Skip to main content

UX and research · Intermediate · 16 min

Include disabled people in your user research

By Padding · Updated 18 Nov 2026
Any document editor, a spreadsheet and your usual video call tool · November 2026

You will have a research plan and a per-participant session setup that let disabled people take part on their own terms, from recruitment and consent to session length, assistive technology and payment.

On this page

Most teams say their product is for everyone and then test it with people who look, see, hear, move and read like the team. Disabled people are left out less by decision than by default: the recruitment agency's panel has few of them, the screener asks nothing about access, the session is 60 minutes back to back, and the prototype is a set of flat images a screen reader cannot read. Each of those is a small planning choice. Together they mean the research never hears from the people most likely to hit a wall in the product.

This tutorial is about those planning choices. It is not a guide to accessibility audits or to WCAG. An audit tells you whether the product meets a standard; research with disabled participants tells you whether real people with real setups can get their task done, and what they work around. You need both, and this covers the second.

You will build two things. The first is an addition to your research plan: who you are recruiting, through whom, how you pay them and what you change about the sessions. The second is a short session setup sheet, one per participant, that records what each person needs before you meet. Keep them in the same folder as the rest of the study.

A note on language. This tutorial says "disabled people", which many disability organisations prefer; others prefer "people with disabilities". When you write to participants and partner organisations, follow the language they use.

Step 1 · Decide what you need to learn and who can tell you

Start from the research question, not from a list of impairments. Ask which parts of the product depend on seeing, hearing, fine movement, reading, memory or concentration, and who would be affected by each.

For a booking flow, for example: picking a time from a calendar grid depends on vision and pointer precision; a code sent by text message depends on switching apps quickly and remembering six digits; a session timeout depends on speed. That points you towards screen reader users, people who use the keyboard or switch access instead of a mouse or touch, and people with memory or attention difficulties.

Add a table to your research plan with four columns:

  • Part of the product (the screen or task)
  • What it asks of the user (for example, "read and pick from a dense grid")
  • Who is likely to struggle (described by what they use or experience, such as "screen reader users", "people using screen magnification", "people who find it hard to hold several steps in mind")
  • Number of participants you aim for

Aim for a small, deliberate mix rather than one token participant. Two or three people for each of your two or three main groups is enough for a qualitative study to show patterns. One blind participant can tell you a lot about their own setup, but not whether a problem is common to screen reader users.

Disability is also not separate from your other recruitment criteria. If your product is for parents booking appointments for their children, recruit disabled parents booking appointments for their children, not disabled people in general.

Check it worked

Your research plan has a table with at least three rows, each linking a specific screen or task to a named group and a target number. No row says only "disabled users".

Step 2 · Recruit through community organisations

General panels and recruitment agencies often have few disabled participants, and the ones they have may already have been through many studies. Local and national disability organisations, user-led groups, charities for specific conditions, sight loss and Deaf community groups, and university disability services reach people who rarely see a research invitation.

Treat these organisations as partners, not as free recruiters. They are approached by many projects and get little back. Before you write to them, decide what you can offer: a payment to the organisation for its time, a summary of findings they can use, or a short session for their members on how the research shaped the product.

Write a short outreach message. Keep it in plain language and include:

  • Who you are and what the product does, in two sentences
  • Who you are looking for, described by what they do ("people who use a screen reader on their phone"), not by diagnosis
  • What taking part involves: format, length, dates and whether it is remote or in person
  • The payment, stated as an amount and how it is paid
  • That you will make adjustments, and one or two examples ("you can use your own device and software", "you can bring a support worker")
  • How to reply in more than one way: email, phone or text message

Send the message as plain text in the email body, not as a PDF attachment or an image. Ask the organisation how they prefer to share it with members; some use mailing lists, some use social media, some pass it on by phone.

Keep a recruitment log in a spreadsheet: organisation, contact, date contacted, what you offered, reply and number of participants referred. You will need it to thank people properly, and it becomes your contact list for the next study.

Check it worked

You have a recruitment log with at least three organisations in it and a sent outreach message that names the payment, the format and at least two ways to reply. The message contains no attachment that holds information found nowhere else.

Step 3 · Make the screener and consent accessible

Your screener and consent form are the first parts of the study participants use. If they fail, you lose people before the session and never find out why.

