Nota: este artigo é o texto original. Caso queira ler este texto em português, acesse a tradução deste artigo.


This time, I have a special article for you, with some great tips for any UX Designer, even if you already are an expert one.

This one is a question and answer article, and the big thing is: the answers were given by Joe Natoli, one of those that I consider one of the best User Experience teachers.

Joe Natoli has launched six successful online courses, with more than 70,000 students.

The website LoudProgrammer, mentioned his course as one of the best User Experience Design courses on Udemy. His courses are listed at the end of this article.

Before you start to read, I’d like to thank Joe Natoli for these so profound answers. I’m sure this will be one of the best articles in this blog. I really appreciate for your time and efforts in bringing us this great content.

So, let’s get it started!

Below, you read the questions I asked him and the answers he provided me.

UX Design Tips by Joe Natoli

Joe Natoli - ISA 2016

Joe Natoli – ISA 2016

1. UX Design is all about understanding the users’ needs. What are the best ways to understand the real needs of users?

Joe Natoli: I think UXers and Designers and Developers of all experience levels wrestle with this almost more than any other issue; it seems to cause people a great deal of undue stress! I say that because not a day goes by that I’m not asked for a single, best-practice process to do this. As such, I want to start by saying this:

There is NO single right way to do user research.

The methods you use can and should change depending on the context of your situation: the industry, the type of product, audience expectations, technology constraints, legal regulations etc. There are a number of variables that affect a user’s experience, and not all of them come from Product Managers, Designers and Developers.

Now, the next part of the answer is that the best way to find out what users expect, want, need and might be willing to use is to observe them interacting with a wireframe prototype or a build. Giving them scenarios, tasks (“how would you…”) and cutting them loose to go do it. Recording that path if possible, asking for commentary from them as they go, without any interruption or interference from you.

What people say they have done or will do is, as we know, almost NEVER what they actually do. So the shortest distance to the truth is seeing that interaction with your own eyes.

That said, there’s a very large number of individuals and teams who don’t have the luxury of user observation. And that’s OK too, because you can absolutely still get close to what’s appropriate and useful for people.

User interviews are great, provided you ask open-ended questions that don’t lead people toward a conclusion. The other key there is that once you ask the question, you let it hang in the air and resist the urge to speak again. Give the person time to work through their answer; let the silence do the heavy lifting.

Desk research — even if it’s nothing more than observing what people complain about most often via social media, or reviewing the most popular competing products to see what they offer and how — is still very, very valuable.

And although I’m not a fan of surveys, if they are well-crafted and open-ended, it’s still possible to glean useful insights.

To sum up, I will say again: there is NO one right way to do this. And research does NOT have to be formal! If you’re digging into why people care about something or what they hate about it, or what obstacles they face — in any way at all — you’re researching.

2. Since a product (mobile app, website etc.) is brand new, it is not always easy to know the profiles of the people who will actually use the product. Many designers try to predict what the profiles are and end up creating personas based on their predictions, which can be dangerous. How can you find out who the real users will be?

Joe Natoli: You definitely have to do some things to get yourself (or your team) into a user’s headspace. There are a few specific exercises I like, because in addition to moving designers away from their preconceptions, they also cut through the fallacy inherent in surveys and interviews.

The two I recommend to teams most often are situation and empathy mapping. They’re quick, simple processes and they do a great job of getting people out of their own heads. I talk about that in detail, along free downloadable templates for doing it, in a blog post here: http://www.givegoodux.com/creating-better-user-personas-tips-templates/

User Personas themselves are always valuable, IF they follow Alan Cooper’s original model. There are a lot of very poor facsimiles of what a persona is and how it should be constructed, but Alan’s version is the truest and most accurate, in my experience.

The one mistake I see most often with personas is that there are too many of them. Alan has said this more than once and my experience proves him right. I’ll explain, but the rule, to me, is this:

The more personas you create, the less likely you are to satisfy any of them.

Cooper’s original prescription for personas basically says that if you are able to design something that delights a single persona, the rest will follow. And as he puts it, “widening your target doesn’t improve your aim.”

So if you have a wide, diverse user audience, the logical thing to do would be to develop a large representative set of personas that mirror this diversity. And then you’d develop a feature/functionality set that accommodates that large set. On paper, it sounds perfectly reasonable.

In reality, it fails terribly. Try to satisfy everyone and you’ll end up satisfying no one. Your design and your product are watered-down, middle-of-the-road, nothing that people can’t get elsewhere. So although you’ve  designed and built something that no one hates, the consequence is that no one loves it either.

What you learn over time is that your primary persona is carrying the same expectations, wants and needs of those you’d identify as secondary or tertiary personas. And if you hit those marks for your primary user, you’ll hit them as well for everyone else.

3. The User Experience Design process is relative to the project and each designer may have distinct processes for building a product. In your opinion, in general terms, what are the key steps in the UX Design process for developing an e-commerce application?

Joe Natoli: Well, attempting to outline an entire process would likely take more space than we have here, but I do think there are three things in particular that are mission-critical when it comes to e-commerce design.

