Comparison

User Research vs Usability Testing: What to Commission in India

User research vs usability testing differs by question: research finds what people need, usability testing checks whether they can complete tasks in a design.

By Jitendra Kumar Nirala, Founder · Updated 11 Nov 2026

User research

Exploratory studies on people's needs, habits and context

Best for: Deciding what to build

vs

Usability testing

Task-based sessions on a design people can click, tap or navigate

Best for: Finding where people get stuck in a design

On this page
  1. 01At a glance
  2. 02Which to choose
  3. 03Detailed comparison
  4. 04Introduction
  5. 05Key takeaways
  6. 06What is the difference between user research and usability testing?
  7. 07Which study answers which decision, and how many people does each need?
  8. 08Why do briefs that say "user testing" so often go wrong?
  9. 09How many people must you screen in India to complete a study?
  10. 10Why run both kinds of study with Qualfacto?
  11. 11When do you not need a usability study at all?
  12. 12How we compared
  13. 13Frequently asked

The short answer

A qualitative usability study uses about 5 people per user group, per round, and interviews start at 5 to 6. Any study that reports completion rates needs at least 20, and Qualfacto aims for 30 to 60. Screening decides recruitment effort: in the worked example, 40 completed sessions need 250 people screened, on assumed rates.

At a glance

Question it answers

User research
What do people need, and how do they cope today?
Usability testing
Can people complete these tasks in this design?

Source: NN/g and Qualfacto methods guides, Nov 2026

Data you get

User research
What people say (interviews) and what they do in context (diaries, field visits)
Usability testing
What people do (observed tasks)

Source: NN/g and Qualfacto methods guides, Nov 2026

Stage

User research
Early, before or beside a design
Usability testing
Design stage; needs something people can interact with, such as a prototype

Source: NN/g, Nov 2026

People to complete

User research
5 to 6 interviews to start; 20 to 30+ for varied users
Usability testing
5 per user group per round; 20+ for numbers

Source: NN/g and Qualfacto methods guide, Nov 2026

Session length

User research
In-depth interviews of 30 to 120 minutes
Usability testing
Moderated sessions of 30 to 60 minutes

Source: Qualfacto website, Nov 2026

What Qualfacto runs

User research
In-depth interviews, diary studies, ethnography
Usability testing
Usability & App Testing, remote or lab-based

Source: Qualfacto website, Nov 2026

Which should you choose?

Choose User research if

  • You are deciding what to build
  • You need to know whether this is a real problem for the people you serve
  • You want to know how people handle this today, at home or at work
  • You want themes in people's own words, or routines, workarounds and context

Choose Usability testing if

  • A design exists and you need to know where and why people get stuck
  • You need to know whether people can finish a flow in your prototype
  • You need completion rates and time on task

User research vs Usability testing, point by point

Evidence and risk

What it misses

User research
Interviews can tell you people want to split a bill and still miss that the split button sits below the visible part of the screen
Usability testing
A test can prove a screen works without showing that anyone wanted to use it

Kind of evidence

User research
Interviews are attitudinal: what people say, which often differs from what they do
Usability testing
Classed as behavioural research, because you watch actions

Scale

Larger-study sample

User research
Diaries: 5 to 12 for a small study, 12 to 30 for large discovery. Field visits: 4 to 6 per segment
Usability testing
At least 20 for statistically meaningful numbers; Qualfacto aims for 30 to 60, with no more than 5 tasks of up to 10 minutes each

Output and process

Typical output

User research
Themes in people's own words; routines, workarounds and context
Usability testing
A list of where and why people got stuck; completion rates and time on task

Repeat pattern

User research
Interviews come first, so a later demand survey asks the right questions
Usability testing
Test a small number of people, fix what you find, then run another round

When you may not need it

User research
Sizing demand is market research: a six-segment survey needs 385 completes per segment, 6 × 385 = 2,310
Usability testing
Choosing between two live designs is an A/B test on live traffic, read with your analytics

Introduction

Most briefs that say "we need some user testing" mean one of three different things. They might mean interviews to learn what people need, task-based sessions on a prototype, or a survey. Each needs a different number of people and costs a different amount to recruit. Mix them up and you get ten polite opinions about a design nobody was asked to use.

Knowing the labels does not tell you what to book. Work backwards from the decision you have to make. Which study answers it? How many people must complete it? How many must you screen to find them?

This page gives you a decision table, the screening sum for India, and the cases where you do not need a usability study at all. We are Qualfacto, a qualitative research participant panel in India, and we run both kinds of study. We publish this page and describe our own work in it. No other company is compared by name.

