The question reaches us the same way almost every time: the product exists, somebody said a shop was needed, and now a supplier and a price are being looked for.
The price comes back quickly and it is usually an honest one. It just answers the wrong question. An online shop is not a website with a basket bolted on. It is an operation, and the website is the cheapest part of it - the only part the quote covers.
The question is not whether a shop can be built. It can. The question is what you will be doing every day once it exists.
Three honest options
A real shop. A basket, stock levels, payments, parcel machines, returns. It suits you when there are more than a handful of products and orders come in often enough that taking them by hand starts eating into other work.
A product page and a form that a person answers. The product is properly written up, the price or a price range is on the page, and underneath it there is an order form. It suits you when orders are few and each one is large, or when every order needs one clarifying question anyway.
Somebody else's marketplace. The buyers are already there and you build nothing. In return you pay a percentage, you play by their rules, and the customer stays their customer. A good place to test whether anybody buys your product at all. A bad place to be the only place five years from now.
Most small companies start with the third, get to the second, and reach the first when the number of orders demands it. Starting at the other end is the most expensive way to learn the same thing.
The costs that arrive after launch
The quote covers the build. Everything else starts on launch day and does not stop:
- The payment provider's fee on every order, every month, for as long as the shop runs.
- Delivery. Parcel machine contracts, waybills, a price list that is not yours, and a customer who wants to choose between three couriers.
- Returns. In distance selling the buyer has a 14 day right of withdrawal. That means money back, the product back on the shelf, and somebody there to receive it.
- Packaging and packing. The box, the tape, the label, and the time one parcel takes.
- Stock that has to be true in two places at once. If the same product also sits on a shop shelf, one day it gets sold twice.
- Accounting. Every order is a document, and there are exactly as many of them as there are orders.
- The person who answers at nine on a Sunday evening. The shop is open even when you are not.
None of these is large on its own. Together they are the work an online shop actually is, and not one of them gets easier because the website is prettier.
The question that decides almost everything
Before picking an option, answer two things: how many orders a week do you expect, and how large is one average order.
The first decides whether doing it by hand is still tolerable. Automation only pays off once the manual work costs more than the tool that replaces it. You know that line better than any agency does: it is the moment when taking orders starts eating into other work.
The second decides whether the buyer even wants to check out on their own. On a cheap repeat purchase they do. On an expensive and rare one most people want to talk first, and then the basket is in their way rather than helping them.
Four products and two orders a week is not a sales channel
Here we argue against our own interest.
At four products and two orders a week a shop is not a sales channel but a job you bought for yourself: software to update, stock levels to correct, and a shop that is open every day whether or not you have something else to do.
The most common correct advice is a boring one. Sell the first fifty orders by hand, by email and invoice. Build after that.
Inside those fifty you learn the things no quote can guess for you: which questions keep coming back, which product actually moves, what packing costs, how many returns arrive, and where your own patience runs out. Every one of those answers is later a line in the shop's specification. Without them the specification is guesswork you are paying for.
The underrated middle option
A proper product page with a form beats a shop for a low volume of high-value orders, for three reasons.
It is finished in weeks rather than months. It needs no new monthly fee beyond ordinary hosting. And it is not wasted work: if the volume ever grows, this is the same product page the shop needs anyway. The basket goes around it, the words stay.
The only thing missing is automatic payment. At a low order count sending an invoice is faster than configuring a payment provider.
What to ask before you sign
A cheap quote answers "a shop, price X". A real quote answers these six questions, and its price is higher because of them. That is the point.
- Which payment provider, and whose name is the contract in? You pay the fee, not the builder.
- Which couriers are included in the price, and what does a fourth one cost?
- Where do stock levels come from, and what happens when the same product is also sold over the counter that day?
- How does an order reach the accounts? If the answer is "you export a file", that is your monthly job.
- Who updates the shop software, and what does that cost per month? Leaving a shop unmaintained is not the same as leaving an ordinary website unmaintained, because payments and customer data leak out of this one.
- What happens when a payment is left half finished? Who sees it, and who calls the customer?
If the supplier answers all six without pausing, they have built shops before. If the answer is "that is all included in the package", ask the same thing a second time.
When the answer is "not yet"
That is not bad news. A product page that genuinely answers every question a buyer has, with a form underneath it, is an ordinary website whose price you already know. And it starts selling this week rather than in the autumn.
If you cannot tell which side of the line you are on, ask us. We will also say so when the answer is that one good product page is enough for now.
Frequently asked questions
There are three honest options. A real shop suits you when there are more than a handful of products and orders arrive often enough that handling them by hand eats into other work. A product page with an order form that a person answers suits you when orders are few and each one is large. Somebody else's marketplace is good for testing whether anybody buys your product at all, but the customer stays their customer. For a low volume of high-value orders the product page beats the shop: it is finished in weeks, it needs no new monthly fee, and it is the same page a shop would need later anyway.
The quote covers the build. Everything else starts on launch day: the payment provider's fee on every order, courier and parcel machine contracts, returns (in distance selling the buyer has a 14 day right of withdrawal), packaging and the time packing takes, stock that has to be true in two places at once, accounting for every order, and a person who answers on a Sunday evening too. None of these is large on its own, and none of them gets easier because the website is prettier. Together they are the work an online shop actually is.
Nobody can name the exact number from the outside, but the rule is simple: automation only pays off once the manual work costs more than the tool that replaces it. You know that line better than any agency does, because it is the moment when taking orders starts eating into other work. Below it a shop is extra work rather than a sales channel: four products and two orders a week means software to update and a shop that is open every day whether or not you have something else to do.
Six questions that separate a real quote from a cheap one. Which payment provider, and whose name the contract is in, since you pay the fee. Which couriers are included and what a fourth one costs. Where stock levels come from and what happens when the same product is also sold over the counter that day. How an order reaches the accounts. Who updates the shop software and what that costs per month. And what happens when a payment is left half finished - who sees it, and who calls the customer. If the answer is "that is all included in the package", ask the same thing a second time.
Read next
It is twenty to midnight and you are on the second page of the quote, the one that says what the number buys. Scope, prototype, staging, acceptance, change request. Here are those words in a sentence each, along with the ones a designer says beside a draft and the ones that arrive in the yearly domain letter. And the single word that costs the most.