By Nguyen Tran, Founder, Ministers.io — five years implementing Odoo across ecommerce, retail, manufacturing, and MRO.
Most guides to choosing an Odoo implementation partner are written for a generic ERP buyer. They will tell you to check references, ask about methodology, and confirm post-launch support. That advice is correct and it will not help you, because every partner on your shortlist will pass it.
An ecommerce build fails or succeeds on integration work that a generic ERP evaluation never asks about. Can they build a product feed pipeline that survives a catalogue restructure? Have they rated freight against real dimensional data? Do they know what happens when the same SKU oversells across two channels?
This guide covers what to ask instead.
First: what “Gold Partner” actually certifies
This is the most misunderstood signal in the market, and it is worth being precise about because the badge does real work in most buying decisions.
Odoo runs three published partner tiers — Ready, Silver, and Gold. Odoo states the criteria openly on its partner programme page: new Odoo Enterprise licences sold in the trailing period, number of certified staff, and customer retention rate. Commission on Enterprise sales scales with tier, reaching up to 20% at Gold.
Read those three criteria carefully. Two of them — licences sold and certified headcount — are scale metrics that say nothing about delivery quality. The third, retention, is an indirect outcome signal, and it is the reason you should not dismiss the tier system entirely. But retention is not a published quality score, and Odoo does not publish per-partner satisfaction data, project success rates, or independent delivery audits.
Treat partner tier as evidence of scale and Odoo revenue commitment — not as certification that they can build your store. A Gold partner has sold a lot of Odoo. Whether they have ever integrated a freight carrier is a separate question, and the badge will not answer it.
There is a second-order effect worth knowing: because commission scales with Enterprise licences sold, a partner has a structural incentive toward the Custom plan and more users. That does not make anyone dishonest. It does mean you should understand why a recommendation is being made, particularly around plan tier.
The seven criteria that actually matter for ecommerce
Each of these closes with the red flag that should end the conversation.
1. Have they built the specific integrations you need?
Not “do they do integrations.” Name yours and ask for the specific build. Product feed automation for Google Merchant Center or Meta. Live carrier rating. Freight and LTL. Marketplace sync for Amazon, eBay, or your channel. POS unified with online inventory.
Ask what broke on that project and how they fixed it. Anyone who has genuinely delivered a marketplace connector has a story about inventory buffering or order status mapping, because everyone hits it. Our own worked examples include automating Google Merchant Center feeds with Odoo and automating freight quoting.
Red flag: They answer at the category level — “yes, we do integrations” — and cannot name a specific carrier, marketplace, or feed spec they have worked against.
2. Do they ask about your catalogue before quoting?
Ecommerce implementation cost is driven by catalogue complexity far more than by user count. Variants, kits, configurable products, channel-specific attributes, and units of measure all change scope materially.
A partner who quotes from a user count and a module list has priced a generic ERP deployment and will discover your catalogue in week three.
Red flag: A fixed-price quote arrives before anyone has looked at your actual product data.
3. Do they separate data cleansing from data migration?
Migration cost is driven by data quality, not volume. Clean, consistently attributed products move cheaply. Fifteen years of inconsistent SKU conventions do not.
These are two different jobs and should be two different lines on a quote. A partner who has looked at your export knows this. One who has not will quote migration as a single number and raise a change order later. Our WooCommerce to Odoo migration case study walks through what this looks like on a real catalogue.
Red flag: One line item labelled “data migration” with no discussion of what state your data is actually in.
4. Do they understand the SEO stakes of a replatform?
If you are moving from WooCommerce, Shopify, or Magento, your existing organic traffic is an asset that can be destroyed by a technically successful migration. URL structure, redirect mapping, metadata preservation, and structured data all have to survive the move.
Most ERP partners have never thought about this, because in a traditional ERP project it does not exist. In ecommerce it can be the single most expensive mistake in the build. We have written about preventing traffic and revenue loss during an ecommerce migration separately.
Red flag: Blank looks at “what’s your redirect strategy?” — or treating SEO as something your marketing agency handles after go-live.
5. Have they told you which plan tier and hosting you need, and why?
Odoo Online does not run modules containing custom Python code, which covers essentially every real shipping, freight, and marketplace connector — Odoo’s administration documentation is explicit about it. Running custom code requires Odoo.sh or on-premise, and both require the Custom plan rather than Standard — with Odoo.sh hosting billed separately from the licence, metered by worker, storage, and staging environment.
