
An online store for a small business: what you actually need and what you don't
Before you build an ecommerce site, do the maths: real costs, how many orders you need to break even, and when selling without a store is a better move.
The online store nobody writes about
Almost everything written about ecommerce is aimed at a brand shipping thousands of orders a month. Size filters, personalised recommendations, cart recovery, loyalty points.
But if you run a corner shop, a bakery, a small winery or a craft workshop, that is not your problem. Your problem is simpler and more uncomfortable: is this going to be profitable, or will I spend two years maintaining a website that takes three orders a month?
That is what this article is about. The maths first, the technology second.
What an online store really costs
The classic mistake is looking only at the build quote. A store has three layers of cost and only one of them is creating it.
What you see
The build. Paid once, and the only thing most people compare before deciding.
What you don't
Payment fees on every sale, hosting, the domain, product photography, the time to keep the catalogue current, and the logistics of every order. This is the part you pay forever.
The recurring costs that belong in the spreadsheet before you start:
| Item | What it is |
|---|---|
| Payment processing | A percentage of every sale plus a small fixed fee per transaction. Check your provider's current rates: they change. |
| Hosting and domain | From very little on a static site to serious monthly fees on a paid platform. |
| Platform | If you pick a hosted SaaS platform, that is a fixed monthly fee whether you sell or not. |
| Product photography | The most underestimated cost of all. Without good photos, nothing sells online. |
| Your time | Packing orders, answering questions, handling returns. Not free just because you do not invoice it. |
The calculation that decides
Add up your fixed monthly costs and divide by your margin per order. That number is how many orders a month you need just to stop losing money.
An example with round numbers: €90 a month in fixed costs across platform, hosting and tools, and a €12 margin per order after fees and shipping.
8 orders/month
just to cover fixed costs. Order number nine is the first one that leaves you anything
If seeing that number makes you think "I can do that with the customers I already have", go ahead. If it makes you think "I have no idea whether I would get there", that is the most valuable piece of information you will get today, and it arrived before you spent anything.
When you do NOT need an online store
There are three cases where building ecommerce solves the wrong problem:
Alternatives that work better than a store
You sell few products at high prices. A strong product page, an order form and a payment link you generate yourself sell just as well without maintaining a whole store.
Your sales are local and collected in person. If the customer picks up, what you need is a booking system or a well-organised WhatsApp ordering flow, not a cart and a payment gateway.
Your product sells through conversation. Bespoke furniture, configurable products, quoted services. Here the cart gets in the way: what converts is a form that captures what the customer needs.
Selling online is not the same as having an online store. You can sell with a good catalogue, a phone and a payment link. Plenty of businesses that jumped to ecommerce would have earned more by spending that money on making their current site convert.
What every store does need, however small
If the maths works, these are the non-negotiables. Notice that almost none of them is a flashy feature.
1. Decent product photography
This comes first, ahead of the technology. In a physical shop the customer touches the product; online they only have your photos. Clean background, natural light, several angles and something for scale. A modern phone next to a window is enough if you take it seriously.
2. Product pages that answer questions before they are asked
Exact dimensions, materials, weight, what is included, how long delivery takes, what happens if it does not fit. Every unanswered doubt is an order that does not complete. That is writing, not programming.
3. Payment without friction
Card, local payment methods your customers actually use, and a checkout that fits on one screen. No forced registration before buying: requiring an account in order to pay is one of the most expensive ways to lose sales.
4. Speed
A slow store loses sales in a way you can measure in your own analytics. If the product page takes four seconds to load on a phone, a large share of people never see it. We cover it in why website speed matters.
5. The legal side, which is not optional
Selling online to consumers in the EU means concrete obligations: full seller identification, terms accessible before payment, clear total price including taxes and shipping, a GDPR-compliant privacy policy, and the 14-day right of withdrawal with its corresponding form. This is not optional small print: it is what keeps you out of a fine and out of an argument with a customer.
6. Analytics from day one
You need to know how many people arrive, how many add to cart and how many pay. Without those three numbers you cannot improve anything, only change things and hope.
Platform: the decision that matters least and gets debated most
It is the question everyone asks first and the one that least determines the outcome. Still, the short answer:
| If... | Then |
|---|---|
| You sell fewer than 50 orders a month and want to start now | A hosted SaaS platform. Quick to launch, fixed fee, little control. |
| You want to control cost per sale and already have a site | A self-hosted solution: cheaper long term, more maintenance work. |
| Your catalogue is small and your traffic comes from Google | A custom-built store, light and fast, which also ranks well. |
| You sell services or configurable products | You probably do not need a store, but a good sales page. |
No platform sells for you. I have seen technically flawless stores with zero sales, and bare-minimum catalogues invoicing every month. The difference was never the software.
The mistake every small store makes
Building the store and waiting. A freshly launched online store has no visitors: it is not a unit on a busy street, it is a unit in the middle of a field.
The budget should be split roughly like this: if you have €3,000 for the project, do not spend €3,000 on the store. Spend part of it on the store and keep the rest to bring people in during the first months, whether through content and organic search, your existing customer list or paid advertising. A store with no acquisition plan is an investment that sits still.
Frequently asked questions
How much does an online store cost? It depends heavily on the catalogue and the integrations. The realistic range for a small business runs from a simple store to a custom project with stock and invoicing integration. We break it down in how much a business website costs.
How many orders do I need to break even? Divide your fixed monthly costs by your margin per order. That number, not an industry average, is your answer.
Can I start by selling through Instagram or WhatsApp? Yes, and for many businesses it is the sensible move. Validate that demand exists by actually taking money, then build the store knowing what sells.
Do I need an online store if I sell services? Almost never. What you need is a service page that explains clearly, builds trust and makes contact easy.
What are my legal obligations selling to consumers? Identify yourself as the seller, show terms and the total price before payment, comply with GDPR and honour the 14-day withdrawal period. Have those texts reviewed before you open.
In short
An online store is a tool, not a magic channel. It works when the margin per order covers the fixed costs, when the photos and product pages sell, and when somebody is responsible for bringing people in. It fails when it gets built because it seemed like the thing to do.
If you want us to run that calculation with you before deciding anything, that is exactly where we start on online store projects. If the number does not work, we will propose the alternative that does. Get in touch and we will look at your figures.


