Conversion rate optimisation (CRO)
What is conversion rate optimisation (CRO)?
Conversion rate optimisation is the systematic work of getting more of the visitors you already have to complete the action that matters — an order, an enquiry, a registration. It starts with data about where people stop, then tests one change at a time and measures it. It is cheaper than buying more traffic, because it stops the leak instead of refilling the bucket.
How it works
CRO is a loop, repeated for as long as the site runs:
- Find where people stop. The step in the funnel with the largest drop: the product page, the cart, the delivery options, the payment.
- Find out why. Session recordings, heatmaps, the questions customers ask the shop by phone, the reasons given for abandoned orders.
- Change one thing. The delivery price shown earlier, a shorter form, a clearer button, an answer to the question that holds people back.
- Measure it. As an A/B test where traffic allows, otherwise by comparing comparable periods.
Example
An online shop for coffee and coffee machines has 40,000 visits a month and a conversion rate of 1.1%. The funnel shows the largest drop at checkout: 62% of people who reach it leave. The recordings show why. The delivery price appears only on the last step, and registration is required before paying.
The shop shows the delivery price on the product page, adds a free-delivery threshold of €50 and lets people order as guests. The share leaving checkout falls to 44%, and the conversion rate rises to 1.6%. At an average order of €42, that is 200 more orders and €8,400 more in sales a month, from the same visitors and the same advertising budget.
Why it matters for a business
More traffic costs more every month; a better conversion rate keeps paying without further spend, and it makes every future campaign cheaper per order. In most shops the largest gains are not in design but in the basics: delivery cost and time shown early, payment options people trust, a short checkout, a site that works on a phone.
CRO needs enough data to tell an effect from chance. With a few hundred visits a month, tests take too long, and the better approach is to fix the obvious obstacles first and watch what customers do and say.
What to ask
- Where in our funnel do we lose the most people, and do we know why?
- When does a customer first see the delivery price and time?
- Can someone buy without registering, and pay the way they prefer?
- Do we measure each change, or change several things at once and guess?
In an online shop, conversion work often brings more than a new advertising budget. We start with it in our online store development, and on a working shop with the pages that already have traffic.