I'm at the beginning of a design effort for a web application that my team is undertaking. I have an older design concept, but really we're looking for some quick information to help drive the design forward. Is there a specific type of user research or testing that should be conducted?

  • Is the design effort to address particular UX concerns that have already been identified, to identify potential issues, or a general redesign effort (because we just redesign things sometimes)? Is the end goal closer to a "complete redesign" or "tweaks to the current experience"? Jun 16, 2014 at 20:11
  • Isn't specifically for concerns that already exist. Its a new screen concept. Due to it's volatile nature, I can't give specifics. I think I'll check out the Lynda resource mentioned below.
    – eeklipzz
    Jun 16, 2014 at 22:24
  • I hope it helped! There are several good, solid UX resources on Lynda if you're just getting started in the field (not uncommon, given how hot of a buzzword it's become this year)
    – Imperative
    Jun 20, 2014 at 22:46
  • ughh... buzzwords. haha. It's funny. I just wrote a blog about buzzwords. uxfindings.blogspot.com/2014/06/drop-buzzwords.html. Thanks again for your answer though Imperative. I have to say that I we have actually recently pulled Lynda.com as a resource into our office. There's SOOO much stuff there.
    – eeklipzz
    Jun 21, 2014 at 4:18

1 Answer 1


If you have a Lynda subscription, I strongly recommend reviewing the course on the Foundations of UX Information Architecture. It will introduce you to the basics of card sorts, then walk you through the details of how to set up + analyze an Optimal Sort online.

The typical UX process can be boiled down into plain English like this:

  1. What am I building? This seems like a question with an obvious answer, but in practice it's like a game of telephone. You need to codify the answer to this question and make sure everyone knows it.
  2. Why am I building it? At this point, I've had newbie clients actually get annoyed so be circumspect in how you ask this one. It's a critical follow up to part one though. "I'm building matchsticks.com in order to help people who need fire discover the best tools for starting one. I'm building this site because so many people believe that an apocalypse is near, and are looking for ways to be prepared. Also, a lot of people like to BBQ at the park during the summer and want to know what the best matches are."
  3. Who am I building it for? Right here? This is where the journey of UX begins. We're going to go on and define all sorts of things from market segments to current beliefs and how we hope to change them. We're going to do a heuristic review, based on this question, in order to learn what our core audience expects from a website about matchsticks. We're going to make a list of all the questions, actions, and objects a person might encouter on a site and we're then going to find the people from this question, and we're going to ask them how they relate to that data.

At this point, you've got a lot of choices opening up. With complex clients, I've been known to break out a copy of Business Model Generation and set up a canvas in order to define key players, the audience, and the value proposition. You'd be amazed how often a small business or a startup has skipped this step. At the very least, you should be familiar with Jesse James Garrett's Elements of User Experience. It's basically a textbook for the UX process.

The simplest approach though, once you have your three core questions satisfactorily answered, would work like this.

  • First, run a card sort and analyze it. Establish an abstract information architecture, then compare it to a minimum viable product (established in the business plan, or in your business plan review).
  • Establish a basic sitemap, based on your abstract IA
  • Do you need just one navigation or is there a case for multiple types? Consider an e-commerce site where the top bar is hierarchical, but the sidebar contains a filter. The sidebar is still a navigation scheme, it's just much better at answering abstract questions like "What should I buy for a 15 year old boy?" than a top navigation; which is good at "Toys -> Action Figures -> X-Men".
  • Set up a Balsamiq project and sketch up several versions of the core pages
  • Present this data to the client
  • Now, using Optimal Workshop, run a "Chalkmark" type analysis. If the intent is to have the customer locate the best match for a particular action. Eg: What type of matchstick should I use in hurricane conditions during a zombie attack? Maybe you show them the landing page sketch. Is their first click correct?
  • And seriously, buy the monthly subscription. You're going to do this several times and $149/each adds up. You can get away with a $149 optimal sort one time, but Chalkmark wants a monthly sub.

At this point you are ready to start either wireframing in something like Axure, or moving into a high fidelity prototype.

There are many other tests you could run, from Conjoint analysis to an a reverse card sort. There are dozens of ways to collect data on your end user from collages to mall intercepts. The main thing you need to know though is: What, Why, Who? / How do they relate to the stuff I am putting on the site? / What is the best way to navigate this? / Does my layout make sense to them?

TL;DR : Buy a subscription to Optimal Workshop

  • Thanks for the information. I'm going to check out the "Foundations for Information Architecture" that you mentioned on Lynda.
    – eeklipzz
    Jun 16, 2014 at 22:25
  • It's a great start. You can read Chapter 2 of Garrett's book here as well: jjg.net/elements/pdf/elements_ch02.pdf
    – Imperative
    Jun 17, 2014 at 0:39
  • This answer is BAD-ASS! and one of the best i have ever seen on here
    – colmcq
    Jun 17, 2014 at 12:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.