Conversion rate: why your e-commerce ads are not selling

The campaigns are running, the clicks are coming in, the sales are not following. The reflex is to change the ad or raise the budget. Yet an ad that generates clicks has done its job: it has brought an interested person to your store. If that person leaves without buying, it is the page that lost the sale.

Why conversion rate sets your acquisition cost

Conversion rate is the share of visits that end in a purchase. Customer acquisition cost (CAC) is spend divided by the number of new customers. The two are mechanically linked: for the same cost per click, a store that converts twice as well pays half as much for each customer. As long as the page leaks, every franc added to the budget buys visits that leave again.

If your margin per order is CHF 70, the first scenario loses money on every customer and the second makes money, without touching the campaigns. To know where to set that ceiling, see how to calculate an online store's maximum CAC.

Reading the funnel stage by stage

An overall conversion rate does not tell you where the leak is. You need to split it into four stages and look at the passage from one to the next, ideally for paid traffic alone and separately on mobile and desktop.

1. Product page viewed. The starting point: how many visits actually reach a product page.
2. Add to cart. The share of product page visitors who add an item.
3. Checkout started. The share of carts that move on to checkout.
4. Purchase. The share of started checkouts that end in a paid order.

In GA4, these stages match the recommended e-commerce events (view_item, add_to_cart, begin_checkout, purchase). A funnel exploration chains them together and can be filtered by source or device. In Shopify, the analytics reports show the store's conversion rate broken down into sessions with an add to cart, sessions that reached checkout and sessions that converted. The two tools do not count in exactly the same way: compare each tool's trends with itself, not one tool's figures with the other's. If the events are poorly installed, the diagnosis is wrong from the start: reliable conversion tracking comes before everything else.

What a drop at each stage means

Typical conversion leaks in Switzerland

• Shipping costs and delivery times revealed at checkout. The visitor accepts a price, then sees the total rise at the last step. Show shipping, the free-delivery threshold and the delivery time from the product page onwards.
• No TWINT, no invoice. Many Swiss shoppers are used to paying with TWINT, the Swiss mobile payment app, or by invoice. If only card payment is offered, some of them stop at the moment of paying.
• Unclear prices. Show prices in CHF, VAT included, with no approximate conversion from euros. A price that changes between the product page and the cart breaks trust.
• Customs for stores shipping from abroad. If the Swiss customer risks paying VAT and customs clearance fees on delivery, they hesitate. Say clearly who pays what, or include these costs in the price.
• Site not translated into German. A campaign running in German-speaking Switzerland that lands on a French page loses a large part of its audience before it even reaches the product page. The translation must also cover checkout, emails and terms.
• Not enough trust. Visible customer reviews, a simple returns policy, an identifiable Swiss company (company name, UID number, contact): an unknown business selling online has to prove it exists.
• Mobile speed. Traffic from social media arrives mostly on phones: check the mobile share in your own statistics. A slow page or heavy images drive the visitor away before they see the product.
• Ad ↔ landing page inconsistency. The ad shows a product, a promotion or a price; the landing page must show exactly the same thing, not the home page.

The method: one leak at a time

Fixing ten things at once may give a better result, but without knowing which change mattered, or whether one of them did harm. The method looks slower and is faster in practice.

1. Measure. Record the rate at each stage over a stable period, for paid traffic.
2. Pick the biggest drop. That is where a gain brings in the most orders.
3. State a hypothesis. For example: “visitors leave the cart because they discover the shipping cost”.
4. Change one thing only. Show shipping on the product page, add TWINT, translate the checkout.
5. Measure again. Over a comparable period, with a stable ad budget, then move on to the next leak.

With little traffic, a statistically sound A/B test takes time. In that case, a before/after comparison over comparable periods is still useful, provided nothing else changes in parallel, neither the campaigns nor the prices.

When to raise the ad budget

Once the main leaks are fixed, every franc of advertising brings in more orders. That is the time to raise the budget, in steps, while watching the cost per customer acquired. If the store itself needs a rebuild rather than tweaks, a Shopify store designed for paid traffic fixes several leaks at once: Swiss payment methods, bilingual content, mobile speed and dedicated landing pages.