A payment provider’s conversion study describes its own users, not the payment step on your site

PayPal has commissioned Nielsen more than once to measure what accepting PayPal does for a merchant, and the editions PayPal has published report large gains. The UK edition was published by PayPal in April 2024. Nielsen looked at 182 large UK enterprises and 4,037 UK shoppers, combining a behavioural panel of 142,000 desktop purchase transactions from 5,000 consumers between July 2022 and June 2023 with a survey of recent buyers in September 2023. The findings are therefore three to four years old, and they come from one market with its own payment habits.

Studies like this are useful, but not in the way their headlines suggest. A figure such as a conversion lift tells you how one group of shoppers behaved under one definition. Whether the same lift would appear in your checkout depends on three things the headline leaves out: where the measurement starts, who is being compared and which device the purchase was made on.

Where the count starts decides the size of the number

Conversion is a ratio, and the starting point of the ratio changes everything. PayPal’s own studies show this across editions. A 2018 comScore study of PayPal Checkout measured checkout conversion from the moment a user clicked on a payment type. The US Nielsen study published in January 2021 measured both from the moment a shopper entered the merchant’s site and from the start of checkout, in each case through to the confirmed purchase page. The 2024 UK study measures its sector figures, which come from the behavioural panel, from the point at which the customer starts to pay.

Each starting point filters out a different group of shoppers before the count begins. A shopper who has already clicked on a payment method has made most of the decision, so a rate measured from there will look very different from one measured from the first page view. The same 2024 article also gives two headline figures: an average conversion improvement of 35% and an average lift of 10% over alternative checkout methods. The text presents both as results among surveyed enterprises, but the footnotes trace them to the survey of 4,037 shoppers, and neither states where the count starts. Before either number is used in a business case, it matters which comparison stands behind it.

The behavioural panel adds a second limit. It covers desktop transactions only. The payment step on a phone, where the shopper switches between your site, a wallet and a banking app, is not part of that measurement at all.

Users of a method are not a random group of shoppers

The same article reports that PayPal users buy 34% more often from large enterprises than other digital buyers. That compares two groups of people, not two versions of a checkout. Shoppers who have set up a wallet and use it regularly may well be the shoppers who buy online most often in the first place. A comparison between users and non-users cannot separate the effect of the method from the habits of the people who choose it. Only a test on your own traffic, in which the same kind of shopper meets two versions of your payment step, can do that.

This is not a reason to dismiss the research. It is a reason to read it as a description of PayPal users in large UK webshops between 2022 and 2023, which is what it is.

The reasons shoppers give are the part that travels

The most usable findings are not the multipliers but the motives. Asked why they chose PayPal, 61% of the surveyed users named trust in the brand, 54% checkout speed and 50% a simple payment experience. Those answers describe what a shopper wants from the payment step itself, whatever method sits behind it: a name they recognise, few taps, no card number to type and a sense that their details are safe.

That shifts the question from whether a particular wallet belongs in your checkout to whether each method you offer delivers those four things. In the Netherlands the method most shoppers reach for is a different one: the Dutch Payments Association puts iDEAL at 71% of online purchases. A UK finding about wallet users says little about what belongs at the top of a Dutch payment step. It says more about the shopper who does not use the default method, such as a buyer from abroad or someone paying with a card, and about the reassurance that shopper looks for when the familiar option is missing.

Trust at the payment step is also built outside the payment method. A shop name the shopper recognises on the payment screen, the amount and the order shown while they confirm, and a clear way back to the basket if they change their mind all work on the same motive. A logo does not do that work on its own.

Measure the method on your own orders before you move it

Your order data can answer the question the study cannot. For each payment method, take the share of shoppers who select it and then reach a confirmed order, and split that by device and by first and repeat orders. A method with a low completion rate after selection is losing shoppers inside its own flow, whatever any vendor study says about it. A method that completes well but is chosen rarely may simply be placed where nobody looks.

Then test changes on your own traffic rather than adopting someone else’s multiplier. Move a method up or down the list, change its label, or add a line about what happens after the shopper selects it, and compare completed orders per payment attempt over a full sales cycle. Start this month with the completion table. It takes an afternoon in your analytics, and it tells you which methods in your payment step deserve attention first.

Sources: paypal.com, Nielsen UK study 2024, newsroom.paypal-corp.com, Nielsen US study 2021 and betaalvereniging.nl

Related reading