A few weeks ago, I bought a Levoit air purifier for my room. I set it up, plugged it in, and turned it on. It lit up, made a soft humming noise, and everything seemed to be working perfectly. For almost a month, I thought I was breathing clean air. I wasn’t.
Figure 1: External view of the air purifier. Nothing suggests additional setup is required.
I wanted to improve the air quality in my room. The setup felt straightforward. The top panel had touch controls with icons for power, fan speed, sleep mode, timers, and a few others that weren’t immediately obvious.
Figure 2: Control panel with multiple icons, some of which are unclear.
I pressed the power button and the lights turned on, and I could hear air moving, so to me it felt like everything was working. Based on what I’ve used before like those standing fans, my mental model was that if a device turns on and air is coming out, then it’s doing its job. So in my head it was like: air purifier is on, air is moving, so the system must be working. Because of that, I didn’t really question anything or check further.
Figure 3: Button layout and a specific icon that appears to suggest rotation but does not behave as expected.
The control panel had a bunch of icons arranged in a circle. Some of them made sense right away, like the power button and fan speed, but others were honestly confusing. There was one button that looked like it might rotate the airflow or do something like oscillation. Based on my past experience with fans, my mental model was that this button would probably make the air move in a different direction. But when I pressed it, it actually didn’t seem to do anything at all, which made it even more confusing.
This is a problem with mapping. Ideally you should be able to predict what will happen, but here that wasn’t the case. I had to guess what the button did instead of just knowing.
It also relates to affordance, which means how a design suggests what action is possible. For example, a handle suggests pulling, and a button suggests pressing. In my case, the icon looked like it was suggesting rotation or movement, but the result was not clear. It looked like it should do one thing, but either did something else or just wasn’t obvious. This increased my cognitive load, which means the extra mental effort I had to use just to understand the control. Instead of using the purifier naturally, I had to stop and think about what the button meant, which broke the flow of the experience.
After about a month, I randomly came across an Instagram reel in my feed where someone was opening their air purifier and removing the plastic from the filter. That made me pause for a second, because I didn’t remember doing that when I set mine up. So I went and checked mine. When I opened the purifier, I found the filter still wrapped in plastic. At that moment, I realized I hadn’t been purifying air this whole time.
The biggest issue was feedback. Feedback means the system gives the user information about what is happening. The purifier gave me feedback through lights, sound, and airflow. Those signals told me the device was on.
However, the feedback was incomplete. It did not tell me whether the filter was ready to use, whether the plastic had been removed, or whether air was actually passing through the filter correctly. In my experience, the purifier basically said, “I am on,” but it never confirmed, “I am filtering air.”
That difference matters because the main purpose of the product is not just to turn on. The main purpose is to clean air. It clearly told me the machine was powered on, but it did not tell me the one thing I actually cared about: whether my air was being cleaned.
The clean design was helpful because the purifier looked simple and not intimidating. The touch controls responded quickly, and once I removed the plastic, the airflow felt much better.
At the same time, the clean design also hid the problem. Since the purifier looked finished from the outside, I had no reason to think I needed to open it. A simple warning sticker or setup reminder would have prevented the mistake.
The biggest improvement would be a clear label on the outside that says: “REMOVE FILTER PACKAGING BEFORE USE.” This label should be placed somewhere the user cannot miss before turning the purifier on.
The purifier could also give better feedback during setup. For example, the app or control panel could show a first-time setup message reminding the user to check the filter. Even a small red tag attached to the filter cover would help.
The unclear button icons could also be improved with more familiar symbols or small text labels. That would make the controls easier to understand without forcing the user to guess.
Overall, this experience taught me that a product can look like it is working while still failing the user. Good UX is not only about making a device look clean or modern. It also has to guide the user through the parts they might not know to check.