The Shopify-or-custom argument usually gets settled by whoever is louder. It is actually a question with a clear answer, and the answer depends on three specific requirements rather than on taste or budget.
Shopify is right more often than developers admit
If you sell a fixed catalogue at published prices to anyone who turns up with a card, Shopify will do it faster and cheaper than a custom build, and it will keep doing it while you sleep. A studio that talks you out of it is usually selling its own hours. We build on it happily when it fits.
The three requirements that break it
| Requirement | Why it breaks | What it needs instead |
|---|---|---|
| Account-specific pricing | Two customers see different prices for the same product, sometimes negotiated per line | A pricing model tied to the customer record, not the product |
| Credit terms | The order is placed now and paid in thirty days, against a limit | Credit limits, exposure tracking, and a rule for what happens at the limit |
| Approval flows | A buyer raises an order, someone senior approves it before it is real | Order states before "placed", and a notion of who may approve what |
The middle option nobody mentions
You do not always have to choose. A common shape is a storefront on a platform for the public-facing catalogue, plus a separate ordering portal for account customers, with both talking to the same inventory. It costs less than a full custom build and avoids bending a retail tool into a shape it resists.
What actually drives the cost either way
Not the storefront. The integrations. Whatever you build has to agree with the system that holds your stock and the system that issues your invoices, and keeping three systems agreeing about one number is most of the work. Ask any prospective vendor how they intend to reconcile those, and judge the project on that answer rather than on the design mockups.