I hope by now you have an idea for what project you’d like to be working on. You also should have a solid problem definition to work with. Like I said before it is extremely important to have one. Here is what Jesse James Garrett, the author of The Elements of User Experience, has to say about it:

“Together, product objectives and user needs form the strategy plane, the foundation for every decision in our process as we design the user experience. Yet, amazingly, many user experience projects do not begin with a clear, explicit understanding of the underlying strategy. The key word here is explicit. The more clearly we can articulate exactly what we want, and exactly what others want from us, the more precisely we can adjust our choices to meet these goals.

Jesse James Garrett in The Elements of User Experience (page 36-37)

So please, please make sure you have a working problem definition before you move on.

Now it’s time to get to the actual design part of the process. You might be thinking, “Finally! What software should I use? How can I make those fancy looking animations that I see everywhere on Dribbble and Instagram?” I hate to break it to you, but we are not there yet.There are few more steps that we need to take before we are ready to produce final designs that you can proudly share with others.

First, we need to decide how we are going to solve the problem on a high-level. Let’s look at Spotify again. Our problem definition was “Improve song suggestions feature of Spotify”. How are we going to do that? Fancy animations definitely won’t do the trick! In fact, this particular problem might not require any UI changes at all, but let’s not jump ahead of ourselves.

When we have a task to improve an existing product it is best to start with figuring out what it does already. You might have done that as part of your research already. It might be that you are already identified particular problem users are having, and you have an idea for how to fix it. That’s great! But let’s assume for a moment that we don’t know. One way to find out would be to reproduce the process that users are going through, their journey through the app, and walk through it step by step seeing what works and what doesn’t. This process can be represented as a customer (or user) journey map. I encourage you to look it up, and try to make one for your project. It is one of the things you can include in your case study.

But I got ahead of myself. In order to construct the user journey map you need to decide who that user is that is taking the journey. This is where personas come in.

Personas are fictional characters that represent a particular segment of users. Defining personas is a big subject. It is tricky to do it right. It is beyond the scope of this newsletter, but let me know if you are interested in personas, and I will address it in one way or the other. I’d say for your first project you don’t need personas. What you can do is use yourself as the hero of the journey.

Let’s come back to improving song suggestions in Spotify. You don’t need to map out the journey ahead of time. Just carefully note steps that you are talking along the way as well as any questions you might have running through your head, or any thoughts of frustration, delight, etc. I put those in brackets below. The journey might look something like this:

• I have a song in my head (Let me find it in Spotify)
• I open the app, and search the artist or album (I don’t remember the name of the song)
• The search suggestions bring the exact album I was looking for (That’s cool. Probably Spotify remembers what I searched or played before)
• I scan the song list (I think I found the right song)
• I play it (It is the one! I am happy)
• Song is done playing (I want to hear more songs like that. How do I do it?)
• I look around the interface (Looking for something like Explore…)
• I don’t see anything in the sidebar (Hm…)
• I go to the main page and see Made for you (That looks promising!)
• I open the playlist (I don’t recognize most of the songs here)
• I play some (I don’t really like them)
• What’s next?

You can see that some steps might not be strictly relevant to the problem at hand. You can also see that some steps already suggest a solution (adding an Explore link to the sidebar). It would solve the problem of discovering the playlist, but the content of the playlist is not satisfactory, so merely adding a link won’t really address the problem.

In this particular example, I’d have to think of the ways to make better suggestions. Some might rely on modifying a UI. For example, prompting the user to provide feedback on the playlist, or something like that. Some might be about figuring out how to use song and search history to improve suggestions. It might get fairly technical, and I might not be able to offer any real solutions without consulting with Spotify engineers to see what kind of data they have and what is possible to add. But remember this is not a real project. It is OK to assume things, and propose solutions that might not be viable or feasible to implement in a real product.

I hope this makes sense. I am trying to make a point here. As a UX designer you have to be comfortable with thinking beyond the UI. It might be that the solution to the problem will call for minimal changes to the UI, or no changes at all.

For your homework, I’d like you to create a user journey map for your project. You can do one for yourself, and one for a user who might have different needs, or is using the same application differently.

The example I gave was about improving an existing app. You can apply the same process to a brand new product too. But instead of mapping a journey through the existing product, you will have to look at how the problem is being solved now. Chances are this journey will involve a number of products, or no products at all. But the same line of thinking applies.

← Previous – TOC – Next →