The cycle of UX design

- A brief intro to UX design

Testing one user is 100 percent better than testing none.
Testing one user early in the project is better than testing 50 near the end.
— “Don’t make me think” Steve Kung

The first time I caught myself thinking about UX design was when I started reading the book called “The design of everyday things” by Donaly Norman. It made me realise, that it is not enough to have just “pretty” things — they also have to be practical and convenient to use. And this is what UX designers do. We make sure, that everything a consumer uses, makes sense and is as practical/useful and simple as possible.

Example: We experience UX designs everywhere in our daily lives.Take a look at this teapot. Do you notice any mistakes?

I used to work as a freelance web designer before I joined EasySize. One of the many great things about working in a startup company is the opportunities and challenges, that surround you in your everyday work. In this blog post, I would like to present the cycle through which my latest UX project in EasySize went through. Use it as guide or inspiration for your own work, and make sure to share your feedback.

Step 1: Get to know your product WELL

Flow chart

Make sure of having a clear picture of the product, that you are working on. Answer such question “what it does”, “how it works”, “what issue is it solving?” and make sure you understand the overall concept behind. What is more, make a flowchart and then a wireframe to help you understand the logic behind of user flow. This step is the foundation, the skeleton of your product, therefore it has to be strong and convincing. The most important thing - it has to make sense. If you are not sure about something, talk to your boss, your colleagues, ask their opinion or even conduct some pilot tests.

For me personally, it took around 1 month to get to know the product and the concept of the EasySize product. I was lucky enough, that EasySize product is clean and clear in algorithm and the logic behind the technology is simple and easy. What is more, the product was already up and running, therefore I was able to save loads of time and just go and test some different prototypes (the next step).

During the first weeks in EasySize, I have learned the data that we collect, how we use it and what the different scenarios in customer experience are. Moreover, it was easier for me to begin working at EasySize, because I do online shopping for myself quite often. I do believe, that UX designers have to somehow relate to the project he/she is working on, as this broadens his/her knowledge, experience and adds value to the design process.

When I have gathered all the information about the product, I work on, I am then ready to put this knowledge to the first interactive prototype.

Step 2: Make your first interactive prototype

Prepare your first idea. During this stage, you will need to put all the gathered information to your first prototype. The idea of this prototype is to present your concept. It doesn’t need to be perfect. But try your best to make sure 60%-70% of your general concept is presented via this prototype. However, it does not mean that you first interactive prototype can be just a piece of paper with random sketches and a few words. In other words, the prototype has to be clear enough to present the core concept without too many explanations. Moreover, from my own experience, depends on the scale of the product. If it is not time-consuming, I would also recommend to include the basic interactions and navigations. This interactive prototype will have a better representation of your design idea, therefore, you will be able to get the most out of it after the first test (step 3).

Frist prototype made by Axure

There are many great UX design software online or you can download for free, most of them have free trial for 30 days. Why not use them, if they are really easy to learn and extremely fun to play with (these include Axure, UXpin, Precursor and many more.) My personal favorite software is Axure, because it allows me to upload demos on the Axure Share, which I am then able to share with my EasySize team. Because of this, I always receive quick and valuable feedback from my coworkers.

Now, let’s move to the step 3 to test your prototype.

Step 3: Test

It is time now for the most important part of your UX project. This is the stage where you need to start exploring “outside the box” by asking for people’s feedback and insights. In other words, testing is the core asset of the UX design. This step is divided into to two main parts: use usability test to identify main problems and A/B test to find out the solutions.

  • Conduct an usability test

Generally speaking, usability test is a qualitative test. It includes interviews, open questions and observations. The test result analysis is not aiming to find statistical patterns, but rather the problems like: whether the tester can understand the concept or the words he/she is seeing, whether they are looking for it in the right place and etc. Therefore, if the test participants ask such questions as “what’s that?” or “what’s next?” , you need to be alerted and write them down for further analysis.

From my own experience in running tests for the EasySize products, I have noticed that usually 3–5 participants are enough to realise the main issues. The methodology I use for creating such questions is rather simple- the questions must be broad and general. This allows the participants to think outside the box and suggest insights, which would never be possible, if you have asked them something specifically.

Once you have finished the usability test, list all the feedback and sort them out accordingly to the degree of impact and mark their complexity.

Pick the element with high impact and low complexity and make the change through the A/B testing. (See below)

  • Conduct A/B test

A/B testing is also called “the split test”. The main idea behind it is to have a different solution for the same part of the product and then run the test to check which one performs better. The performance is usually measured by the conversion rate.)

There are many great online A/B test service tools worth to try. I personally rely on the Optimizely split test, which requires at least 100 unique visitors’ interactions to draw the performance conclusions.

The example of the EasySize A/B test

The A/B testing provides valuable statistics which give the confidence for the decision making process. For example, a while ago, I was testing how convert button in EasySize lightbox (“Find my size”) performs versus no button (automatic conversion) (look at the example below.) We found out that the original button (“Find my size”) performed better than no button by 42 %.

…and now what?

This blog post wouldn’t have been called UX cycle if you didn’t have to do…a cycle again. Simply put, you will need to test your product again and again, once in a while. Adding new feature to the lightbox? Test different options and which one performs the best! Introducing a new website feature? Run a test to understand,whether users understand it, and if they don’t, then why? My honest advice for young UX designers would be:

  • Have as many tests as possible in the very early stage of your project/product
  • Find the balance between the feedback from others and your own insights.

What is more, charge yourself with creativity and inspirations while reading books about the UX design, attend events, read various articles online… Take a look at those small usability details — why there is one solution, and not the other? How easy for users is to use this website, and why? This will train your mind to think more critically for the next UX project.

And now, go to one of the shops EasySize is available at, choose “What’s my size button” and use the pop up lightbox. Share your insights with me! I am waiting for your feedback - let’s talk!

Haochen Xu, UX designer at EasySize