What a rented platform costs by year three

Every comparison between a rented platform and a built one is conducted on the wrong number. The subscription is quoted, custom development is quoted, the first is smaller, and the conversation ends. Nobody adds the other four lines, and the other four lines are where the money is.

Take the published ladder first, because at least it is public. At the time of writing Shopify runs at thirty-nine dollars a month on Basic, a hundred and five on Grow, three hundred and ninety-nine on Advanced, and Plus starts at two thousand three hundred a month on a three-year commitment or two thousand five hundred on one year. Very large merchants move off the flat fee onto a variable one somewhere around eight hundred thousand dollars of monthly volume, in the region of a third to four tenths of a per cent of sales with a cap in the tens of thousands. Note the currency while you are reading those, because a Greek merchant carries the exchange rate on every one of them.

Now the line nobody quotes: the apps. A serious store does not run on the platform, it runs on the platform plus fifteen to thirty subscriptions, and in Greece the count is at the top of that range because the entire local layer arrives that way — myDATA transmission, invoicing, instalments, cash on delivery, four courier integrations, a locker map, a Skroutz feed, a marketplace connector. Individually they are ten to three hundred dollars a month and each one feels trivial when it is approved. Collectively they routinely exceed the platform fee itself, they are billed per store rather than per company, and not one of them is negotiable at renewal.

The third line is the one that surprises people who have not run this project before: the theme is not the store. A brand that cares about how it looks does not launch on a purchased theme, so there is still a design phase, still a front-end build, still an agency invoice. The rented platform did not remove that cost. It removed your control over the part underneath it, and left the visible part exactly as expensive as it always was.

Fourth, the surcharge, which compounds silently because it is a percentage. If you take instalments in Greece you are on a Greek gateway, and a third-party gateway carries a platform fee per order on top of the acquirer's own. On Advanced that is six tenths of a per cent. On a store turning over two million euros a year, six tenths of a per cent is twelve thousand euros annually, paid to the platform for the act of accepting a payment it did not process, on top of a subscription you already pay. Move to Plus to cut the surcharge and you have just committed to roughly twenty-eight thousand dollars a year in platform fees to save twelve thousand euros in surcharge, which is the kind of arithmetic that only makes sense in one direction and it is worth checking which.

Add the four lines over three years for a mid-sized Greek store and the number lands, comfortably and repeatedly, in six figures. That is not a rhetorical figure — it is a subscription, plus twenty app subscriptions, plus a build you paid for anyway, plus a percentage of every order, multiplied by thirty-six months. And at the end of those thirty-six months you own the design and nothing else. The engine is not yours, the data model is not yours, the checkout is not yours, and the ceiling on what the business can ask for next is exactly where it was on the first day.

This is the point at which the honest comparison becomes possible, and it is not the one our industry usually makes. Custom is not better because it is custom. Custom software written once for one merchant has a well-documented ending: it is correct on launch day and then decays, because a payment provider changes an API, a browser drops a behaviour and a security advisory lands on a Friday, and four years later somebody proposes a rebuild and the merchant pays for the same thing twice. Anyone selling you a bespoke platform without telling you that is leaving out the expensive half.

So the answer we arrived at splits the store rather than choosing a side. The bespoke half — the design, the storefront, the content model, the integrations into your ERP, your warehouse, your acquirer, your couriers — is drawn for you, belongs to you outright, and leaves with you. The product half — the commerce engine, the CMS, the dashboard, the mobile app, and the Greek layer inside all of them — is licensed and maintained by us, patched continuously, and improved for every store on it at once. That split is Nextcommerce, and the terms are written down in plain language on that site rather than described in a meeting: what is yours, what is licensed, and how you leave with one month's notice and a complete export.

And the limit, stated plainly because it decides the reader's actual answer. Below roughly a quarter of a million euros a year, with a small catalogue, no ERP, one courier and no marketplace, the rented platform wins the arithmetic outright and we will say so. The crossover is not a revenue number by itself — it is the moment the store stops being a website that takes orders and becomes an operation that a platform is either helping or throttling. Most people know precisely when that happened. It is usually about eight months before they start asking whether the platform is the problem.