Everyone quotes Odoo implementation timelines. Very few of them describe the real ones.
The honest baseline
For a mid-market business (50-200 users) covering CRM, sales, purchase, inventory, and accounting: 10-16 weeks end-to-end. Here's how it breaks down.
Weeks 1-2: Discovery
Two weeks of workshops and process mapping. If you're finished in one, you skipped something. If you're still discovering in week four, scope is too broad — cut and defer.
Weeks 3-8: Build
Configuration, customization, integrations. Weekly demos every Friday. This is where slippage happens: unclear requirements from discovery come back as change requests, third-party APIs turn out to be worse than documented.
Weeks 9-10: UAT
Real users, real data, on staging. If your UAT is one person clicking through happy paths for a day, you're going to have a bad go-live.
Weeks 11-12: Migration + go-live
Data migration dry runs, then final cutover. Weekend cutover is the norm. Reconciliation reports run automatically for the first 48 hours post-go-live.
Weeks 13-16: Hypercare
Your team on Odoo full-time. Our team on Slack. Issues found now are cheap to fix; issues found after AMC starts are more expensive.
What slips
Discovery slips because business processes turned out to be more variant than the initial workshop revealed. Build slips because third-party integrations were more complex than expected. UAT slips because it's the first time users actually try the system. Migration slips because source data is dirtier than anyone admitted.
The way to protect against slip is not to pad timelines. It's to demo weekly, fail small, and adjust scope early instead of pushing on hopelessly.