How to conduct mobile app user research with Maze

Mobile now drives the majority of web usage, accounting for 55.9% of global traffic. Teams need to focus on mobile-specific behaviors, from how users tap, swipe, scroll, and complete forms to how they move through key flows on smaller screens.

In this guide, we look at how to test prototypes, live apps, and mobile websites. We also explain how to recruit the right participants, use AI-powered user insights to validate flows, and improve the mobile experience.

But first, let’s start with why it matters.

Mobile experiences are shaped by context. People use apps on the move, with interruptions, limited attention, and varying network conditions. This changes how they navigate screens, choose the next steps, and complete tasks—meaning mobile flows need to be tested for this unique context.

Here are some of the reasons user research is crucial for mobile product development:

Mobile apps are used in a different context from desktop products. Smaller screens, touch-based navigation, changing connection quality, device permissions, interruptions, and shorter sessions can all affect how users move through a flow and complete tasks. That context makes mobile-first UX research essential for spotting friction that desktop testing can miss.

An AI‑first user research platform like Maze supports mobile app testing across research methods, so you can observe user behavior, validate design decisions with qualitative and quantitative data, and optimize your mobile experiences.

Maze brings your mobile UX research into one place. You can test native apps and mobile web flows, from early prototypes to live experiences, and combine behavioral data with direct user feedback.

Maze also helps you recruit the right participants for mobile app research through Maze panel. You can access a pool of millions of B2B and B2C participants across 150+ countries, then narrow your audience with 400+ filters and custom screeners. This helps ensure your mobile feedback comes from people who match your target users.

You can also add qualifying screener questions directly to your panel order, so bad-fit participants are filtered out before they reach your study. To get fresh eyes on each study, use the ‘Exclude past participants’ filter to control whether repeat participants are eligible. This helps you capture cleaner feedback from people who match your audience and research goals.

Let’s break this down into a clear, step-by-step process for testing mobile experiences with Maze.

Here’s a simple framework to plan, run, and analyze mobile research with Maze.

Start by narrowing your mobile research goal. Are you trying to understand whether users can complete a key task, where they drop off, why they hesitate, or how the experience performs on a real device?

The clearer your goal, the easier it is to choose the right Maze study, write focused tasks, and measure the right outcomes. You could be interested in task success, drop-off points, completion time, user paths, or open-text feedback—all of which provide different insights that answer different research questions.

For example, if you’re testing onboarding, you might look at whether users understand the first few steps and know what to do next. If you’re testing checkout, you might focus on task success, drop-off points, and moments of friction. For a new feature, you may want to learn whether users can find it, understand its value, and complete the intended action without extra guidance.

Once you know what you want to learn, you can get started in Maze. From your dashboard, click ‘New study’ to choose the study format that best fits your mobile research goal.

Maze gives you three study options:

After choosing the study type, add your project details and decide how you want to build the study. You can use the AI study builder to generate a study plan, tasks, and questions; choose a mobile usability testing template to customize a pre-made setup; or start with a ‘Blank study’ and add everything manually.

Now, let’s take a look at how to set up these three types of mobile app research studies.

The first step is to add blocks for the mobile experience you’re testing from the left-hand panel:

Next, open ‘Study requirements’ for your unmoderated study. Choose which devices are allowed. For mobile testing, select ‘Mobile’ or ‘Mobile and tablet’ so participants on unsupported devices see an error and cannot proceed.

Decide how participants should join:

Click ‘Preview’ from the builder and open the preview link on a mobile device.
Step through the entire flow like a participant. Confirm that each block loads, screens fit the device, and navigation feels natural on a small screen.

Choose an AI‑moderated study when you want interview‑style insights about how people use mobile apps, without scheduling and handling every session yourself. Maze AI moderator runs the conversation based on your goals, while participants join from their mobile browsers and talk through their experiences.

When you create an AI‑moderated study, you have two ways to get started:

After setup, you’ll see your ‘Plan’ on the left panel, with an introduction, one or more goals, and a closing. From here, you can edit the existing goals or click ‘Add goal’ to create more questions about your mobile app—for example, first impressions, navigation, or key tasks like sending money or checking an order.

For each goal, choose:

You can preview the experience at any time with ‘Start preview’, then move on to recruitment once the questions about your mobile app feel clear and focused.