For the screener:

  • Use a form tool you have tested with a keyboard alone and with a screen reader, or offer to complete the screener by phone. Say so at the top.
  • Ask about access needs directly and in plain words: "Is there anything we can do to make it easier for you to take part?" with a free-text answer. Do not ask for a diagnosis unless the research question needs it; ask what technology people use and how they use the product type.
  • Ask which assistive technology people use, if any, as a free-text question. People combine tools, and a checklist of five product names will miss most of them.
  • Keep it short. Every extra question costs more for someone using switch access or dictation.

For consent, start from your usual consent form (see the linked consent tutorial) and adjust it:

  • Offer it in more than one format: an accessible document that works with a screen reader, a large-print version, an easy-read version with short sentences, and the option to go through it on a call.
  • Separate each permission ("we may record the screen", "we may record your voice", "we may quote you without your name") so people can agree to some and not others.
  • Let people give consent in the way that suits them: a typed name, a spoken agreement recorded at the start of the session, or a signature. Write down which one you accepted.
  • Say what happens if they need to stop part way through: they still get paid.

If a participant has a support worker or interpreter, the consent form should say that person is present and that what they say may also be recorded. Get the participant's consent for that, not only the support worker's.

Check it worked

You can complete your screener using only the keyboard, from first question to submit, without the mouse. Your consent pack has at least two formats saved in the study folder, and each permission has its own yes or no.

Step 4 · Plan session length, timing and breaks

The default 60-minute session with a 15-minute gap is designed around the researcher's calendar. For many disabled participants it is too short to do the tasks and too long to do them without a rest.

Assistive technology is slower for some tasks. A screen reader user moving through a page by headings, or someone using switch access, may take two or three times as long as a mouse user on the same form. That is not a problem with the participant; it is a finding about the product. Plan for it by cutting the number of tasks, not by hurrying people.

In your research plan, set:

  • Session length: offer a choice. For example, one 90-minute session with breaks, or two 45-minute sessions on different days. People with fatigue conditions often prefer the shorter option.
  • Breaks: a planned break every 30 to 40 minutes, and a line in your discussion guide reminding you to offer it. Tell participants at the start that they can ask for a break at any time.
  • Task list: mark each task as core or optional. Run core tasks first and drop optional ones if time runs out, without saying so to the participant.
  • Time of day: let participants pick. Energy, pain, medication and support worker availability all change through the day.
  • Gaps between sessions: at least 30 minutes, so a session that runs over does not eat into the next person's time, and you can reset any equipment.

Write the same flexibility into your scheduling. A booking link that only shows 60-minute slots between 10 and 4 rules out the choices above. If your scheduling tool cannot offer them, book by email or phone instead.

Check it worked

Your discussion guide marks every task as core or optional, has at least one written break prompt, and the core tasks alone fit into 45 minutes when you run them yourself at a slow pace.

Step 5 · Set up sessions around people's own technology

The most useful thing you can do is let people use their own device, their own assistive technology and their own settings. A screen reader on a borrowed laptop with default speech speed and no custom shortcuts is not how that person uses the web. Magnification, high-contrast modes, voice control and switch setups are personal and slow to rebuild.

That makes remote sessions the default for many studies, because the participant stays in their own setup. Plan these details:

  • Call tool: choose one that works with the participant's assistive technology and ask what they already use. Many screen reader users are fluent in one call tool and not another. Offer to use theirs.
  • Joining: send a direct link with no account to create. Include a phone dial-in as a fallback. Send instructions as plain text in the email, not as a screenshot.
  • Screen sharing: screen readers produce speech, not just pixels. Ask the participant to share their computer sound, or to turn up the speech so their microphone picks it up, and test this at the start. Without the speech, you see the screen but miss half of what they hear.
  • Captions and interpreters: for Deaf and hard of hearing participants, ask whether they want a sign language interpreter, live captions or a text chat. Book interpreters well ahead and pay them separately; do not rely on automatic captions for the whole session.
  • Tech check: offer a 10-minute call a few days before to check the link, audio and screen sharing. Pay for it as part of the session.

For in-person sessions, check the venue yourself: step-free route from the street, an accessible toilet, a quiet room, space for a wheelchair or guide dog, and seating with and without arms. Send the participant directions that cover the whole route from transport to the room, including where the entrance is. If they bring their own laptop, have power sockets and a table at a usable height.

