Often, engagements inherit an existing site – along with all of its baggage. Here are some of the most common problems I’ve found.
Many design engagements are to “fix” an existing site
Very rarely do we design with a clean slate. If we had the luxury of doing it over – and doing it right – then the task of a deep structural re-design would still be substantial, but it would be simpler. Nonetheless, some usability problems are superficial “coat of paint” issues, which can be addressed fairly easily.
The first step in the process is often to make a preliminary site evaluation.
This analysis is often based on heuristic analysis, which offers a fast track to insight.
the Usual Suspects
Here is some of the low-hanging fruit.
No Summary Overview
- You are allowed (or required) to work with a broad range of attributes but are given no concise overview.
- You have to scroll through a looooong page in order to find out what’s there.
- There are no references, no clues and no context.
There are several versions of the same feature.
- They all looked similar
- They all performed variations on the same function
- But they each had different titles.
- Windows proliferate like bunnies
- You see the “same” page at different places in the service – under a different name – but it operates differently in each case.
- Action buttons have different names and appear in different positions from page to page.
- Terminology varies across the site.
- Maintain consistent terminology between control label (buttons & hyperlinks) and the title of the page it invokes. The labels on buttons and hyperlinks are often contradictory, redundant, inconsistent and confusing.
- Some choices have literally hundreds of items to choose from. But there are no filtering tools to assist in making the choice.
- You have a slew of parsing tools cluttering up the screen – but only a few items in the list.
- A well-designed system does not present you with an unmanageable number of choices. How often do you get beyond the first 100 items in a list? (Yes, it can happen – But that is for the most part a special case.)
- The navigation structure imposes unproductive pages and extra mouseclicks on vital processes. I’m always stunned when I encounter an index list with only one item.
- The website structure may reflect a logical hierarchy that is inappropriate or irrelevant. It’s not that it doesn’t make some theoretical sense, but it doesn’t reflect common sense.
The legacy site maintains totally different – yet functionally incompatible – versions of a feature. This meant that you could perform certain functions on one, but not on the other.
- different management tools
- totally different behavior
- totally different page layout & design
Many of the pages in the legacy site are simply badly designed. Some of the most common offenses:
- Excessive white space in the upper regions of the page wastes valuable presentation area
- The focal information area is crammed to the bottom and overflows the page boundaries
- Disjointed layout of data boxes
- Distracting use of color (colorful boxes draw attention away from the focal table of information)
- Inappropriate use of color emphasizes information that is not really important
- There is often no way to tell what to do next – Provide a “look ahead” and employ reasonable assumptions about the next step.
- A blank slate is rarely helpful – and is often a real challenge. Fill in the blanks with some of those reasonable assumptions: Even if you guess wrong 60% of the time, you’ve still made life easier for 40% of your users (and the 60% have some idea of what to do next).
- Indicate where relevant tools / information / assistance might be. No need to clutter up the UI with unasked for “help”, but if I need it, I should be able to invoke it quickly and easily.
- Reduce the visual dominance of the menu areas relative to content area. Your focus should be on the content materials.
- Every page should have a title that confirms what it is.
- Every page should have visual “design clues” that let you know where you are. Employ appropriate shortcuts.
- Break the items in the Navigation menu into sensible and manageable groups. These do not need to be labeled, but should be at the very least separated.
- The organizational context of the selected document is often lost when you click on a document within a MegaProcess.
Vaughan’s Law: Anything – anything at all – can be done badly
Vaughan’s Law is not a challenge to do things badly. It is cautionary. Sometimes you’ll be told that ‘such and so doesn’t work because…’ followed by an example. It’s worth remembering that ‘even good things can be done wrong.’
Scale, Scope, Agenda
Based on our initial evaluation of the legacy website, we already have a thumbnail impression of what’s good, what’s not and what’s missing – as well as a sense of what we need to do about each.
What’s good : Keep it, though possibly rearrange it
What’s not : Fix it
What’s missing : Find it, add it, create it The “Low Hanging Fruit”
Legacy Issues tend to fall in the category What’s Not : i.e. Relatively obvious stuff which might be described as best practices, standards, or … common sense. But that’s another thread.
The probability of being watched is directly proportional to the stupidity of your act. — the Law of Probability
© The Communication Studio LLC