A guide to accessibility

Slava Shestopalov, Design Manager and Eugene Shykiriavyi, Lead Experience Designer, Eleks, share their guide to web accessibility, detailing the importance of using real people to test and why this is an emerging trend.

What is accessibility on the web?

Web accessibility refers to the accessibility of a site, mobile app, smart assistant, or wearable device for all people, regardless of their computer proficiency, usage preferences, and body features (both physical and mental). It’s an important part of inclusion and means that everyone, including people with a disability — the most vulnerable audience — can navigate technology.

Despite its obvious popularity in theory, in practice, accessibility ratings across many businesses are poor. For example, 70% of websites in the UK are inaccessible, which, although better than globally (over 97%), suggests a huge area for improvement. At the same time, almost 13 million people in the UK have some permanent disability (a medical condition like blindness or hearing loss), not to mention millions of people with temporary and situational disabilities (like a broken arm or temporary blurred vision after the ophthalmologist’s examination).

There is room for improvement in understanding and implementing accessibility which has strong links with usability and a company’s bottom line. Testing digital products with people who have a form of disability is now seen as an emerging best practice in 2021.

Who really needs web accessibility?

Web accessibility benefits both able-bodied and disabled users. However, there are many ways to present the same information to different audiences in the digital world and the main goal of accessibility boils down to ensuring that there will be at least one method that will suit the user.

Able-bodied users have the whole spectrum of ways to interact with software. In contrast, people with disabilities (whether permanent, temporary or situational) will prefer one or a few features or methods depending on their ability. In some situations, these features are also beneficial to all users.

Imagine you open an article on your smartphone. One can read it or listen to an audio version if it’s available or turn on an embedded text-to-speech feature to avoid eye fatigue while commuting. These are all features that have been added to enhance accessibility and will benefit able-bodied and disabled users.

What are the advantages of website testing using people with disabilities?

It is normal practice to challenge software solutions before their public launch. A designer will test the prototype of a new app or site with several unprejudiced people to check whether they find the design clear, useful, and comfortable. This method is called usability testing or sometimes user testing and allows you to identify major issues before launching.

Usually, usability testing participants don’t have disabilities and are recruited according to the target audience’s demographic characteristics. Simply put, selection criteria include job title, country of residence, experience in some topic or with certain software, gender, age, etc.

Accessibility testing is testing a live product with users with a range of disabilities and utilising certain assistive technology. Participants are selected due to the primary senses involved in using your product (for example, vision and touch — for a mobile weather app; eyesight and hearing — for a computer game). Apart from this method’s distinct accessibility focus, it also reveals broader usability issues. For example, if a colour-blind user cannot notice a call-to-action button, there is a chance colour-sighted users might not notice it either.

Of course, one can use an automated web tool to check accessibility (many of them are free of charge), but involving users with disabilities gives you a more realistic insight into common scenarios. Accessibility standards and automated tools aren’t perfect, and formal compliance might not account for behavioural patterns or inexperienced users. Human nature is unpredictable, and people do not always take the “designed” path within a software product, but the one that seems more intuitive — testing will identify that.

How do companies go about practising accessibility testing?

There are many types of disabilities (visual, physical, cognitive, auditory, etc.), and people use different types of assistive technology. However, this article will focus on the case applicable to most websites and web or mobile applications — when your interface mainly relies on vision.

Step 1. Who to test with?

Although there are never-ending debates in the professional community about optimal sample size, your very first round of testing can include sessions with the following participants:

  • 1-2 people with blindness who browse the internet using screen readers;
  • 2-3 people with low vision who use zoom and the increased contrast mode while browsing the web;
  • Optionally, also 1-2 people with colour blindness if you have rich data visualisation: charts, graphs, dashboards, maps, etc.

Step 2. Where do I find users with disabilities?

It is not as hard to find people with disabilities for testing as it may seem at first glance. Feel free to look for participants via:

  • Specialised platforms (for example, Access Works or UserTesting).
  • Communities in social media (try the keywords “people with disabilities,” “support group,” “work opportunities for people with disabilities,” etc.).
  • Organisations for people with disabilities and social enterprises (for instance, the network “Social Enterprise UK”).

An organisation or individual tester might name their standard fee, or you should propose fair compensation for their time and effort.

Step 3. How to prepare for the session?