Create the session setup sheet now. One row per participant in a spreadsheet, or one short document each, with:

  • Session format and chosen length
  • Device, operating system, browser and assistive technology, in the participant's own words
  • Communication needs (interpreter, captions, plain language, extra time to respond)
  • Support worker or companion attending, and whether they will help with tasks
  • Venue or call details and any tech check booked
  • Anything else they asked for

Store the sheet with the same access controls as your other personal data. Access needs can reveal health information.

Check it worked

Your setup sheet has a row for each confirmed participant with every column filled in, or marked "asked, none needed". You have sent the call link as plain text and booked a tech check for each remote participant who wants one.

Step 6 · Make the prototype usable with assistive technology

A prototype built as images with clickable hotspots has no text, no headings and no form fields for a screen reader to find. A keyboard user may not be able to reach the hotspots at all. If you test that prototype with those participants, you will learn only that the prototype does not work.

Before recruiting finishes, decide what you will test with each group:

  • Coded prototype or live product: best for screen reader, keyboard and voice control users. Even a rough HTML version with real headings, labels and buttons gives you useful results.
  • Design tool prototype: often fine for people using magnification, colour or contrast settings, and for cognitive accessibility studies, as long as text is large enough to zoom and the flow is simple to follow.
  • Conversation and existing products: when you have no accessible prototype yet, interview people about how they do the task today and ask them to show you a product they already use for something similar. That still tells you what your design must not break.

Test the prototype yourself first with the technology your participants named on their setup sheets. Turn on your phone's built-in screen reader or use only the keyboard, and go through each core task. You are not trying to become an expert user; you are checking whether the task can be completed at all. If it cannot, fix the prototype or change the task before the session.

Write down in the plan which version each participant will see, so the results are read in context later.

Check it worked

For each core task, you have completed it yourself with at least one assistive technology or input method your participants use, and noted the result next to the task in your discussion guide. Any task you could not complete is either fixed or removed from that group's sessions.

Step 7 · Pay fairly and cover the costs of taking part

Pay disabled participants at least the same rate as everyone else, for the whole time you ask of them, including tech checks and any extra time a session needs. If a 90-minute session is paid at the 60-minute rate, you are charging people for their access needs.

Add these lines to the plan's budget:

  • Participant payment at your standard rate per hour, for the full booked time
  • Support worker or companion costs, if the participant has to pay for their time
  • Travel, including taxis where public transport is not accessible, paid in advance or reimbursed quickly
  • Interpreters and captioners, booked and paid by you, never by the participant
  • Partner organisation payment, if you agreed one

Offer a choice of payment method: bank transfer, a voucher or gift card, or a payment to a charity. Some people who receive disability benefits or other support worry that a cash payment will affect it. Do not give benefits advice yourself. Say that a choice of method is available and that they can check with their adviser or a local advice service, then follow your organisation's rules for recording payments.

Pay quickly, ideally within a week, and confirm by the participant's preferred contact method that the payment has been sent.

Check it worked

Your budget has a line for each item above, even if some are zero, and the per-participant amount for a 90-minute session is one and a half times your 60-minute rate. Your outreach message, screener and consent form all state the same payment amount.

Step 8 · Run a pilot and review after each session

Run one pilot session before the real ones. Ideally the pilot participant uses the same kind of assistive technology as some of your participants, and you pay them in full. If that is not possible, run it with a colleague using the keyboard only or the built-in screen reader, and treat it as a timing and setup check.

During the pilot, time each core task and note anything that broke: the call tool, the screen share audio, a prototype control that could not be reached, a question that needed rewording. Update the discussion guide, the setup sheet template and the timings before the first real session.

After each real session, spend five minutes adding two things to the setup sheet:

  • What worked in the setup and what you changed on the day
  • Anything the participant said would make the next session easier

When you analyse the findings, keep track of which problems came from the product and which came from the research setup. A participant who could not complete a task because your prototype had no labels found a prototype problem; a participant who could not complete it because the live product has no labels found a product problem. Report the second kind with the same weight as any other usability finding, and pass the first kind to whoever builds the next prototype.

Finally, close the loop with the partner organisations. Send them the summary you promised, in an accessible format, and thank them by name in your recruitment log.

Check it worked

Your discussion guide has a dated change note from the pilot. Every completed session has an entry on the setup sheet under "what worked" and "next time". Your recruitment log shows the date you sent each partner organisation its summary.

Next steps

You now have a plan and a setup that disabled participants can use. Build on it with these:

Need this done for you?

Padding does this work for growing companies, from the first diagnosis to a system that keeps working.

Digital Transformation

Get new tutorials

New tutorials, about twice a month.