Benchmarking: What It Is, Types, and How to Do It
Complete guide to benchmarking: what it is, the 4 main types, a step-by-step process, and how to compare your conversion rates with your industry...

Since June 28, 2025, selling online to consumers in the European Union requires meeting accessibility standards. The European Accessibility Act (EAA) has turned what was long considered a best practice into a legal obligation. For marketing and product teams, it raises a very concrete question: what happens to conversion if we change colours, buttons, and copy to comply?
The EAA is Directive (EU) 2019/882, which the European Commission summarises here. In Spain it was transposed through Law 11/2023 of May 8. It applies, among others, to ecommerce, consumer banking services, passenger transport services (websites, apps, and electronic tickets), and ebooks. Micro-enterprises providing services, with fewer than 10 employees and an annual turnover or balance sheet of up to 2 million euros, are exempt.
The usual technical reference is EN 301 549, which aligns with the AA level of the WCAG 2.1 accessibility guidelines. This article is not legal advice. If you have questions about your specific situation, consult a specialist.
The annual WebAIM Million analysis reviews the one million most visited homepages. Its 2025 edition found WCAG failures on 94.8% of them, and the most frequent issue was low-contrast text, present on 79.1%.
Level AA requires a minimum contrast ratio of 4.5:1 for normal text and 3:1 for large text and UI components. Many vivid brand colours — oranges, yellows, light greens — fall short of that ratio when paired with white text. And those tend to be exactly the colours used for purchase buttons.
WCAG has dozens of criteria, but not all of them weigh equally on the pages where the sale is decided. From a CRO perspective, these are the ones worth reviewing first.
It is a reasonable fear. The CTA colour is often a brand decision that has been defended for years. At Dogfy Diet, which operates in Spain, France, Italy, and Germany, we decided to settle it with data.
In a test applied across the entire site on mobile and desktop, we replaced the vivid orange of the CTAs with a darker tone offering higher contrast. The hypothesis was framed in an unusual way: we were not trying to improve, but to verify that the change did not hurt performance.
| Version | Visitors | Form entries | Form entry rate |
|---|---|---|---|
| Control (vivid orange) | 5,671 | 1,929 | 34.02% |
| Variant (darker tone, higher contrast) | 5,493 | 1,894 | 34.48% |
The expected uplift was +1.37%, with a 53.34% probability in favour of the variant. In practice, that means there was no detectable difference — which was exactly what we needed to prove. The colour change required by the regulation did not penalise funnel entry.
If the change is mandatory, why test it? For three reasons. First, there is almost always more than one way to comply, and the test tells you which one best preserves conversion. Second, if a particular implementation hurts performance, you know before rolling it out to all traffic. Third, it is documented for other markets and teams.
The key is in the test design. Before launching, you need to set the maximum acceptable loss rather than look for a winner. That is the logic of a non-inferiority test, and it completely changes how you read a "no winner" result.
According to the World Health Organisation, around 1.3 billion people — roughly 16% of the global population — live with a significant disability. On top of that come situational limitations that affect everyone: a screen in direct sunlight, a phone used one-handed, tired eyes... Improving contrast, form labels, or keyboard navigation benefits all of those users.
We are not going to promise that accessibility will skyrocket conversion: our data does not show that. What it does show is that compliance does not have to cost you sales if you approach it with the right method.
Start with the pages where the sale is decided: product detail, cart, checkout, and forms. On those pages, check the contrast of buttons and text, make sure fields have visible labels, that errors are explained in text, that focus is visible when navigating by keyboard, and that relevant images have alt text. Then prioritise changes by funnel impact and validate the ones that affect conversion elements with a test.
A Scan&Boost audit is a good first diagnostic of the visual and mobile experience. If you already know what you want to change, working with our full CRO method lets you validate it with tests in a matter of weeks.
Complete guide to benchmarking: what it is, the 4 main types, a step-by-step process, and how to compare your conversion rates with your industry...
The Facebook Ad Library lets you view all active ads from any advertiser. Learn how to use it to analyze competitors and improve your campaigns.
A CRO audit reveals exactly why your website is losing conversions. Follow this step-by-step checklist to find and fix the leaks in your funnel.