What an online store actually costs to run

Of 27,169 project briefs we hold, 4,342 involve an online store in some form. It is the largest single category we see, ahead of maintenance work, automation and branding. It is also the category where a build quote is least predictive of the eventual cost, because a store, unlike a brochure site, has a cost base that scales with how well it does.

The three recurring lines

Almost every store cost decomposes into a build, then three recurring lines that nobody puts in the original quote because they are not the builder's to charge.

The plan is the only one anybody shops on, and it is the least important of the three. The other two are where stores quietly become expensive.

App creep is the invisible line

A hosted store platform ships with a deliberately incomplete feature set. Reviews, subscriptions, bundles, upsells, advanced shipping rules, loyalty, back-in-stock alerts, a decent search — each of these is an app, each app is somewhere between $10 and $40 a month, and each one is individually too small to argue about.

The pattern is consistent and it is not a platform's fault: the apps get added one at a time, in response to a real need, and none of them ever gets removed when the need passes. A store that has been trading for two years and has never audited its app list is usually paying for at least two things it no longer uses, and often for two apps that overlap.

The percentage that reorders everything

Payment processing is unavoidable and broadly similar across providers. What is not similar is the additional cut some platforms take on top of it if you do not use their own payment product. At low volumes this is invisible. Above roughly $10,000 a month in sales it starts to dominate every other line, and it reorders which platform is cheapest.

Worked crudely: a two per cent additional cut on $10,000 a month is $200 a month, which is more than most platform plans and more than a typical app stack. On $50,000 a month it is $1,000 a month, and at that point the difference between two platforms is no longer a subscription decision — it is a material line in the business's margin.

Where the build money actually goes

Assuming the platform decision is made, store build cost is driven by four things, in descending order of impact:

1. Product count and variant complexity. Fifty simple products is a different job from fifty products with size, colour and material combinations, each with its own stock level and image.
2. Whether product data exists in a usable form. A spreadsheet with clean columns is an import. A folder of PDFs and a supplier catalogue is a data project that happens to end in an import.
3. Shipping and tax rules. Flat rate to one country is trivial. Weight-banded rates across zones with thresholds and exclusions is genuinely intricate, and it is the single most common source of post-launch bugs.
4. Integrations — inventory, accounting, fulfilment, marketplaces. Each one is a scoped piece of work with an owner on your side who has to supply access.

Design, which is what most people assume drives the price, is usually fourth or fifth. A well-implemented theme with a considered content structure outperforms a bespoke design with disorganised products almost every time.

The question worth asking before any of this

Do you need a store, or do you need to take payment? These are not the same requirement. A business selling four things to people who already know them needs a payment link and a decent page. A business selling four hundred things to strangers needs search, filtering, stock control and a catalogue structure. Building the second when you needed the first is the most expensive mistake available in this category, and it is very common, because the second one is what people picture when they say the word shop.