UX Desire-Lines – Find ‘em & Fix ‘em
Friday, April 27th, 2012
UXer : Nice picture of a park, right?
Person : Umm… that park is jacked up. The grass and hedges are all trampled…
UXer : You see people ruining perfectly good landscaping… I see patterns, in human behavior and desire.
Take a closer look:

We’ve all seen this and at one point or another use these desire-lines as a routine part of our commute.
As long as humans have roamed the earth, we’ve carved these paths… “I’m here and I need or want to go there.” If you and others repeat the pattern enough times, a path is forged.
In photo overlays, the red paths are high through-put pathways and the white overlays are low-flow alternate paths. In the landscape, we’re able to see these marks because the trends trample down plants and compact the soil to a point where things can’t grow.
Once the trends emerge, we can get a sense of what the original design deficiencies were and, as designers, set out planning design solutions to meet the expressed needs.
The interesting thing about human behavior is that, while we all have varying degrees of tolerance in design or system deficiencies, we all modify our behavior to make up for system deficiencies in the name of getting things done. Often times, these behavioral changes take place subtly and over longer periods of time such that the users themselves aren’t even aware they have bent themselves around a broken system.
It didn’t take much to see these areas and make some hardscaping recommendations, where the evidence is tangible. But as a system designer, a user interface designer… a UXer, how do you “see” these desire-lines in the digital landscape?
You have a number of alternative methods depending on available data and resources. I’m going to cover the following:
- Usage Metrics
- Usability Testing
- Contextual Research
1) Usage Metrics

2)Usability Testing
In any user dependent or focussed system, we say always say “test early, test often”. 5 users can demonstrate 85% of the problems in an interface, given that the probability a user would encounter a problem is about 31%. Formal usability tests are great for spotting gross system deficiencies, and gathering both quantitative metrics on how a user performs in the system (time to completion, number of errors, task completed, etc) and qualitative metrics on how the system performs in the perception of the end user (subjective satisfaction, ease of use, etc).
The pre-requisite is knowing what the usability goals are, which parts of the system to test (tasks, workflows, features, designs) and how to recruit for the test. Usually a good look at the usage metrics should help point to areas that need testing.
For recruiting, you need to screen appropriately to avoid heavy biases that can skew the study results. This takes some level of understanding about the users and behaviors that trend moderate-to-center of the behavioral dimensions you want to target.
You’ll see which specific design or workflow is tripping up or getting in the way of users task or goal completion. But unless your moderator is asking the right questions, “Why did you do that?”, “What were you trying to accomplish by doing that?”, “What did you expect to happen when you did that?”, “How would you change this design to better meet your need?” It’s difficult to get a glimpse into the human intent behind the usability problem.
Going back to the landscaping image, the “user” intent is quite clear… ”I’m here and I want to go there” – but in the context of system usage, often times the interface is such a complex labyrinth to task completion that it’s difficult to understand all the subtle nuances that impact our behavior. This is particularly relevant point because most often, usability studies are performed in a lab setting and or the user is removed from their natural context of use. This brings me to Contextual Research.
3) Contextual Research

In the past 10yrs we’ve conducted a good deal of these studies and the results are always extremely insightful, not only for the system stakeholders, but often times for the subject who realizes how contorted their processes have become to make up for system deficiencies.
If you could draw marks in time and space, the general trend lines go from in-system workflows to jumping in and out of other systems, documents, emails, notes on paper, phone calls, IMs and back to the system to complete the any given task. Most times when we ask, “Why are you doing that…” they’re unaware that they’re doing anything out of the ordinary, and once it dawns on them, they realize… “Well… I need this piece of information to complete this task.”
BOOM!
This is a perfect example of a user jumping off and blazing their own to make up for gaps in the system design. These contextual inquiries have been priceless engagements for our clients who not only discover opportunities for enhancements they were never aware of, but often times entirely new interaction models that had never been considered that bring widespread organizational efficiencies that can save millions of dollars.
Next time you find yourself walking along one of these…
Think about what other things you do (in your job, online, or at home) interacting with any technology that feel “off the designed path” and know that you’re probably not alone… that somewhere in time and space is a beaten path waiting to be discovered and designed for.
I once heard Steve Smith (@orderedlist) say:
“Design is: Intentionally making things Amazing!” I totally agree, but you need to know the problem space first.
By Jon Fukuda
Quick update: I just found a nice related post by Bella on www.uxgroundswell.com







