We've all been there at some point, frustrated and annoyed, we hear that bubbly voice on the other end of the line, "Well, how's the weather out there, today?" This is the fourth time you've called customer service to return yet another pair of shoes you bought online that doesn't fit, and the last thing you need is a bubbly, exaggeratedly happy voice trying to cheer you up. The customer service reps at Zappos really try to live by the company's mantra of happiness, but sometimes the way they go about it doesn't really jive with your mood.
This is a fundamental problem when trying to design for experience. You can deploy all the design patterns you want, but each one might be perceived or used differently depending on a user's state of mind.
Experience with any product or service can be broken down in this way:
- Perception - What a user senses, and how they actually interpret the events around them.
- Mindset - What the user expects, how they approach solving a problem, and their goals.
- Emotion - This is the essence of experience, and unfortunately, what we can't control. But a user's positive emotional reaction is (usually) our ultimate goal as designers.
As designers, controlling what a user perceives is core to what we do. Visual designers and interaction designers rule here; they deploy principles and patterns to prompt users to action, make affordances evident and intuitive, etc.
But users' mindsets are what we should be paying closer attention to and attempting to prescribe. If a product's behavior doesn't accommodate or cue a user into an appropriate mindset, he or she can get frustrated really quickly. (In fact, as far as interfaces go, one could consider this a definition of frustration.) We've all tried to author important emails only to be interrupted by an ad or app update.
While designers can't control a user's mood, they can direct users into particular mindsets tailored for the goals they have and then design accordingly. So how can we start to describe mindsets, which can provide a crucial link between perception and emotion?
A Mindset Example - Learn by Doing
We've all heard the phrase, "learn by doing." This is a familiar mindset that when phrased this way links goals ("to learn") to methods ("by doing"). Its effectiveness comes from leveraging the power of association. By doing (instead of watching or listening), we start to associate gesture (swipe, tap, click, etc.) to action (follow a link, send an email, etc.).
To turn this into an actionable tactic, however, we first must decide that "learn by doing" is the right mindset to design for. Maybe what we're designing is new enough that we need to associate gesture with actions, but not intuitive enough that standard visual cues do the job for us. The drag-down-to-refresh behavior is something we were all taught by clever tutorials in the first iPhone apps that deployed it.
Once we've picked the right mindset, we can craft an interaction model that helps user adopt it. One technique is to drop a user into a "walk-through tutorial." This interaction pattern prompts users to tap, swipe, pinch, etc. to learn important actions and employs such techniques as lightboxing to encourage them to focus on specific areas. In this sense, we can say that a "walk-through tutorial" pattern embodies a learn-by-doing mindset.
In general, we can evaluate an interaction model by whether it contributes to particular mindset or detracts from it. If it does, we can design in order to telegraph to a user what mindset we're attempting to employ. If we do our job correctly, they'll pick up on it, and thus will (hopefully) have appropriate expectations.
But most important is the fact that whatever mindset we employ, it must help a user achieve his or her goals.
Some Common Mindsets
If we had a catalog of mindsets to play with, derived from our experiences with common products, websites, and apps, we could start understanding, categorizing, and redeploying them. Here's a first stab:
The learning experience is always present in a piece of software; the more complex the software, the more important learning becomes. Some people like to avoid instructional materials (e.g. videos) and attempt to learn a product by actually using it, and others like to study collections of tutorials and documentation:
- Learn-by-doing: Discussed above, this is often represented as a guided tour that leverages the power of association, interactively teaching important features.
- Learn-by-playing: Enable people who like to avoid instructions to play to discover behaviors on their own. This relies heavily on discoverability and convention.
- Learn-by-studying: For complex or mission-critical applications, this involves providing detailed instructional and reference material.
- Learn-by-engaging: Interaction among users (in places like online forums) can often be an effective way of providing help for learning. Stack Overflow serves as a great resource for aspiring programmers. Interacting with people takes a different mindset than interacting with a help system; some users will prefer one over the other.
- Learn-by-observing: This often means watching a presentation or video of someone else operating a piece of software. Can be boring if not well crafted.
