Let Your Users Show You What's Not Working6 min read
When you're getting ready to build a new website, app, or software suite, there is no substitute for observing the real people who will use it. These people already do the things your digital product will help them with - your goal is to make those things easier, faster, and more intuitive.
A sure path to this success is watching real people navigate the frustrations, work-arounds, and processes of the digital tools they're already using. You'll do this through user observation, the act of actually watching people use digital products to perform day-to-day tasks. It is the most important thing you can do for your project's Discovery.
Pssst… What is Discovery? It's the crucial data-gathering first step in creating any digital product. It's also a step teams often skip. When they fail to properly diagnose problems, they waste time and money in the long run building websites, apps, and software suites that don't work as they should. Read "Successful Digital Products Start With Discovery" for a more in-depth look.
Diligent User Observation Will:
- Show you what users actually do. Sometimes what they say and do are two very different things.
- Give you first-hand experience with user tasks, step-by-step, and the thought processes behind those tasks.
- Make it painfully obvious what's just not working for users.
- Reveal things you would never find out any other way, things you would never even think to ask about.
The Science of Conducting User Observation Sessions
Where to start with user observation? Do you just walk into your users' offices or homes and hover over their shoulders? Grab someone off the street and drag them in front of a computer screen? Bring users in for a focus group and ask them how they use the current site, app, software, etc? Nope. This is a scientific undertaking. No focus groups or user kidnappings here. We need a controlled environment and a consistent process.
Follow these guidelines for observing users to gather accurate, usable data that will help inform the right path forward for your digital product.
1. Observe users in their own space.
Our controlled environment is the users' own office, call center, work phone in the field, home, or wherever they typically are when they use their sites, apps, and software.
Hang out with them in their space, all day if you can. Take in as much as possible about the place where they do their work. Is it hectic or calm? Are they mobile or stationary? Do they have multiple places they might use the digital product or is it always the same? Lots of noise or is it quiet? Are there windows? Do they have work-arounds plastered on the wall and their machine: notes to self about how to do a certain task or who to call when they're stuck? How much paper is around in general? Do they still have a fax machine?
All these observations give you data about your users' daily life and the myriad processes they use. These could be directly related to your project or only tangentially, but they are all a part of what influences your users. Don’t put your new Mac laptop in front of people who use an ancient Windows machine.
2. Users need to use their own equipment.
Another control - the people you're observing should use the equipment they usually use in real life. Don't put your brand-new Mac laptop in front of people who typically use an antiquated version of Internet Explorer on an ancient Windows machine with a tiny screen. Your data won't be accurate.
There are other considerations here too, both design and functional. Does each user have their own machine or do users share computers or consoles? Are they still using a black screen with green text that reminds you of the days of floppy discs? Is their screen enormous or ridiculously small? Do they have multiple screens? Do they whip out their cell phone whenever they need to do something? Do they predominantly use shortcuts on the desktop to get things done? Can they even access the internet from their machine? If so, how many privacy settings are in place? Can they get to document sharing, video sites, etc?
3. Put your users at ease.
It's nerve-wracking when people hang over your shoulder watching your every move as you do your job or complete tasks online. It's also common for the people you're observing to have no idea what you're there for. A typical first question from users will be, "What do you want me to do?" Reassure them that they're in control here - you want to see what they do every day, no more and no less. No pressure, no wrong answers. They might even find themselves excited to tell you - it's likely that no one else has ever asked them this obvious question. If your user goes silent, remind them to tell you what they’re doing.
4. Have users talk out loud.
Ask the people you're observing to tell you what they're doing as they're doing it. Have them narrate their actions from the start of a task to its completion. Or, if they can't finish it, have them tell you about that too. (Example: "Well, this is about the time I always call Bob to ask what to do next because I can never remember.")
If your user goes silent, remind them to tell you what they're doing. If they get stuck, ask: "What are you thinking about right now?" It's important to understand actions and the thought process behind those actions.
5. Ask task-related questions (but don't lead the user).
Ideally, you want to watch your users complete their usual tasks without prompting them. Begin with: "Show us what you do in a typical day." You can tell a lot about what's most important to them by what they show you first.
If someone does something unusual or something you're unclear about while completing a task, ask them about it. It's tempting to ask leading questions here - don't. For example, ask: "Why did you do that?" or "Why did you do it that way?" instead of "Why didn't you click the navigation? or "Did you see the home button in the navigation?" See the difference?
Use your questions to clarify what users are doing. If someone has navigated to a page and you're not sure of the steps they took to get there, ask them: "How did you get to this page?" If they start telling you about a process and they're not actually doing it, ask them: "Can you show me how you would do that?" Make sure you understand the steps they're taking to try to get the job done and the rest will take care of itself. Document everything. Bring a person along whose sole job is to take notes.
6. Take copious notes (and photos/video if possible).
Like any good scientist, you should write down your findings. Document everything. Bring a person along whose sole job is to take notes. Take photos and videos of areas and equipment when you can. Just be sure to ask permission and be careful to avoid photographing sensitive information.
Your notes should include:
- Steps users take to complete their tasks. (Examples: Clicked page in main navigation, used search, started from button on home page, etc.)
- Quotes. (Example: "I hate this thing, it never works!")
- Descriptions of the environment. (Examples: Users stand to do work including on computer, noisy, customers interrupt work constantly, etc.)
- Notes about the equipment. (Examples: Shares computer with 8 other people, still using IE 9, heavy firewalls for websites, etc.)
- Work-arounds. (Examples: Printed out a huge paper copy of site content because easier to use, IT number stuck to front of computer, etc.)
7. Don't just delegate this exercise.
Yes, bring your team members with you to observe users, take notes, and ask questions. But you need to be there too. You can't lead your team to UX victory if you didn't personally interact with the users as well.
Watching Real Users Will Show You What Has to Change
User observation reveals what works with existing systems and what doesn't. It shows what hacks users have created to complete something that's difficult and what absolutely makes them want to tear their hair out. Furthermore, it will make every single team member and stakeholder who participates an instant believer in the value of user experience. There's nothing like watching someone struggle and complain about an interface in real time to convince everyone that change is needed.