Experience Obsolescence
Jim Lentz and Mark Marrara
Modern technology changes quickly. It doesn’t just advance. It rushes ahead of us. It forces us to constantly adapt to the shape-shifting stuff we make for ourselves.
Technophiles find this state of affairs to be exciting and overwhelmingly positive. However, others often find it less than desirable and display a range of responses - from mockery to outright rejection of the latest gadget or app trends. For many of us “progress” has limitations.
More than 50 years ago, the futurist Alvin Toffler coined the term “future shock”, in the eponymous title of his best selling 1970 book. He defined it as a stressful state experienced by individuals and societies when people perceive there has been “too much change in too short a period of time” [1]. He wrote about the cumulative effects of many such changes in a short period of time. We want to examine individual human responses to ongoing technology changes observed “mid-stream” and up close. Understanding our reaction to these changes should help to develop insights for human interaction design.
Adaptation to change requires both the ability to embrace the new and let go of the old. By better understanding adaptation we may be able to control and manage change in order to enhance technological experiences.
“Experience” and “obsolescence”
In this article, we often use the terms “experience” and “experience of technology” rather than “user experience” to emphasize that the phenomena described here refers to more than just people interacting with products. To be sure, the things that we use include digital and material tools and devices. It includes acquired behavior patterns (e.g. blogging), group activities (e.g. barn raisings, e-meetings, flash-mobbing) and social structures such as organizations, rules and policies. Technology has downstream influences on our behavior that are more or lesss decoupled from the direct interactions with user interfaces.
“Obsolescence” is a process in which things, tools, devices, and social constructs are rendered more or less useless and unavailable by changes in technology. Because this article is about the human side of the technology-human experience relation, we focus on the experience of obsolescence. When obsoleted things become less useful and available, human experiences are affected. Rendering a device obsolete also renders our experience of it obsolete and our experience is what really matters.
Technology obsolescence has knock on effects that go beyond the changes in user experience that are bound up in direct interactions with devices and software. The invention of writing changed and devalued the experience of remembering that was associated with song, poetry, and mnemonic tricks. Digital writing technology has largely obsolesced calligraphy and cursive writing skills and even the ability of young people to read cursive script. When newer media formats replaced large vinyl disks the experiences of enjoying album cover art and inner liner bonus ephemera went away. Technology experience obsolescence has wide ranging effects.
Drivers of experience change
We often think of technology as the internal mechanisms that make things work —the physical components, materials, gears, batteries, electrical circuits and software. It also includes external features such as the human input and output interfaces, their appearance and the tasks they help us with.
Engineers change the internal aspects of technology for many reasons. These include improved reliability, durability and production costs. Changes to internal technology sometimes bleed through to impact human experiences regardless of design intentions. Tool handles once made of wood are now usually made of plastic. This primarily reduces manufacturing costs. In addition, changes like this may improve or detract from the experience of use — the tool may be lighter, more or less durable, stronger or weaker.
Engineers, of course, also intentionally change technology externals to modify human experience. As the pace of innovation has quickened, the power to define these externalities has been delegated to designers in various subspecialitiations — “visual”, “interaction”, “information” or most generally, “experience”. Most design work deals with updating existing products so in a very real sense, experience designers are responsible for managing experience obsolescence.
Positive experience obsolescence
Belief in technological progress assumes that technological change is for the better. Old technologies produced experiences that were tedious, effortful, boring or error prone. In modern society we no longer fetch water in wooden buckets from a well. We have indoor plumbing. It is difficult to argue that turning on the kitchen faucet is in anyway inferior to trudging outside to fetch a heavy bucket of water. We don’t need to go to the mailbox or telegraph office to send a message to a friend. We just pick up our phones and call or text. There are innumerable obsolete experiences like these that no one misses.
Negative aspects of experience obsolescence
However, like most things in life, UI changes may also have negative consequences for users. If you doubt this, just google “unnecessary UI change”.
The most common negative consequences involve the experiences of learning and relearning tasks. Learning to use a new device or app takes time, burdensome mental effort and delays its intended use. Moreover, when people interact with a system repeatedly they learn to become more efficient. When a learned task is modified or made partially obsolete by a design change, prior learning and efficient interactions are lost. This adjustment in user behavior is cognitively expensive and therefore annoying.
This aggravation is directly proportional to the frequency with which the changing task is performed. Frequent interactions become more automatic, requiring minimal attention and thought. This automaticity, better known as “skill”, is valuable because not only produces better results but it also frees the mind to concentrate on other things. Keyboard typing is an example of a one such skill. People who are highly skilled at typing can do so without looking at the keyboard, listen to music, and even follow conversations without interrupting their work. They acquired the skill through considerable practice and it now requires little conscious involvement.
Changing the machine interface for an automatic task like typing is disruptive. In the early days of computing, alternatives to the QWERTY keyboard were proposed. These included the efficient Dvorak keyboard and the easy to learn alphabetic keyboard. They all failed because they disrupted to skilled QWERTY typing.