A partner who knows ecommerce will raise this unprompted in the first scoping conversation, because it determines your licensing budget. One who does not will discover it during development.
Red flag: A quote that assumes Odoo Online while also promising custom connectors. These are mutually exclusive and the quote is wrong.
6. Who owns the code, and what happens at the next version?
Odoo ships a major version annually. Custom modules need review and testing at each upgrade, and third-party App Store modules are generally licensed per major version — upgrading often means repurchasing unless the vendor bundled versions.
Establish two things in writing: who owns the custom code, and what the upgrade path costs. A partner who owns your code and charges to move it forward has you in a weak position every year.
Red flag: Vague answers on code ownership, or an upgrade position that leaves the cost to be quoted later.
7. Can they say no to you?
The most reliable quality signal in a scoping call. A good partner will push back on at least one thing you have asked for — because ecommerce operators routinely request customisation that recreates a process they should be changing instead, and every unnecessary custom module is a permanent upgrade liability.
Red flag: Enthusiastic agreement to everything. A partner who will build anything you ask has no opinion about what you should have, and heavy customisation is the single best-documented cause of Odoo timeline overrun and post-launch instability.
The question a good partner asks you
Everything above is about evaluating them. This one runs the other way, and how a partner handles it tells you more than most of their answers.
Who on your side will be the system administrator, and do they know all your workflows?
A partner who does not ask this has not scoped the engagement, because the answer changes the price more than almost anything on your requirements list.
If you have that person — one individual who can answer process questions definitively rather than convene a meeting — a large share of the configuration can sit with your team, with the agency bridging gaps: connectors, specific workflow customisations, document templates. In our experience that engagement shape runs around $5,000–$10,000.
If you do not, the partner runs the project end to end, and the cost typically moves to $10,000–$100,000. The reason is worth understanding because it is not what most quotes imply. In a complex operation, process knowledge is spread across several people — warehouse knows fulfilment, finance knows reconciliation, someone in sales knows why that customer group is priced differently. Before configuration starts, all of it has to be discovered, documented, and reconciled where two departments describe the same process differently. That discovery phase is what you are paying for, not extra software complexity.
What it costs you when your partner fills the role
We do this, and it works — but understand what you’re buying. Your agency becomes the gateway to knowledge about your own system, a genuine dependency that makes the engagement unusually sensitive to communication quality: poor collaboration risks the project and internal adoption together, since a system nobody understands is a system nobody trusts. See what an Odoo ecommerce implementation actually costs for the full picture. Without an internal system admin, negotiate the knowledge transfer plan before price — ask what documentation you get and when you could run it without them.
Red flag: A partner who never asks who owns the system on your side, or who is comfortable being the only party who understands it.
What the person approving the budget should ask
The seven criteria above are delivery questions. If you are the one signing the contract rather than running the project, these four are yours.
How much of the quote is fixed, and what triggers a change order? Ecommerce scope moves when catalogue complexity or data quality turns out worse than assumed. Ask what specifically would trigger a change order and what the approval process is. A partner who cannot answer has not scoped defensively.
What is the year-two number? Year one is the implementation. Year two is licence renewal at the standing rate rather than the promotional one, the annual version upgrade, module re-purchases, and connector maintenance. Ask for year two in writing before signing year one.
What is the exit cost? If this relationship ends, what do you keep and what does it cost to move? This is the code ownership question in financial terms, and it is the one most likely to be answered vaguely.
What is the total cost across three years, not the quote? Odoo’s licence advantage is real but it is the smallest line. Comparing partner quotes on implementation price alone ignores upgrade exposure, which is where a heavily customised build gets expensive. We break the full cost picture down in what an Odoo ecommerce implementation actually costs.
A partner who volunteers the year-two and exit numbers before you ask is telling you something useful about how they run projects.
