Find out why people leave your store, then fix it

What conversion rate optimisation actually is

Your store gets a certain number of visits and a certain share of them buy. Conversion rate optimisation, or CRO, is the work of raising that share: same traffic, more orders. It is not changing button colours and hoping. It is finding the specific reasons people leave without buying, removing them one by one, and checking that each change did what you expected.

It is also the cheapest growth most stores have, because every improvement applies to every visitor you already pay for, from search, ads and email, for as long as the change stays in place.

Research first, always

Anyone can guess what is wrong with a store. The point of CRO is to stop guessing. We build the picture from four places.

Your analytics

Shopify's reports and Google Analytics 4 show where people arrive, which pages they leave from, how mobile compares with desktop, how far through the checkout they get, and which products get looked at but not bought. The funnel from product page to basket to checkout to payment tells us where the leak is before we know why.

Session recordings and heatmaps

Watching real visitors use the store on a phone is humbling. People miss the size guide, fight the variant picker, tap things that are not buttons, and scroll straight past the add-to-basket button because an app pop-up covered it. The patterns repeat quickly.

What customers ask

Your customer service inbox is a list of the questions the product page failed to answer. Delivery times, returns, whether it fits, whether it comes assembled. Every one of them is a reason someone else left without asking. We read the inbox, the reviews and the chat logs, and add a short survey where it helps.

The checkout itself

Shopify's checkout is excellent and mostly locked down, which is a good thing. But what feeds into it is yours to fix: delivery costs that appear late, discount fields that send people off hunting for codes, express payment buttons missing, and addresses typed by hand.

Hypotheses, not hunches

Each finding becomes a written hypothesis: because we saw this, we believe changing that will improve this number. Then we rank them by strength of evidence, how many visitors are affected and how much work each takes. The list you get is short and ordered, and it usually contains several things so clearly broken that they need no test, just fixing.

Testing, and the honest truth about sample sizes

A/B testing means showing two versions of a page to different visitors and measuring which sells more. It is the gold standard, and it only works with enough orders. As a rule of thumb a test needs a few hundred orders per version before you can trust the result, so a store taking a handful of orders a day would wait months for a single answer, with seasonality and promotions muddying it the whole time.

So we are straight about which approach fits your traffic:

• High-traffic stores: proper A/B tests on the biggest hypotheses, one at a time, run to completion.
• Most growing stores: fix the clearly broken things without a test, then make bigger changes one at a time and measure before and after, with the caveats written down.
• Every store: judge by orders and revenue per visitor, not by clicks on a button.

An agency that promises a testing programme to a store with a few orders a day is selling you noise. We would rather tell you that up front.

Product pages, where most of the gains are

The product page is where the decision is made, and on a phone the first screen does most of the work. Price, delivery date and cost, stock, the variants that matter and a visible add-to-basket button, all before anyone scrolls. Then the questions in the order customers ask them: what is it exactly, will it fit, what is it like in real life, what happens if it is wrong. Photography that shows scale and use, specifications in text, reviews easy to find, and no pop-up covering any of it.

Basket and checkout

Shop Pay, Apple Pay and Google Pay on the product page and in the basket, because a saved address and card is the difference between buying now and buying later. Delivery costs and dates shown before the checkout, not discovered inside it. A discount field that does not send everyone to a voucher site. Address lookup. And trust signals at the right moment, usually smaller and quieter than people expect. Where the checkout needs more than the settings allow, our checkout and payments page explains what is possible on each Shopify plan.

What you get

A CRO audit as a fixed-fee written report with the evidence and the prioritised list. Then, because we build Shopify themes, the changes are made by us rather than handed to someone else, tested on real phones, and measured. Ongoing CRO runs monthly with a short report: what changed, what it did, what is next. Fixed quote before we start.

Who this is for

Shopify retailers with steady traffic and a feeling that more of it should be buying. Especially stores selling considered purchases, where the product page has to do the job a salesperson would do on the shop floor. If you are about to spend on ads, do this first; it makes every pound you spend afterwards work harder.