NVDA Is the Screen Reader I Test With—Here's Why Your Team Should Switch

There's a persistent myth in web development that JAWS is the only screen reader worth testing with. It's wrong. I've spent years in accessibility testing, and I'm here to tell you: if you're not testing with NVDA, you're missing the point.

This isn't about hating JAWS. It's about understanding what real screen reader users actually use, and building a testing workflow that reflects reality—not legacy assumptions.

The Reality of Screen Reader Usage

Let's start with the data. In WebAIM's 2024 Screen Reader User Survey, 65.6% of respondents reported using NVDA, making it the most commonly used screen reader—even ahead of JAWS, which came in at 60.5% (WebAIM). And when it comes to primary desktop screen readers, NVDA holds 37.7% share, nearly tied with JAWS at 40.5% (WebAIM). So the idea that JAWS is the overwhelming standard is simply false.

Imagine you're a developer at a mid-sized company, tasked with making your web app accessible. You've been told to test with JAWS because 'that's what blind users use.' But the survey says otherwise. By focusing only on JAWS, you're testing for a minority—and ignoring the free, open-source screen reader that's used by more than half of the screen reader community.

Why NVDA Deserves Your Testing Time

NVDA (NonVisual Desktop Access) is developed by NV Access, a non-profit that's been supporting it since 2007 (NV Access). It's free, which means no licensing headaches for your team. It's also translated into over 55 languages and used in more than 175 countries (NV Access). That global reach means you're testing with the same tool that a huge portion of the world's screen reader users rely on.

And here's the kicker: NVDA is open source. That means you can inspect the code, understand exactly how it parses your HTML, and even contribute fixes. Try doing that with JAWS, which is a closed commercial product.

The Mobile Reality: It's Not Just Desktop

But here's where things get really interesting. The same WebAIM survey found that 91.3% of respondents use a screen reader on a mobile device (WebAIM). And among mobile users, 70.6% use Apple's VoiceOver, while 34.7% use TalkBack on Android (WebAIM). So if you're only testing on desktop, you're ignoring the vast majority of how people actually access your site.

Now, I'm not saying you should ditch desktop testing entirely. But you need to expand your toolkit. VoiceOver is built into every Apple device, and TalkBack is on Android. Both are free. So your testing matrix should include NVDA on Windows, VoiceOver on iOS, and TalkBack on Android—at minimum.

What Actually Breaks on Your Site

Let's get concrete. The WebAIM Million analysis of over one million home pages found that 95.9% have detectable WCAG failures, with an average of 56.1 errors per page (WebAIM Million). The most common issues? Low contrast text (83.9%), missing image alt text (53.1%), and missing form input labels (51%) (WebAIM Million). These are the exact problems that screen reader users encounter every day.

Imagine you're testing a checkout form. With NVDA, you can quickly navigate through the fields and hear whether each input has a label. If the label is missing, NVDA will say 'blank' or nothing at all—immediately signaling a problem. With JAWS, you might get similar feedback, but why pay for JAWS when NVDA does the same job for free?

The Cost Argument Isn't Just About Money

Yes, NVDA is free, and JAWS licenses aren't cheap. But the real cost argument is about efficiency. NVDA is lightweight, starts up fast, and doesn't hog your system resources like JAWS can. In a fast-paced development environment, that speed matters.

But here's a deeper point: by testing with NVDA, you're aligning with the tools that the majority of screen reader users actually rely on. That's not just a cost saving—it's a correctness issue. If you test only with JAWS, you might miss bugs that only appear in NVDA's rendering engine. And given NVDA's popularity, those bugs will affect a huge number of your users.

What About the Other 90%? A Reality Check

Now, I want to zoom out for a moment. The WHO estimates that over 2.5 billion people currently need assistive technology, and that number is projected to rise to 3.5 billion by 2050 (World Health Organization). But here's the sobering stat: about 90% of people who need assistive technology don't have access to it (World Health Organization). Screen readers are just one tiny piece of that puzzle.

So when you're testing with NVDA, you're not just checking a box. You're contributing to a larger effort to make the digital world accessible to everyone. And while NVDA is free, it's still a luxury that many people can't access because they lack the hardware or the internet connection to download it. But that's a problem for another day. For now, let's focus on what you can do: test with the tools that real users use.

What I'd Actually Do

Here's my concrete recommendation: if you're a web developer or QA engineer, download NVDA today. It's free, it's powerful, and it's what your users are using. Set up a testing routine that includes NVDA on Windows, VoiceOver on iOS, and TalkBack on Android. Run through your key user flows with each one, at least once per sprint.

And don't stop there. Use the WebAIM Million's most common errors as a checklist. Fix your low contrast text, add alt text to images, and label your form inputs. These are the basics, but they're the basics that 95.9% of home pages are getting wrong (WebAIM Million).

Testing with NVDA isn't just about saving money—it's about testing with the tool that the majority of screen reader users actually use. That's the kind of pragmatic accessibility that makes a real difference. So go ahead, download NVDA, and start testing. Your users will thank you.

Sources

• WebAIM Screen Reader User Survey - https://webaim.org/projects/screenreadersurvey10/
• WebAIM Million - https://webaim.org/projects/million/
• NV Access - https://www.nvaccess.org/about-nv-access/
• World Health Organization - https://www.who.int/news-room/fact-sheets/detail/assistive-technology