Questions to take into a vendor call
Print this. Ask all of them. The second column is what to listen for.
| Question | What you’re testing |
|---|---|
| What does this quote assume we’ll do ourselves? | Whether the engagement model is explicit |
| Who should our system admin be? | Whether they’ve scoped the engagement at all |
| What’s the knowledge transfer plan? | Whether you’ll ever own the system |
| Which of our integrations have you built before? | Specific delivery, not category claims |
| What broke on that project? | Real experience produces real stories |
| Can we see our catalogue in your quote? | Whether they’ve looked at your data |
| Is cleansing priced separately from migration? | Whether they’ve seen your export |
| Which plan tier and hosting does this assume? | Odoo.sh awareness before development |
| What’s the redirect strategy? | Whether they understand replatform SEO |
| Who owns the custom code? | Lock-in exposure |
| What does the next version upgrade cost? | Year-two budget honesty |
| Which of our requests would you push back on? | Whether they have professional judgement |
| Who is on the team, and are they in-house? | Outsourcing and continuity risk |
| Can we speak to a client directly? | References beyond curated case studies |
| What happens if we’re off track at week eight? | Risk maturity |
The three that separate ecommerce specialists from generalists are integrations, redirects, and plan tier. A generalist ERP partner will handle most of the rest well. The three engagement questions at the top separate partners who have scoped your project from partners who have priced a template.
A note for Australian buyers
The Australian Odoo partner field competes largely on generic ERP breadth — localisation, GST and ATO payroll compliance, and industry range. Those matter and you should confirm them.
But the same ecommerce-specific gap applies, and arguably more sharply: freight is a harder problem in Australia than in most markets, because domestic freight pricing, regional surcharges, and carrier coverage vary in ways that generic carrier integrations handle badly. If you ship anything bulky domestically, ask specifically about Australian freight rating rather than accepting “we integrate carriers.”
Where partner directories fall short
A large share of search results for Odoo partners are “top 10” roundups. They are useful for building a shortlist and close to useless for choosing from it, because they rank on partner tier, company size, and review counts — none of which tell you whether a partner has built a marketplace connector.
Use the directories to assemble a list. Use the questions above to cut it down. The single most useful filter is asking each shortlisted partner to name an ecommerce integration they have delivered that matches yours. Most lists shorten quickly.
Frequently asked questions
What does Odoo Gold Partner status actually mean?
Odoo’s Ready, Silver, and Gold tiers are set by three published criteria: new Odoo Enterprise licences sold, number of certified staff, and customer retention rate. Commission scales with tier up to 20% on Enterprise sales. Two of the three criteria measure scale rather than delivery quality, and Odoo does not publish per-partner project success or satisfaction data — so treat the tier as evidence of scale and Odoo commitment, not proof of implementation capability.
How do I evaluate an Odoo implementation partner for an ecommerce project specifically?
Ask which of your specific integrations they have built before — product feeds, carrier and freight rating, marketplace sync, POS-to-online inventory, catalogue migration — and ask what went wrong on those projects. Confirm they have separated data cleansing from migration, that they have a redirect strategy for the replatform, and that they have told you which Odoo plan tier and hosting your build requires.
How do I evaluate a partner if nobody internally knows all our processes?
Look for a partner who treats discovery as a scoped, priced phase rather than something bundled in for free — and who proposes a knowledge transfer plan before you ask for one. That is exactly what the system-admin question above is designed to surface: a partner who has handled this scenario before will raise it unprompted. One who quotes a flat implementation fee without asking who knows your processes has not scoped the engagement.
Should I hire a freelance Odoo developer instead of an agency?
For a single-app deployment with low risk, a freelancer can work. For an ecommerce build with multiple integrations, the risk is continuity: connectors need maintenance at every annual Odoo version, and a single developer is a single point of failure. Ask any agency the same continuity question — an agency that subcontracts the work has the same exposure without the transparency.
Do I need a partner in my own country?
For ecommerce, less than you might think — most of the work is integration rather than on-site. Local presence matters most for tax, payroll, and compliance localisation, and for onsite POS rollouts. Odoo bills travel time separately on its own onsite consulting, and most partners do the same.
What’s the biggest mistake buyers make choosing an Odoo partner?
Selecting on price or partner tier without confirming ecommerce integration experience. The cheapest quote is usually the one that has not yet discovered your catalogue complexity, your data quality, or the fact that your custom connector requires Odoo.sh.
Evaluating partners for an Odoo ecommerce build?
We implement Odoo for ecommerce businesses — feeds, freight, marketplaces, POS, and migrations. Ask us the questions above.
About the author
Nguyen Tran is the founder of Ministers.io. He has spent five years working on Odoo ERP implementations — five delivered directly and more than twenty advised on — across ecommerce, retail, manufacturing, and technical services (MRO).
That range is deliberate context for this guide. Most of an ERP implementation looks the same whatever the industry; the parts that don’t are the parts that decide an ecommerce project.