In this COVID-pandemic time, many people will prefer remote testing to face-to-face testing in the office or lab. Testing remotely brings many benefits:

  • Native environment Participants test with their usual equipment and configured assistive technology. As a rule, people who live with permanent disabilities don’t utilise standard embedded screen readers and prefer professional tools with finely tuned settings.
  • Cost and time efficiency Usually, you would reimburse expenses for travelling to your office. It might be costly if people with disabilities require an accompanying person or accessible transport.
  • Broader recruitment It is more likely you’ll find a participant that meets your criteria around the world, so neither you nor the testers have to travel every time you need to test.

Step 4. What tools should we use?

There are many tools for remote usability testing (UserTesting, Lookback, etc.) that allow you to talk to participants while they share their screen and video. They also help to take notes and create highlights in session recordings. Usually, these tools are pretty costly and require additional time to install and set up.

Fortunately, you can use almost any video conferencing tool: Skype, Zoom, Google Meet, BlueJeans, MS Teams, etc. They all have video calls, screen-sharing, and recording functionality. The main choice criterion is whether your participants are familiar and comfortable with the tool.

⚠️ Tip: Unlike usability testing sessions, accessibility testing usually requires a longer intro part. For example, visually impaired participants need time to configure videos and share their screens.

Step 5. How to formulate testing tasks?

Unlike usability testing, accessibility testing implies you have working software — not just mockups or design prototypes consisting of interlinked pictures. As a result, you are not constrained by limited functionality and can formulate realistic tasks. The best task is a broad real-life task that matches a natural way of thinking and doesn’t include interface-related hints.

✅ Broad realistic tasks:

  • “You want to send a £100 gift card to your friend Owen Chan, 43 Boughton Rd, Wickersley S66 3PZ. Please do that on this site and feel free to think aloud and comment on all your actions.”

🚫 Too granular robotic tasks:

  • “Enter the request ‘Wholegrain Bread’ in the search bar… Add the first item on the page to the shopping cart… Select ‘Visa/Mastercard’ as a paying method…”

Task wording is important because it can frame the user’s thinking, influence their choice, and distort the results. There are many things to keep in mind, but here are the five key pieces of advice to start with. 

⛔️ Things to avoid:

  • Promotional language: “Check the new smart reporting feature, which generates a report in just one click.”
  • Leading tasks: “Open the ‘Map’ section and find the nearest branch office to your current location.”
  • Overly simplistic tasks: “Find your profile page.”
  • Strange tasks: “Add Elizabeth Windsor to the list of customers.”
  • Complex terminology: “On the dashboard, there is data visualisation with logarithmic scales.”

A couple of words about screen readers

Accessibility is a vast professional area that requires years of training and practice, but here is some basic knowledge that can help you organise the first testing sessions with visually impaired users or get insight into accessibility testing.

Screen readers are programs that convert on-screen information into a speech that plays through speakers or earphones. People who can’t see the screen use a keyboard or touchscreen to navigate through websites or applications. Before testing with visually impaired users, it is strongly recommended to familiarise yourself with screen readers.

Standard readers are embedded into desktop and mobile operating systems and can be launched in the “Accessibility” settings: VoiceOver — Mac and iOS; TalkBack — Android; Narrator — Windows. As mentioned above, people with disabilities usually use advanced third-party readers, for example NVDAor JAWS (Windows).

⚠️ Note: Screen readers rely on webpage markup and what they read doesn’t always equal what a sighted person sees on the screen.

People with normal vision scan the screen and can quickly jump from one place to another, while users of screen readers can “see” only one element at a time — a button, input field, or headline — and cannot easily skip some portion of content.

If you want to help a person navigate through the website during a testing session, beware of giving hints that don’t make sense.

🚫Not helpful hints:
“At the top of the screen.”
“Scroll to the bottom.”
“Look at the item in the corner.”
“Press the cross icon.”
Helpful hints:
“Please go to the next/previous item.”
“Go to the last heading/link/button.”
“Navigate to the first element in the list.”
“Please close the modal window.”

After all the testing sessions, the team organises all the received feedback, prioritises it, and plans to implement improvements.


We are advising our clients to test accessibility with end-users who have disabilities. At Eleks, we like to take a human-first approach to technology, and accessibility testing is a good step in this direction. It will ultimately benefit both the consumer (whether disabled or not) and the company itself.

For more news from Top Business Tech, don’t forget to subscribe to our daily bulletin!

Follow us on LinkedIn and Twitter

Amber Donovan-Stevens

Amber is a Content Editor at Top Business Tech