How to Read an Ebook Like a Tester – A Quick 10-Minute Accessibility Test
Written by
Monika Zarczuk-Engelsma
Posted on
Jul 23, 2026
Category
Accessibility
Many myths have sprung up around the accessibility testing of e-books. For some publishers, it sounds like a process requiring specialized equipment, knowledge of standards, multiple applications, and—ideally—the involvement of a blind person at every stage of production.
Indeed, a full accessibility audit is a distinct field. At the same time, however, there are things that can be done much earlier.
Ten minutes.
Often, that is enough to catch a large portion of the most annoying errors that would otherwise bother readers.
Step 1. Open the Book and… Stop Looking
It sounds strange, but it's a very good exercise.
Open the e-book and try not to judge it by its appearance for a moment. Instead of wondering whether it looks aesthetically pleasing, start asking questions:
Do I know where I am?
Do I know where I can get to?
Do I understand the structure?
Many publishers know their books so well that they don't notice problems. Just like an author doesn't notice typos after the tenth reading.
Try approaching the publication as someone who sees it for the first time.
If, after just a few seconds, it is difficult to figure out where a chapter begins or how to return to the table of contents, it is worth pausing.
Step 2. Open the Table of Contents and Do Not Scroll
This is one of the fastest tests.
Don’t scroll manualny.
Enter the table of contents and try to:
go to a selected chapter,
return to the beginning,
open a subchapter,
go to a footnote and back.
If something doesn't work or requires scrolling through dozens of screens, that’s a warning sign.
For many blind people, the table of contents is not an extra feature; it is the primary way of navigating a book.
Imagine a 500-page textbook. A sighted person can quickly skim through the publication. A screen reader cannot do that.
A lack of functioning navigation very often means a real loss of functionality.
Step 3. Enlarge the Text to an Absurd Size
Enlarge the text as much as possible.
Not 120%.
Not 150%.
Until the moment you decide that "nobody reads like that."
And see what happens.
See if:
the text still fits on screen?
nothing overlaps?
fragments don’t disappear?
horizontal scrolling doesn't appear?
This quickly reveals whether the e-book was designed as a digital publication or as a static image of a printed page.
Interestingly, such a test often reveals problems for sighted people as well—when using phones or small screens, or when reading in the evening.
Step 4. Turn on Text-to-Speech
This is a step that many publishers never take.
There is no need to install specialized software right away.
Built-in reading features are available on computers and phones.
On Windows, you can launch Narrator using the shortcut:
Ctrl + Windows + Enter.
On iPhone and iPad, you can turn on VoiceOver in:
Settings → Accessibility → VoiceOver.
On Android phones, this usually is:
Settings → Accessibility Settings → TalkBack
(the name may vary slightly depending on the manufacturer).
It’s not about learning how to use a screen reader. You simply need to launch it for a few minutes and try opening one of your own ebooks. Listen to two or three pages of the book.
Don’t focus on the content.
Listen to what technology is saying.
Do the following messages appear:
"graphic"
"graphic"
"graphic"
Are strange numbers being read out?
Do the headings sound like ordinary text?
Are the footnotes clear?
This is often the moment when publishers hear their own book for the first time the way a screen reader perceives it.
And it very often turns out that things that look visually perfect sound completely chaotic.
If you want to take it a step further, it is worth downloading NVDA—a free, open-source screen reader for Windows that is widely used by blind people for daily reading, as well as for work and study.
Download page:
https://www.nvaccess.org/download/
Importantly, NVDA does not require immediate installation. You can run a temporary version and simply check how the book behaves during playback.
Step 5. Open the First Table and Try to Answer One Question.
Don't analyze the entire table.
Try to answer one specific question.
For example:
Which year had the highest sales?
Which column contains the results?
How is the data related?
If you're starting to get lost, your screen reader will likely have trouble too. It's also worth practically testing how your screen reader handles this type of data.
In particular:
whether there are column headers,
whether there are no empty cells serving a visual function,
whether the data is in a logical order.
A well-prepared table provides answers.
A poorly prepared one forces the user to guess.
Step 6. Look for Anything that Says "See below"
This is a small test that reveals a lot.
Look for the following phrases in the book:
"in the illustration"
"as shown below"
"see the chart"
"highlighted in colour"
"we in the photo we can see"
And then check:
Can someone who cannot see this illustration still understand the content?
If the answer is no, probably something is missing.
Not every graphic requires a long description.
But any important information should also be accessible in ways other than visual.
Step 7. Open the Same Book in Different Applications
This is a very underrated test.
Open the same EPUB file in at least two different programs or on two different devices.
For example:
Thorium Reader (Windows, macOS, Linux),
Apple Books (iPhone, iPad, Mac),
PocketBook Reader (Android, iOS),
Kobo reader,
Kindle (if you are also preparing a version for this platform),
the distributor’s or sales platform’s app.
Don't just check the appearance.
Check:
if the table of contents works,
if the text size can be changed,
if illustrations disappear,
if it is possible to navigate via headings,
if footnotes work correctly,
if the book behaves similarly after changing settings.
If something starts working differently, it is worth checking why.
The point isn't for the book to look identical everywhere; the point is for it to be comfortable to use everywhere.
This is particularly important for EPUBs, as different applications may interpret the same code and CSS styles slightly differently.
Step 8. Ask Yourself One Uncomfortable Question.
If I didn’t know this book, would I still be able to use it?
That’s a very good filter.
Because publishers often unconsciously fill in the gaps with their own knowledge.
They know that the chapter begins here.
They know that this illustration shows a diagram.
They know that this footnote should lead elsewhere.
The reader doesn’t know this.
And that's why testing isn't about checking whether everything was done according to plan.
It involves testing whether the book works for someone who is not familiar with the plan.
Lastly, the most important thing
This ten-minute test does not replace an audit or user testing.
But very often it is enough to pick up on things that later determine the entire reading experience.
Working table of contents.
Logical headings.
Annotated graphics.
Flexible layout.
Listening and navigation options.
These are not additions for the few.
This is very often the difference between a book you can buy and a book you can read.
Useful links:
https://atonce.pl/jak-sprawdzic-dostepnosc-cyfrowa-dokumentu/
https://mintconcept.pl/dostepnosc-cyfrowa-e-bookow-co-nalezy-wiedziec/
https://www.a11y-collective.com/blog/ebook-accessibility/
https://accessibility.huit.harvard.edu/manual-testing-accessibility
https://elearning.adobe.com/2025/08/inclusive-interactive-ebooks-a-guide-to-accessible-elearning/