I think the first thing you have to do, the minute you dip your toe into the world of e-commerce, is spend some time talking stock of recent innovations and improvements in ease of use — the ways in which we look for and buy products. And, of course, the place you start is with the undisputed industry leader — amazon.com.

Right now, for example, they apparently decided that one-click ordering wasn’t fast enough (!) — so they came up with this idea of a “Dash” button.

This is a physical button that you can stick on your refrigerator, or in your pantry, laundry room, etc. When you’re running low on a certain product, you just push the corresponding dash button and a refill is on the way.

Now, is this more sizzle than steak? Maybe, and it may not even be necessary. But what matters here is the intent — ease of use. A frictionless user experience, with virtually zero effort on the part of the user. Interaction in context with the user’s scenario, by allowing them to act at the moment of need.

So while your e-commerce site may not need a dash button, the first thing you need to concern yourself is ease of use, removing every possible roadblock to finding, evaluating and purchasing a product.

The second thing that requires relentless focus is product and product category labeling. If I had a dime for every e-commerce search engine I’ve been asked to evaluate where common words and phrases customers use returned a “no items found” result, I’d be a billionaire.

Organizations often default to their own internal terminology, phraseology, labeling and organizational models. And in far too many cases, their buying audience does not use that same language to describe what they’re looking for. What the customer calls a specific product and what they call it internally are often two different things.

Add in the fact that Google has essentially spoiled us all, because it’s insanely good at guessing what we mean when we enter words and phrases. The net result of this is that your product data, your database design and your search functionality must be stellar in terms of flexibility, adaptability and accuracy.

So this starts with a healthy dose of user research, specifically things like tree testing and card sorting, in order to get a handle on what people call things and how they assume categories and subcategories would be linked.

The next thing is to make sure that your database designers and architects are intimately involved and collaborating in your information architecture and taxonomy efforts — because unless you’re on the same page with them, your best laid plans will fall flat on their faces. It always surprises me that in so many teams, designers never talk to the database architects. I can’t think of two groups who need each other more.

The last thing I have to say about e-commerce is this: make checkout insanely easy. Follow conventions. Don’t reinvent this wheel. People’s expectations are formed by all the other things they use, and many of them use amazon. So follow their lead unless you have a spectacularly good reason to deviate.

And for the love of all things good in this world, never require someone to create an account to check out!

4. Perhaps one of the most complex points is measuring the success and failure points of a product in terms of UX Design. There are tools like Google Analytics that help we get key data, but it is not always an easy task. How can you measure the qualities of UX Design of a product effectively?

Joe Natoli: Successful UX, to me, is an experience where both sides of what I call the UX Value Loop are satisfied. That means users get something they find useful, usable and valuable. It also means that because people use that product or service, the business or creator benefits as well; increased internal efficiency or external market share, money made, money saved.

Companies and product teams tend to focus too much on the product itself, and with the metrics typically included with analytics software. The problem there is that this data, for the most part, isn’t very good at telling us whether or not the experience was a good one for someone. For example, you may have a metric that tells you that people are spending an average of 7 minutes on your site. That may mean that they’re engaged and interested — but it can also mean that they’re having a hard time finding what they’re looking for.

The only metric that really matters, for users and for the business, is the fulfillment desired outcome. What did people want or need to happen, and did it actually happen? What did the business want to achieve, and was that metric met or exceeded?

Here’s the thing about that: you can only measure those desired outcomes if you’ve done a good job of defining them in the first place 😉

If you’ve done that, then analytics can certainly tell you a great deal, even if you’re not using Google Analytics. If you make a change to a checkout process, and revenue jumps 150% over the next 8 weeks, or cart abandonment goes down, that’s something you can hang your hat on. If you streamline a self-service app workflow, and help desk calls go down by 43% the first three months after launch, that’s a meaningful outcome.

But again, you have to know what you’re trying to achieve in the first place, and in many cases those goals are poorly defined. “We need to be seen as more current, incorporate best practice” isn’t something you can measure. “We need to get account cancellations down to 4% within the next 3 months” is a measurable goal: either cancellations are going to decrease or they’re not.

So I guess the moral here is that UX can be measured, provided you (and your clients/stakeholders) take the time to define what outcomes you and your users need, in a very specific, measurable way.

5. What are the 5 must-read User Experience Design books in your opinion?

Joe Natoli:

  1. The Design of Everyday Things, Don Norman
  2. The Inmates Are Running the Asylum, Alan Cooper
  3. Universal Principles of Design, William Lidwell, Kritina Holden, Jill Butler
  4. Don’t Make Me Think, Steve Krug
  5. Think First, Joe Natoli (yes, I’m that proud of it)

Joe Natoli’s UX Design courses and books (and others)

If you are interested in learning more UX Design with Joe Natoli, which I strongly recommend, here are his courses:

My UX, UI, and Career Training Courses:
http://learn.givegoodux.com

The Making UX Work Podcast:
http://givegoodux.com/podcast

Tuesdays with Joe Video Show (UX Q&A):
https://www.youtube.com/c/Givegoodux

Think First: My No-Nonsense Approach to Creating Successful Products, Memorable User Experiences + Very Happy Customers (Book)
http://givegoodux.com/think-first