Even the simple, mundane task of answering the phone can be disrupted. We do it reflexively with no more cognitive involvement than expressing intention to respond. But those of us old enough to remember old telephones with separate receivers experienced initially fumbling with wireless and mobile phones to find their answer button. The lifting of a receiver to the ear had been automatic. It was made obsolete by portable phones in which it was necessary to learn to lift the phone with one hand, find and press a button with a finger of the other hand and then move the phone to the ear. Of course, this too was learned but it took a few times to become automatic.
Driving a car is another example. Prior to the invention of automatic transmissions, learning to shift a manual transmission was required. This is a requires the coordination of multiple actions involving two feet moving at different rates while moving the shifter. If you learned to do this yourself or better yet, had the experience of teaching another to do it, you know how difficult and attention demanding it is. Yet with practice, it becomes automatic.
Users sometimes react strongly to these experiential changes. Roxanne Aberchrombie [5] referred to this as “change aversion”. When a “new and improved” technology disrupts the learnings and adaptations users have made to the prior technology, they often resist the change. The persistence of the love some have for manual automobile transmissions is maintained by change aversion as well as emotional attachment to the obsolete shifting experience.
Mitigating the negative effects of change
Fortunately, we have techniques to deal with the some of the unpleasantness of experience obsolescence and change aversion. These include knowing when change aversion is likely, knowing when an experience should be changed and knowing how to design those changes that are worth making.
Identifying tasks susceptible to change aversion
Change can be unpleasant when people need to relearn how to reach their goals with a new technology. But not all tasks benefit from learning and acquired skills. The impact of change is felt less for rarely performed tasks that don’t require investment in learning. The perceived effort required to focus attention when performing these tasks may be forgotten. Furthermore, changes in them are unlikely to be noticed because there was so little learning involved in rare prior uses. And finally, if designed properly, infrequent tasks can be designed to direct the user and thus reduce the cognitive effort required.
Because one-time and rarely performed tasks are relatively immune to change aversion they afford the designer with considerable freedom to make improvements. These can take the form of reduction in the number of steps to be performed or the reduction in effort required to perform steps.
Identifying and handling useful and necessary changes
More care is required when redesigning interfaces for repetitive tasks. First, it is important to consider whether and which experiential changes are warranted. This should always be framed in terms of benefits and costs for users. Will there be sufficient user benefit to outweigh necessary relearning costs? Are there areas of the experience that users are perfectly satisfied with?
Two operations of user interfaces are highly susceptible to change aversion — navigation and identification. Experienced users hate it when they can no longer find a function because someone has decided that it makes more sense to put it in a different area of the user interface or label it differently. Apple did this in IOS 14 when they moved the “Bedtime” feature from the clock app to the health app [3]. Unsurprisingly, many users (including this one) believed the function had been eliminated when it had merely been hidden.
True, designers don’t always have the luxury of avoiding risky experience changes. Sometimes this is out of their control. These may be forced by the underlying technology, the integration of multiple products into a single coherent platform, or the whims of product owners. In these cases, they should mitigate experience obsolescence impacts in the best possible ways for their users. Designers should thus consider how existing users will respond to change, address how they will perceive and understand the changes and ensure that novel aspects will be readily learned in order to limit unpleasant experiences.
Of course many changes can be beneficial for user experiences. Some existing experience designs leave room for optimization. These cases provide potential benefits that outweigh the cost of user adaptation. Note however, that they don’t eliminate the need for the designer to address change experience itself. While there are many design improvements that are instantly recognized by users as beneficial there are also many in which users don’t immediately or even ever see the value in the change.
Some experiences can be improved directly with little relearning required. These include changes in which the designer can reduce user effort in tasks that require multiple and/or cognitively demanding steps by automating them [4]. Still, even in these cases users may be disrupted if they recall obsolete steps but don’t see that the system is now performing them automatically.
Finally, some redesigns that improve status and feedback information can improve experiences without imposing significant relearning costs.
Product releases provide the opportunity to make many changes at once. It is important to think about those changes in aggregate as well as individual ones. Not everything should be changed. Things that work well for users should be preserved. Things that don’t work so well should be candidates for change. Cost-benefit tradeoffs should be used to prioritize changes. Lastly, when considering a product in its entirety, there can be a limit to how many changes, even positive ones, should be made in a single release. Sometimes improvements are best when delivered incrementally.
Cost and benefit data used to prioritize improvements can be gleaned from traditional user research, customer support reports, product forums and discussion groups.
To sum up, designers should limit change aversion by reducing the amount of egregious change in a product. We all must continually learn and adapt to changing interfaces in our daily lives, and in many cases, the changes provide little value. This is typically because a product development team neglected to consider potential impacts of technology changes from the user’s perspective.
Dealing with the most disruptive product changes
The most complex experiential changes are those that force a change in the user’s mental model so that it no longer provides a frame of reference for understanding what to do. These kinds of changes are by far the most disruptive and challenging to design. They can be the béte noir of experience obsolescence.
Not to sound like a broken record[5] but again, ensure that benefits of change are sufficient to outweigh costs. Reforming the users’ mental model is expensive and difficult, therefore the benefits to the user must be large.