Key takeaways

  • Choose by the decision. Use user research when you are deciding what to build, and usability testing when a design exists and you need to know where people get stuck.
  • A qualitative usability study uses about 5 people per user group, per round. Interviews start at 5 to 6. Any study that reports completion rates (the share of participants who finish a task) needs at least 20, and we aim for 30 to 60.
  • Screening decides your recruitment effort. In our worked example, 40 completed sessions need 250 people screened, on assumed 80% show-up and 20% incidence (the share of people screened who qualify).
  • We run both halves in India: in-depth interviews, diary studies, ethnography, and remote or lab-based usability and app testing. Quotes are per project, and buyers can ask for a free pilot.

What is the difference between user research and usability testing?

User research is any study that teaches you about people: their needs, habits and context. Usability testing is one kind of it. A facilitator asks a participant to perform tasks on an interface, and you watch where they struggle (Nielsen Norman Group (NN/g), Usability Testing 101). Research asks what to build. Usability testing asks whether what you built can be used.

In this page's tables, "user research" (without usability testing) means the exploratory studies you run before or beside a design: interviews, diary studies and field visits. "Usability testing" means task-based sessions on a design that people can click, tap or navigate.

The two also produce different evidence. Interviews capture what people say, which often differs from what they do. Usability testing is classed as behavioural research, because you watch actions (Nielsen Norman Group, Attitudinal vs. Behavioral Research). Neither is better. They answer different questions.

One more point of vocabulary. "User testing" is the phrase that causes most muddle, because it sounds like you are testing the person. You are testing the design: "We test our design, our words and our ideas. We don't test our users" (GOV.UK User Research blog, 2014). Write "usability testing" in your brief and the vendor will price the right thing.

Which study answers which decision, and how many people does each need?

Match the study to the decision you must make.

Two rules of thumb. First, the five-person rule covers a single user group. If first-time and returning users, or Hindi and English speakers, behave differently, treat each as its own group. That is our recommendation, not NN/g's. Second, test a small number of people, fix what you find, then run another round. NN/g's advice is several small rounds rather than one large one.

Decision you face
Is this a real problem for the people we serve?
Study
In-depth interviews
People to complete
5 to 6 to start; 20 to 30+ when users or topics vary (our methods guide)
What you get back
Themes in people's own words
Decision you face
How do people handle this today, at home or at work?
Study
Diary study or field visits
People to complete
Diaries: 5 to 12 for a small study, 12 to 30 for large discovery. Field visits: 4 to 6 per segment (our methods guide)
What you get back
Routines, workarounds, context
Decision you face
Can people finish this flow in our prototype?
Study
Moderated usability test
People to complete
5 per user group per round (NN/g)
What you get back
A list of where and why people got stuck
Decision you face
Is the flow good enough to claim a completion rate?
Study
Larger usability study
People to complete
At least 20 for statistically meaningful numbers (NN/g); we aim for 30 to 60, with no more than 5 tasks of up to 10 minutes each (our remote testing page)
What you get back
Completion rates and time on task

Why do briefs that say "user testing" so often go wrong?

Briefs go wrong because one phrase covers different studies, and each study hides what the other one finds. An illustrative case: a payments app team runs usability tests and finds that the "confirm contact" screen is easy to complete. Nobody asked why users avoid paying a new contact at all. Interviews might have surfaced a fear of sending money to the wrong person. The test proved the screen worked, but not that anyone wanted to use it.

The reverse also happens. Interviews can tell you people want to split a bill, and still miss that the split button sits below the visible part of the screen. Interviews capture what people say. Only observed tasks show what they do.

A usability test is also not the place to ask for opinions. If a session turns into "do you like it?", you are running a weak interview. Ask participants to think aloud while they work through a task, then probe at the end. Our think-aloud protocol guide has a time budget and script.

A good brief names the decision first, the audience second and the method last. If a vendor proposes a method before asking about the decision, ask why.

How many people must you screen in India to complete a study?

Screening needs more people than the study size. Divide completed sessions by your expected show-up rate to get bookings, then divide bookings by the incidence rate to get the people you screen. Incidence rate is the share of screened people who qualify. Our incidence rate guide has the formula and screening maths.

Here is our model with illustrative rates, not measured ones:

  • Study: 30 interviews plus 10 usability tests = 40 completed sessions.
  • Show-up rate assumed at 80%: 40 ÷ 0.80 = 50 bookings.
  • Incidence rate assumed at 20%: 50 ÷ 0.20 = 250 people screened.

