The First Law of Usability: Your design will be tested by users — your only choice is whether to run the test yourself before launch so that you can fix the inevitable problems while it’s cheap instead of playing expensive catch-up later.
– Jakob Nielsen
By now you have an implemented interactive prototype. It’s time to run evaluations on this prototype. Evaluations come in two forms:
Heuristic Evaluation: Heuristic evaluations are based on previously agreed rules of thumb.
User Evaluation (User Testing): User evaluations are based on usability tests, heuristics, and user requirements.
This article is aimed at Heuristic Evaluation. This evaluation is necessary before user testing to knock out the big and obvious problems with the interface. User testing is useless if you already know that your design has issues.
Heuristic evaluations are one of the most used methods of evaluation in the early stages of design. Rolf Molich defines Heuristic evaluation as, “A usability inspection in which one or more evaluators compare an interactive system to a previously agreed list of heuristics and identify where the interactive system does not follow those heuristics.”
With heuristic evaluation, we do not aim to perfect the design and interface altogether, but find big interface issues before user testing.
Jacob Nielsen and Rolf Molich
developed a set of usability heuristics in 1990, and Nielsen released them in his book Usability Engineering in 1994.
Following are the 10 Commandments as developed by Nielsen and Molich:
VISIBILITY OF SYSTEM STATUS: Can the users always tell what is going on with the system? This can be achieved with appropriate feedback in a reasonable time.
MATCH BETWEEN SYSTEM AND REAL WORLD: Are the
messages displayed in a way that users understand rather than what developers speak? Does the information appear in the natural order?
AESTHETIC AND MINIMALIST DESIGN: Does the design allow users to easily
browse through the information, or do the users have to move around complicated design?
USER CONTROL AND FREEDOM: Do users feel that they are in control while operating through the interface? Is it easy for them to back out if they make a mistake?
CONSISTENCY AND STANDARDS: Are inter-design and intra-design consistency maintained? Does the design interface match the industry standards?
RECOGNITION RATHER THAN RECALL: Do users have to remember things? Or does the design present them with information whenever and wherever it’s needed?
FLEXIBILITY AND EFFICIENCY OF USE: Are
customizations and shortcuts available for expert users?
ERROR PREVENTION: Is there a confirmation option before a user takes an action that might lose his/her data? Is it possible to make it absolutely impossible for users to get into trouble?
HELP USERS RECOGNIZE, DIAGNOSE, AND RECOVER FROM ERRORS: Do error messages precisely indicate the problem, use plain language and constructively suggest the solution
HELP AND DOCUMENTATION: Is documentation easy to
search, task-focused, and brief?
HOW TO APPLY HEURISTIC EVALUATION TO YOUR DESIGN?
Choose a part of your interface.
Walkthrough it with fresh (evaluator’s) eyes.
Apply heuristics to each step.
If you find a potential issue: Note down the issue, its
location, and consequence. Move on.
As per Jakob Nielsen and Rolf Molich heuristic evaluation is to be done by at least 5 evaluators. They base this on the observation that 1 individual can find up to 1/3rd of potential issues, but a group of 5 people can find almost 3/4th of potential issues.
Remember: Your main goal is to identify potential issues during evaluation. Not to fix them. Fixing the issues comes after the evaluation is completed.
Another set of heuristics are the cognitive engineering principles developed by Jill
Gerhardt-Powals for better human-computer interfaces. In her 1996 paper, Gerhardt Powals described cognitive
engineering of an interface as,
“Cognitive engineering of a human-computer interface is the leveraging of empirical findings from the cognitive
sciences and application of those findings to the design of the interface.”
The 10 cognitive design principles, or heuristics, developed by Gerhardt Powals are:
AUTOMATE UNWANTED WORKLOAD: Eliminate mental calculations, estimations,
comparisons, and any unnecessary thinking, to free cognitive resources for high-level tasks.
REDUCE UNCERTAINTY: Display data in a manner that is clear and obvious to
reduce decision time and error.
FUSE DATA: Bring together lower-level data into a higher level summation to
reduce cognitive load.
PRESENT NEW INFORMATION WITH MEANINGFUL AIDS TO INTERPRETATION: New information should be presented within familiar frameworks (e.g., schemas, metaphors, everyday terms) so that information is easier to absorb.
USE NAMES THAT ARE CONCEPTUALLY RELATED TO THE FUNCTION: Display names and
labels should be context-dependent, which will improve recall and recognition.
GROUP DATA CONSISTENTLY AND IN MEANINGFUL WAYS: Within a screen, data should be
logically grouped; across screens, it should be consistently grouped. This will decrease the information search
LIMIT DATA-DRIVEN TASKS: Use color and graphics, for example, to reduce the time spent assimilating raw data.
INCLUDE IN THE DISPLAYS ONLY THAT INFORMATION NEEDED BY THE OPERATOR AT A GIVEN TIME: Exclude extraneous information that is not relevant to current tasks so that the user can focus attention on critical data.
PROVIDE MULTIPLE CODING OF DATA: The system should provide data in varying formats and/or levels of detail in order to promote cognitive flexibility and satisfy user preferences.
PRACTICE JUDICIOUS REDUNDANCY: Principle 10 was developed to resolve the possible conflict between Principles 6 and 8. In order to be consistent, it is sometimes necessary to include more information that may be needed at a given point.
In the 1996 paper Gerhardt-Powals concluded that the lack of applying these design principles results in poor performance; whereas, the application of these cognitive-design principles will enhance human-computer performance.
Heuristic evaluation prior to user testing will help in identifying major issues with the interface and reduce the severity of issues discovered by users.
The next step in the process is Usability Testing.
This article belongs to a series of articles on UX design. Read other Art Attackk blog’s which will help a lot.