If you run Odoo and you want to sell online in this region, you have two honest options. Buy a ready-made connector for a platform like Salla, Zid, Shopify or WooCommerce, or build the integration yourself against a storefront you control.

Almost everything written about this online is either a module listing or a sales page. Neither tells you which one you should choose. This is the version we would want if we were the ones deciding.

What a connector actually gives you

A connector is a mapping. It moves products, stock, prices and orders between Odoo and a platform, on a schedule, using the fields both sides already agree on. When your business fits inside that agreement, a connector is the right answer and building anything custom is a waste of money.

Buy the connector when most of these are true:

  • Your catalogue is straightforward — one price per product, standard tax, no customer-specific pricing.
  • You are happy living inside the platform's checkout, exactly as it ships.
  • Your order volume is comfortable with scheduled syncing rather than immediate.
  • You do not need a mobile app that behaves differently from the website.
  • Nobody in the business is going to ask for a workflow the platform does not have.
Worth saying plainly

Most small catalogues should buy a connector. If this describes you, stop reading and go buy one — you will be live in weeks for a fraction of the cost.

Where connectors run out of road

The trouble starts when your business does something the mapping did not anticipate. In practice it is almost always one of five things.

  • Price lists. You charge trade customers differently from retail. A connector that assumes one price per product cannot express that, and the workarounds — duplicate products, manual overrides — create exactly the second source of truth you were trying to avoid.
  • Tax that has to match the ERP exactly. Fiscal positions, tax groups and rounding rules configured for how you already invoice. Close enough is not close enough when finance reconciles.
  • Payment rails the platform does not carry. Cash on delivery with a variable fee, Fawry codes, local wallets, instalments — and all of it landing back in Odoo in a shape the finance team recognises.
  • A catalogue that is large or image-heavy. Thousands of full-resolution images pulled from the ERP will make a storefront unusable on mobile data unless something sits in between resizing and caching them.
  • A mobile app that is not just the website in a wrapper. The moment you want native behaviour, you are building against an API anyway.

The honest cost comparison

A connector is cheap to buy and cheap to start. Its real cost shows up later, in three forms: the workarounds you accumulate to express things the mapping cannot, the manual reconciliation someone quietly does every week, and the ceiling you hit when the business wants something the platform will not do.

A custom integration costs more up front and less to bend. It is the right choice when the thing you are building is a real channel for the business rather than an experiment — and when the cost of the storefront being wrong is measured in reconciliation hours rather than in mild annoyance.

Buy the connector to test whether the channel works. Build the integration once you know it does.

A test you can run this week

Before you commit either way, write down the five things your business does that a generic online store would find strange. Different prices for different customers. A delivery fee that depends on the governorate. An approval step before certain orders ship. A product that is sold by weight. Anything at all.

Then take that list to whoever is selling you the connector and ask them to show you each one working. Not describe it — show it. The list is your specification, and the answers are the cheapest research you will ever do.

What we did

For SG-Mart, a distributor with more than 110 categories, customer-specific pricing and cash-on-delivery economics, the connector route would have needed workarounds around every one of those. We built the storefront and the Flutter app against Odoo directly, with Odoo remaining the single source of truth for catalogue, price and tax.

That was the right call for them. It is not automatically the right call for you — and any vendor who tells you otherwise before asking about your price lists is selling, not advising.

More on what actually breaks when you connect a storefront to Odoo, and on how we approach ERP integration. If you want a straight read on which side of this you fall, tell us what you are running.