odoo (any hosting) that already run Microsoft 365 and want cross-module KPIs shared with people outside Odoo — CFOs, boards, franchise networks, group finance consolidating multi-company Odoo tenants, distributors tracking margin and stock ageing together, manufacturers watching OTIF and WIP. Less of a fit for teams that only need per-module views inside Odoo; for those, Odoo's built-in reporting or the AI BI Dashboard app is cheaper and faster.

Frequently asked questions

Can Power BI connect directly to an Odoo Online database?

No. Odoo Online and Odoo.sh do not expose PostgreSQL to the internet; Odoo's own staff state that direct access conflicts with their security, performance, and uptime commitments. For those hostings we use Odoo's external API (XML-RPC / JSON-RPC) through a connector or on-premises data gateway, or scheduled exports into a warehouse. Direct PostgreSQL connections are only an option for on-premise or self-hosted Odoo.
Will Power BI slow down our Odoo?

Not if the data path is designed for it. On self-hosted Odoo we point Power BI at a read replica, never the transactional database. On the API path we use incremental refresh keyed on write_date, batch requests, and run pulls off-peak, so the load is a fraction of what a naive full-refresh connector would generate.
How fresh is the data?

Read replica: near real time with DirectQuery, or hourly with import mode. API path: typically hourly to every-4-hours incremental refresh; Power BI Pro workspaces allow up to 8 scheduled refreshes a day, Premium / Fabric capacity more. Warehouse path: nightly is the norm. Truly real-time operational screens belong in Odoo, not Power BI.
What Power BI licences do we need?

Report authors and most viewers need Power BI Pro; Premium Per User adds larger models and more frequent refresh; Fabric capacity lets viewers consume without per-user Pro licences at scale. Microsoft's pricing page (checked 13 August 2026) lists Power BI Pro at USD 14 per user per month and Premium Per User at USD 24 per user per month, billed yearly, with regional variation. Many Microsoft 365 E5 tenants already include Pro — we check before you buy anything.
How do you handle multi-company Odoo and data security?

One dataset, with a company dimension and row-level security roles mapped to Microsoft 365 users or groups. A country manager sees their company; group finance sees all. Credentials into Odoo or the replica are read-only, and nothing personal beyond what the report needs (e.g. no payroll fields) enters the model unless explicitly scoped.
Should we use Power BI or TechUltra's AI BI Dashboard app?

Different jobs. The AI BI Dashboard app runs inside Odoo 19 and lets a user type an English question and get KPI cards, charts, and a backing record list — ideal for quick, in-Odoo answers by Odoo users. Power BI is for governed cross-module models, history, DAX measures, and sharing with people who don't have Odoo logins. Several clients run both.
What does it cost and what is the first step?

Scoped fixed-price after a 30-minute call. Bring: Odoo hosting and version, the KPI list (even rough), number of companies, and whether you have Microsoft 365 / Power BI already. We reply with a data-path recommendation and a week-by-week plan.