Look and Feel

No - this is not an article about stalking or sexual harassment. It's about Web GUI interfaces.

Here is one of our Web interfaces. It has been operating for about 25 years. It has been and continues to be effective at accomplishing its objectives - being a high density information site where we can refer people for content and make it widely available. Among the comments we get are that it looks like it is from the 1980s (before there was a Web) and that we should update it to make it look more modern.

Look and feel is important because it effects cognitive processes in the reader/user. Cognitive dissonance will create an inability to grok the content, while cognitive resonance will tend to cause the user to believe the content and wish to use the site. But the real question for the company is:

What am I trying to achieve?

Yes, that's right, before you consider the look and feel of your Web site, you should start with a business objective in mind. If you are trying to close sales for a small number of low-priced items in high volume to consumers worldwide, you should likely take a different approach than if you are providing B2B back-end technology support functions to a small number of large corporate customers in Western Europe.

The look and feel of a Web site is, in many ways, like the look and feel of a commercial. For those who have watched commercials over many years, you might recognize or remember things like the "Popeel Pocket ..." I don't remember the name of the product, but I certainly remember Popeel, and more importantly, those ridiculously amateurish commercials sold lots and lots of stuff. Just look at the home shopping network. It has none of the production values of a lot of commercials or television programs that are widely touted, but it sells tons of stuff all the time.

How do I know I am achieving it?

Open loop systems ultimately fail as/if they do not adequately adapt to the changes in the environment they operate in. That feedback comes in many ways, but for Look and Feel of a Web site, you will never know if you could do better (or worse) unless you test. Testing means making more than one Web site version and trying them out to compare results. But as a side effect, you will create dissonance in your customers who might be sent to one or the other site, unless you do much more expensive market studies like the big corporates do.

I have occasionally tried variations on all.net, but none have produced as much of my desired outcomes as the current look and feel. Nevertheless, I have sought ways to be responsive.

Give the customer what they want

One of the approaches I have taken with my software for a long time is to add options so the customer can set the look and feel they like. Of course once I did this I found out that they don't know what they like, but that's a separate issue. A good example of users not choosing what they like is their default browser font settings. I always let them use their default font, and every once in a while I get a complaint about the font. So I tell them that they are using the font they have set in their Web browser and that is their current preference. At that point, if I'm not careful, I will end up being their support person for browser configuration.

In my current interface users can select the "LookFeel" mode and change settings for font size, face, italics, bold, small caps, and color, background, spacing, padding, corner-radius, and border type, width, and color, and in the advanced mode, do this for each of left, right, top, bottom, and each corner. And for those who love to play, I have the random LookFeel generator that generates a new look and feel every time they click on a button. Of course I also have a set of about 8 standard color sets and total look and feel setups as starting points.

My current policy is that if they complain about look and feel (after they were told in the manual, the video, the live onboarding meeting, the help menu, and the help videos), I will walk them through setting it up however they like it. Historically, when doing the same thing for users before, two results appeared; (1) they chose the default, and (2) they say something to the effect of "Oh..." and make no changes. I suspect they want me to choose for them and have the good taste to give them something they like. That's of course what I do with the default mode, but they don't know they like it till they try out other things and find out that they don't actually care. Having said that, the most popular setting is "Random", for about 20 screens worth, at which point they realize something about...

The cognitive issues with interfaces

My approach to understanding interfaces with people is that the object of the interface is to make the people efficient. That is, spend less time to do what they have to do and make the results as good as they can be. Or in the manner of the cognitive interface:

Good results - no wasted time

The quality of the results stems largely from the availability of just the information they need to do whatever they are doing, presented in a way that makes it easy for them to understand and act upon.

The wasted time is largely spent in having to look for something and having to push numerous buttons before being able to decide and act.


Going to the observe, orient, decide, act (OODA) loop, as the old fighter pilot saying goes, "see it, kill it, get away quickly". My application Web sites are not there for the enjoyment of the experience or the entertainment value. They exist to get things done. The goal is to (1) see what there is to do, (2) do and document it, (3) identify what needs to be done next/later and set it up, and (4) do the next thing or be gone because you are done.

That means the user has to be able to readily see what there is to do, readily be able to act on it, document what was done in the process, and schedule the next thing to do (or if none, be done with it). For example, as I look at my current interface, I can immediately identify that I have 18 more things to get done today (for A2E) as of now. I could tell that in about 1 second by the color of things to do today (orange) and the counts of everything listed, plus the fact that it is currently sorted by when things are scheduled to get done. Each item shows who/what is involved, contact information, what is to be done, and the situation relative to my overall business. It's all on one screen, readily readable, and I can pick off one or more of them to act on individually or together. When I pick one to act on, it shows me the entire history of interactions going from now back in time (I have to scroll to get to earlier things at some point). I can update anything I need to, setup the next interaction expectation, and click to save then return to the now 17 things to do today.

Like driving or flying?

But like flying a plane or driving a car, it takes time to get used to such an interface and become efficient at it. If you only have 3 things to do this week, this is not an efficient way to go about it. If you use it once a month, this is not a process you are likely to become proficient at doing.

My solution is the "Novice" interface mode that provides fewer options so it is less confusing. You can still see the big picture with its colorings and markings to help see things more clearly. But there are less things you can try to do, and thus fewer options for your decision-making. Once set up, you can eliminate confusing columns, change font sizes and separations, and make it look big and seemingly easier to see the things you are looking for. As simple as possible but no simpler (Occam's razor so to speak).

But you still have to learn how to use it. The so-called obvious way things work in most systems today are only obvious because they are the same as a previous analogous system the user has used. But making decisions involving multiple factors and many options is not that those interfaces were made for, so like flying, you need to learn the user interface and then be able to fly with it from now on. Changing the interface will cause increased mistakes, as has been studied extensively in the literature of ergonomics and operations of complex machines.

A call to action

Want to try out these interfaces? When you present at one of our forums, you will have the opportunity to do so, as part of providing metrics for investors:

Go To Angel

Experience the interface, and get help to grow your company.

In summary

Look and Feel - a subtle art - which I have only started to see and touch in this article.

Copyright(c) Fred Cohen, 2020 - All Rights Reserved