Interface Design is a Life and Death Matter

Book Review: Mica R. Endsley - Designing for Situation Awareness: An Approach to User-Centered Design

The correspondence between this work and that of Alan Cooper (The Inmates Are Running the Asylum: Why High Tech Products Drive Us Crazy and How to Restore the Sanity and About Face 3: The Essentials of Interaction Design) are staggering. Dr. Endsley and her co-writers rail about how a technology focus leads to poor user interfaces, and recommend a user-centered design approach including "Goal-directed Task Analysis." Alan Cooper rails about how a technology focus leads to poor user interfaces and recommends a user-focused approach he calls "Goal-Directed Design." The difference between this book and Cooper's books is the wealth of dramatic examples and underlying research in Designing for Situation Awareness.

Situation Awareness refers to the OO of the OODA loop - getting input from the environment - Observing - and understanding the significance of that input now and in the future - Orienting. It is a concept used widely in aviation, medicine, and the military - areas where life and death decisions are routinely made based on situation awareness.

One study cited in this book identifies flawed situation awareness as the root cause of 88% of aircraft crashes due to human error. In the remaining 12% the wrong decision was made or there was a problem with execution. With these sobering figures, this book lays out design guidelines to enhance situation awareness.

A formal situation awareness design approach would involve realistic prototyping and rigorous testing as you'd expect for anything related to aviation or medicine. This book provides 50 concrete design principles in six different areas to assist this formal design cycle, but as the book says: "These principles can be applied to a wide range of systems from a variety of domains where achieving and maintaining SA [Situation Awareness] is challenging."

Anyone designing interfaces to support situation awareness or quick comprehension - like performance dashboards - can learn from this book. Unlike software design examples, the examples in this book contain flight numbers and phrases like "killing all aboard" that underscore how very critical situation awareness is, and how driven the authors are to help raise the standards of design.

The only minor criticism I can level is a feeling that this book was rushed together; but with the critical importance of the topic I can see why. I look forward to the recently announced second edition.

Organizational Measurement is Hard

Book Review: Robert D. Austin - Measuring and Managing Performance in Organizations

This book is filled with both humorous and chilling examples of measurement dysfunction that make the sometimes academic approach quite palatable. Dr. Austin identifies three different types of performance measurement based on the intent of the measurement - measurement for motivation, process improvement, or process coordination. It is measurement for motivation that causes the dysfunction that this book so convincingly describes.

For example, if we record the fact that 10 widgets are produced on machine A and we are comparing this against the 10 widget benchmark for bonuses, it is very likely that other perspectives like quality will suffer in the drive to make the 10 widget goal. Austin makes the point that the discovery that every time our overall performance is excellent we have produced 10 widgets does not imply that producing 10 widgets will guarantee excellent overall performance.

If we record the fact that 10 widgets are produced on machine A while only eight widgets are produced in the same time using competing technology on machine B, this is measurement for process improvement and can be very useful - provided it is limited in scope and used purely for the stated purpose.

If we record the fact that 10 widgets are produced on machine A and convey this information to the widget packaging department to ensure that enough widget cases are ready, this is measurement for process coordination, and is also potentially useful on its own.

The idea that the intent or goal of the measurement is of paramount importance is one important lesson from this book.

Austin does make some recommendations about developing effective performance measurement systems.

Understanding the costs involved with "perfect" measurements is part of the solution. Substituting a cheaper approximation for a key measurement is bound to cause problems - witness the measurement of nitrogen instead of protein in wheat gluten used in pet food. The incorrect justification for cheaper approaches is a thread surfaces in other areas - reusing financial figures as a proxy for management accounting leads to flawed descision-making emphasizing short term financial gain - reuse of software components leads to products that are hard to use.

One effective technique is using the end customer as the ultimate judge of quality and performance - the kind of approach described 10 years later in Fred Reichheld's The Ultimate Question: Driving Good Profits and True Growth