How to set up a Shopify store properly, and fast

Most stores that take three months to launch were not slow. They were done in the wrong order, so every step kept reopening the one before it. Speed here is almost entirely a sequencing problem, and the parts that decide whether you ever take an order are the dull ones nobody writes about.

What should you decide before you open Shopify?

The single biggest cause of a stalled build is starting in the theme editor. You pick a template, then discover your products do not fit the grid it assumes, then change the template, then rewrite the descriptions to match, and a fortnight has gone on decisions you had to make anyway — only now you have made them three times.

Before you touch the platform, settle four things on paper: the full list of what you sell, one photograph per item that you are actually willing to publish, the price, and how each one ships. That is a spreadsheet and an afternoon. Every later decision follows from it, and none of them can be made properly until it exists.

This is also where most people discover their real constraint. If you have forty products and photographs for nine, your problem is not Shopify. Launch with the nine.

Why is the theme not the real job?

A theme decides how your shop looks. It does not decide whether the checkout works, whether tax is right, whether a customer in Aberdeen can be quoted postage, or whether you find out that someone abandoned a basket. Those are configuration, and configuration is the actual work.

Pick a well-supported theme, including Shopify's own free ones, and change as little as possible. Every customisation is something that can break when the theme updates, and something you have to understand again in six months when you want to change a price.

The exception worth paying for is a catalogue that genuinely does not fit a standard grid — variants with their own images, bundles, or a product that needs explaining before anyone buys it. That is a real reason. Wanting a different shade of green is not.

Why test payments with a real order, not a test order?

Shopify has a test mode, and it will tell you the checkout works. It cannot tell you that your bank details are right, that the payout actually lands, or that the confirmation email leaves and arrives looking like something a person would trust.

So before launch, buy something from your own shop with your own card, at full price, and let it settle. Then refund it. You now know four separate things work: the charge, the confirmation, the payout, and the refund path you will eventually need on a bad day.

It costs you the card fee on one transaction. Skipping it is how shops discover on their first real sale that payouts were pointing at a closed account.

Why do most builds stall at shipping?

Shipping is the step people leave until last, and it is the one most likely to send you back to the beginning. Rates depend on weight and size, which depend on packaging, which depends on the products — so if you guessed at postage early, you will be redoing it.

Weigh a packed parcel of each thing you sell. Not the product: the parcel, with the box and the padding. Then set rates from real weights and pick your bands deliberately. Free delivery over a threshold is a pricing decision, not a shipping one, and it belongs in the same conversation as your margin.

Decide what you will not ship, too. Overseas orders, or anything to an address you cannot get a tracked service to, are better refused at checkout than discovered after you have taken the money.

Which pages is a UK shop expected to have?

A UK store selling to consumers needs terms of sale, a privacy notice, a returns and cancellation policy that reflects your actual practice, and clear contact details. Shopify generates templates for some of these, and the templates are a starting point rather than a finished document — they contain placeholders that stay in the published page if nobody reads them.

The returns policy is the one worth writing yourself. Consumers buying online in the UK have cancellation rights that exist whether or not your page mentions them, and a policy that contradicts the law does not override it — it just tells a customer you either do not know the rules or are hoping they do not.

Write what you will really do. A fourteen-day policy you honour beats a thirty-day policy you argue about.

When should you turn tracking on?

Analytics installed after launch cannot tell you about the traffic that came before it. That sounds obvious, and it is still the most common gap, because tracking feels like a growth job rather than a build job.

The minimum is a working analytics property and, if you intend to advertise at all, the ad platform's tag and its server-side equivalent. Get them in while you are already in the settings, and check they report a real event — a tag that is installed but firing nothing looks identical to a tag that is working until the day you need the data.

Then leave them alone. Collecting from day one costs nothing and gives you a baseline. Deciding in month three that you want to know what happened in month one is not a thing you can do retrospectively.

What does 'fast' actually look like?

With the products decided and the photographs in hand, a straightforward catalogue is a couple of days of unglamorous configuration. Without them it is unbounded, which is why so many builds have no end date.

If you are doing it yourself, the honest advice is to block two consecutive days rather than spreading it over six evenings. The work is full of small dependencies, and every gap between sessions is spent remembering where you were.

If you would rather not, this is what OpenX24 does: a Starter Store Build is £349, one payment, handed over in 2 business days, with the products loaded, payments and shipping tested with a real order, legal pages written and tracking live from day one. You own the store, the domain and the data at the end of it, and a Run plan afterwards is optional.

What to do next

Do it in this order: decide the catalogue on paper, pick a plain theme, configure payments and test with a real purchase, set shipping from weighed parcels, write the legal pages properly, and switch tracking on before anyone arrives. The sequence matters more than the speed, because doing it in a different order is what turns two days of work into three months of nearly finished.

Common questions