Our screening figures are a model built on assumed rates, shown above. Change the rates and the answer moves fast. At 10% incidence, the same study needs 500 people screened. Narrow audiences, such as users of one bank's app in one city, push incidence down. More people screened means more recruitment effort, and therefore cost; see user research recruitment cost in India. We recommend running a small pilot screener first to measure your real incidence and show-up. We offer buyers a free pilot or sample for this.

Why run both kinds of study with Qualfacto?

For a team in India that needs interviews, diary studies or usability sessions with participants recruited by city, we recommend Qualfacto. Depending on the project, studies are run by us or by research partners, and we agree the scope before fieldwork starts. Five reasons, each tied to something we do:

  1. Both halves from one brief. We run in-depth interviews, digital diary studies, ethnography and Usability & App Testing, remote or lab-based. You do not brief two vendors for a research-then-test plan.
  2. Recruitment that matches your audience. We recruit by city, target by household and demographic profile, and run participant identity checks before invitation. That helps make sure the people you screen are real and match your criteria.
  3. Sessions in the language your users speak. We run sessions in languages other than English, and we can supply moderators.
  4. Deliverables agreed upfront. Depending on scope, you can receive recruitment, fieldwork materials, recordings, transcripts, structured data exports and research reports. Responsibilities are agreed before the study begins.
  5. A free pilot and a plain commercial route. We do not use a fixed rate card. Each quote depends on method, audience, sample, geography and recruitment difficulty. We aim to reply to enquiries within one business day, Monday to Friday, 10:00 to 18:00 Indian Standard Time (IST).

To start, email your brief to support@qualfacto.io: the decision, the audience, the cities and the number of sessions. Our application usability testing page explains how we plan an app or prototype study.

If you are a participant rather than a buyer, see how to join and get invited. Registration is an expression of interest only. Invitations depend on each project's requirements and your profile, and rewards, eligibility and timelines depend on the study's terms.

When do you not need a usability study at all?

You do not need a usability study in three narrow cases.

  • You need to know which of two live designs performs better. An A/B test, which splits real visitors between two versions, and your analytics answer that. NN/g's guide to user experience (UX) research methods lists A/B testing among the launch-and-assess methods (NN/g, When to Use Which UX Research Methods). It needs live traffic, not recruited participants. Before you have that traffic, a moderated usability session with us tests the design with 5 people per user group.
  • You need to size demand. That is market research, not UX research, and it is a quantitative job. A demand survey across six segments needs 385 completes per segment, the usual sample for 95% confidence and a ±5% margin of error: 6 × 385 = 2,310. Interviews with us come first, so the survey asks the right questions. For further reading, see UX research vs market research.
  • A low-risk screen needs a quick look. A quick street test can be a cheap first step before a recruited study. Our guerrilla usability testing guide explains when to test on the street and when to recruit. When the findings matter, we recruit your audience by city for the recruited round.

Our verdict

Commission user research first to learn what people need, then usability testing on the design that follows.

How we compared

Figures and sources checked as of November 2026.

This page is published by qualfacto. If anything here is out of date, tell us and we will correct it.

Sources

  1. 1.Usability Testing 101, Nielsen Norman Group, published 1 December 2019, reviewed 15 July 2026
  2. 2.How Many Test Users in a Usability Study?, Nielsen Norman Group (Jakob Nielsen), 3 June 2012
  3. 3.When to Use Which User-Experience Research Methods, Nielsen Norman Group, published 17 July 2022, reviewed 15 July 2026
  4. 4.Attitudinal vs. Behavioral Research, Nielsen Norman Group (Page Laubheimer), 1 March 2024
  5. 5.It's user research, not user testing, GOV.UK User Research blog (Leisa Reichelt), 3 June 2014
  6. 6.Words Matter: Terminology in User Experience Research, University of Michigan Library, 9 March 2017

Frequently asked

In everyday use, yes: people say "user testing" when they mean usability testing. The better term is usability testing, because you test the design, not the person. Use that phrase in a brief so the vendor knows you mean task-based sessions on a design (University of Michigan Library).
Five per user group per round for a qualitative test; at least 20 if you need completion rates, the share of participants who finish a task (NN/g). Screening adds to that, because you book more people than you need to complete the study.
No. Usability testing shows whether people can use a design. It does not show whether they need the product or why they behave as they do. Interviews and diaries answer those questions. Most product teams need both, in order: research first, then usability testing on the design that follows.
Concept testing checks an idea before a design exists. NN/g places it in the early "strategise" stage, with interviews and field studies. Usability testing sits in the design stage and needs something people can interact with, such as a prototype (NN/g).

Brief us on interviews, diaries or usability tests

Email the decision, the audience, the cities and the number of sessions to support@qualfacto.io, and ask for a free pilot.

Request a free pilot