In these highly disruptive cases, changes in Information Architecture (how the product is organized and structured), navigation, terminology, and UI model disrupt a user’s existing mental model of the product. This will inevitably mean heavy reliance on user assistance and documentation — something we are going to disparage in the next section. It will be necessary to be up front with users about the extent of change and to provide encouragement that it will all be worth the transition in the end. It might be useful to compare and contrast metaphors in the old and new versions. When all else fails it might be necessary to provide support for both a “classic” and “new and improved” versions to avert wholesale user abandonment of the application and defection to competitor products. Finally, common functions (e.g. Save, Save As, Copy/Paste, Open, Rename, etc.) should be preserved from the old experience and labelled with same words or icons to provide islands of experience stability and familiarity.
Avoid design arrogance and user assistance mitigation traps
Designers sometimes make changes for the sake of change itself or to demonstrate their skill as designers. This is fine as long as it doesn’t fail the end-user cost/benefit test. No one wants to kill an opportunity for a good designer to grow her career by producing noteworthy designs. However, the design field needs to above all honor the primacy of the user experience. A truly good designer will improve user experiences while considering the impact of any changes made.
Sometimes designers assume that any negative consequences of change can be nullified by cleverly integrating user assistance. Rarely if ever is this the case. User assistance should not be viewed as a get-out-of-jail-free card for design hubris. It is yet another thing a user must attend to and therefore is a detour between the user’s intent and end goals. When a person must look up a topic in user assistance, that time could have been better spent productively using the product.
User assistance varies in size and complexity. Bulky documentation in the form of education, release notes, product welcome or introduction materials are the most burdensome. They are usually added toward the end of development cycles as afterthoughts. No one enjoys reading this stuff and it is often ignored. Even when read, it is often quickly forgotten.
Lightweight assistance such as contextual help and walk thru UI tours are better but are still costs in the cost/benefit equation. Contextual help comes into play only when the user has been defeated in his attempt to understand the user interface. It does not eliminate the cost of unintuitive design. Walk-through tours are better but they also delay and divert the user’s attention from the task at hand. When given a choice, users will often decline such an experience in favor of attempting a task directly or exploring the user interface under their own control. The best improvements are readily apparent without requiring attentional detours.
Conclusions
Designing a new release of a product poses the challenges of coming up with a good design and then ensuring that it is implemented as designed. Surmounting these challenges is no small order. So it is understandable that designers sometimes fail to consider the effects of their change on users. But the growing burden experience obsolescence on users demands more attention.
Change is inevitable but can be managed. It requires proper cost/benefit tradeoff analysis, management of experiential change, and eschewing dependence on user assistance.
It requires knowing how to address different kinds of task modifications. Changes to tasks that required implicit or explicit learning are more risky and difficult to make. It helps to know what to keep and what can be modified. Changes that reduce mental effort and steps in a task are usually safe and welcomed by users. Those that impact UI navigation and operation identification are dangerous.
Designers should consider islands of experiential stability as critical to managing change. This should include familiar terminology, organization and iconography so that users can locate and identify task interfaces.
When changes are large, change mitigation efforts need to be equally large. They will require the tall order of designing learning experiences that engage users as they adapt. This in turn will will require a solid human-centric design process for the bridge to the improved experience.
Above all, designers should question the need to continually “refresh” experiences with the possible exception of some aspects of visual design. It is possible to fulfill the designer’s hope that they are making things better for users. It simply requires paying attention to what should change, what should be kept and how the transition will be experienced.
All of this requires that designers have a solid understanding of the strengths and weaknesses of the experiences within their product. Good redesign means shifting toward the strengths. If changes don’t do this, designers should raise the alarm and push back. Hardly anyone wants change for the sake of change.
Notes
[1] Toffler, Alan. Future Shock 1970
[2] Roxanne Aberchrombie, Change Aversion And The Conflicted User, Usability Geek, 3/15/2020 https://medium.com/usabilitygeek/change-aversion-and-the-conflicted-user-17f6b43e0cb4
[3]RepublicWorld.com. What happened to bedtime in iOS 14? How to set a sleep schedule in iOS 14. 9/21/2020 https://www.republicworld.com/technology-news/mobile/what-happened-to-bedtime-in-ios-14.html
[4] automation: Automation is sometimes assumed to suggest artificial intelligence. However, often a simple calculation or operation performed on behalf of the user will suffice. Presumably you arrived at this footnote by scrolling to the end of the article. This annoyance could have been automated by a simple hyperlink and return-link had Medium provided such a feature.
[5] broken record: A record is an obsolete form of physical media used to store music or the spoken word that existed prior to compact disks, MP3s, Spotify and Pandora. An early form of records were called “78s” because they rotated(!) at 78 RPM. 78s were made from an obsolete material — brittle shellac based resin. As a consequence, 78s often broke. In turn, 78s were made obsolete by 33s and 45s which were made from a plastic called vinyl (33s and 45s have recently been de-obsoleted by certain demographics for reasons of irony alleged sound quality). Regardless of rotation speed and construction materials, all of these media suffered from a mild form of breakage or glitch would cause the playback to repeat incessentally.