Choose a moderated study when you want to observe people using your mobile app or prototype in real time, ask follow‑up questions, and see their screens. Maze uses interview studies together with a built-in conferencing tool to run these sessions end-to-end (but you can also connect your favorite video conferencing tool if you’d prefer!).

When you create a moderated study, you first choose how you’d like to start:

Here’s what to do next, depending on which option you choose here.

On the ‘Recruit’ tab, click ‘Start recruitment setup’ to define your interview details and availability.

Open ‘Study settings’ and, if you’re using Maze conferencing, enable the ‘Conferencing room’ for this study and the ‘Redirect to Zoom App’ option. This sends participants straight to the Zoom app when they join, which is required for screen sharing on mobile.

In your invite, tell participants which app to install (for example, Zoom) and whether you want them to join on mobile only or with both desktop and mobile devices.

For one‑device sessions, ask participants to join the call on their phone, start screen sharing from the conferencing app, and open your mobile prototype or app from the link you share in chat or email. The recording captures their mobile screen and think‑aloud feedback.

For two‑device sessions, ask participants to join on a desktop for conversation and camera, then connect their phone as a second device just for screen sharing. This reduces the everything-on-one-small-screen problem and works well for longer interviews.

This is where you import recordings from your pre-existing mobile interviews. Click ‘Add your first session (or Create session)’ to upload a recording and add participant details such as name, date, and device.

Maze then generates transcripts, lets you document and group highlights, and helps you build reports from your mobile sessions. You can tag moments where people struggled with specific screens, navigation, or gestures in your app.

Once your study is configured, click ‘Go live’. Maze will then take you to ‘Share your study with participants’, where you decide how to bring mobile users into your study.

If you’ve followed along so far, you’ve set up your mobile study in Maze, chosen your preferred study type, and recruited the right participants. Now it’s time to review how they moved through your tasks, where they hesitated or changed course, and what they shared about the experience.

After participants complete your study, open the ‘Results’ tab. Maze automatically organizes the data for each block, so you can review how mobile users completed tasks and what they said about the experience.

Key metrics include:

You can also filter results by outcome, device, and other attributes. This helps you zoom in on specific groups, such as users who didn’t complete a task or participants using a particular mobile device.

Once you’re ready to share findings, open the ‘Report’ tab. Maze creates an editable report with key metrics, charts, and screenshots. You can add commentary, customize the report, and share it through a public link, embed it in tools like Notion or Jira, or export it as a PDF.

Mobile UX comes with its own constraints, including smaller screens, varied devices, and shorter attention spans. Here are a few best practices to help you design Maze studies that respect those constraints and still give you actionable insights for product decisions.

Next, let’s take a look at a real customer example to see how this approach to mobile testing with Maze works in practice.

Doctolib is one of Europe’s largest healthcare platforms. It helps patients book appointments and manage care and works with more than 100,000 health professionals and 2,000 health facilities worldwide.

As Doctolib’s mobile app had grown beyond appointment booking, users could now manage documents, relatives, video consultations, and more. But many people were still using the app mainly for appointments, and internal data showed that at least 70% of users were not engaging with the newer services. The team needed to rethink the app’s navigation so these features were easier to find without complicating the experience.

To validate a new navigation approach, Doctolib used Maze to run unmoderated mobile prototype tests in France and Germany. They tested real tasks—including booking an appointment, checking appointment details, previewing a medical document, and accessing relatives. Within a few days of launching the Maze study, they collected 4,000+ responses in France and 500+ responses in Germany.

The results gave the team clear direction:

This is a good example of how mobile app research helps teams move beyond assumptions, see how people actually navigate on smaller screens, and make confident changes before releasing them to millions of users.

When you’re designing or improving an app or website for mobile, it’s easy to rely on your team’s assumptions about what people notice, tap, or understand. With the right setup, you can test real mobile experiences early, validate key flows, and keep improving based on user behavior.

Maze brings mobile app research into a single workflow. You can test prototypes, live apps, and mobile websites, recruit the right participants, and analyze results without switching tools. Whether you’re refining navigation, improving onboarding, or validating a new feature, you can move from idea to insight at the speed of product development.

With Maze, teams build better mobile experiences, one test at a time.