- See it once, know it for good: This assumes a user will remember an idiosyncratic feature because of its high impact, importance, and high frequency of use.
Perception and Understanding
Helping users quickly grok a product's conceptual model should be one of the primary goals of experience design. Without visual clarity and relatable terminology, interfaces can easily become cryptic and frustrating. Here are techniques that apps have embodied to help users understand:
- Understand at-a-glance: Like a heads-up-display in a fighter jet, information is laid out to push the boundaries of perception, delivering the most pertinent information with the least perceptual effort. It is found in dashboards, data visualizations, icons, street signs, etc. Such clarity of purpose often makes design invisible.
- One-to-one mapping: This is partly a control layout strategy that provides one control per possible action. Like on an audio mixer board that presents all of its controls at once (nothing hidden behind a panel), we can use Gestalt-based groupings and other visual techniques to establish relationships and hierarchy. Users understand that this satisfies the need for immediate ergonomic access to many interrelated things.
- Increased leverage (or minimal effort, maximal results): Instead of requiring multiple steps or controls to perform a large-scale or complex action, provide just one. Users will be more careful when they perceive the power behind this single action. See Amazon's "Buy now with 1-Click®" feature.
- Domain mimicry: When an interface's conceptual model attempts to mimic the domain language of a user's goals. This mimicry can leverage physical metaphors (e.g. flash cards are drawn as cards that actually flip over) or conceptual metaphors (e.g. software for designing buildings using elements for "walls," "floors," etc., or simply using "folder" instead of "directory").
In the days of incessant digital distraction, websites and apps have an opportunity (and perhaps, a responsibility) to cultivate users' mental focus. But it's not only about optimizing productivity.
- Single task, single focus: Like the original iA Writer app, this removes all distractions and helps focus someone on a single task, ideally with a single interaction model. This often implies a minimal visual design.
- Multiple tasks, single focus: This pattern presumes that while you may be doing several things, you can only really do one of them at a time. Demonstrated by the early iOS, each fullscreen interface focuses on one subtask (compose an email, check emails, change settings, and so on).
- Interrupt to warn, but not to inform: This counters two common patterns: notifications and modal confirmation boxes. Most interfaces that use notifications treat them as if they are inherently unobtrusive, but they really are a form of interruption and distraction. Anything that appears in someone's peripheral vision will still be noticed, interrupting a user's train of thought. As polite as they intend to be, notification systems still cannot judge for themselves which ones are truly important. If focus is indeed an important state of mind, avoid modal dialogs and notifications that require manual dismissal, unless the idea is really to take someone’s attention away to attend to a mission-critical matter, like impending loss of battery power.
- Create another world: Like the embedded mini-site for the Mac Pro on Apple's website, this creates impact by putting the user in a completely different "place." Such an experience provides the opportunity for the rules to change, for a different interaction model, and for thus a different mindset. Instead of enabling casual browsing through many products, here Apple encourages a deep dive into the Mac Pro's inner workings in a sales pitch shrouded as an interactive movie.
Artificial Intelligence and Automation
One promise of modern technology is making our lives easier by acting more as we do. Is it possible for a machine to pretend that it knows what we're thinking?
- Anticipated needs: Demonstrated by type-ahead features in IDEs, this experience pattern provides an interactive shortcut to selecting among many well-defined options. Unlike search engines, the choices are predefined and predictable. If a programmer types "NSO", Xcode will reliably show her "NSObject" as a choice in the same manner, each time. As for mindset, she needs to perform very little evaluation, just identification and selection.
- Guessed needs (fuzzy anticipation): Like recommendations on Amazon and type-ahead suggestions in search engines, this makes an educated guess as to what someone might want. When presented with choices of varying and often unpredictable accuracy, a user's mindset changes from one of searching, browsing, or query-writing to one of evaluation and judgment. Instead of working to express what she wants, a user needs to think about multiple choices to find the best fit.
- Finish the job: This is best shown by the "fill" feature in Microsoft Excel or Apple Numbers that extends formulas along rows or columns. Features that finish someone's work will work well when the pattern is known and easy to express, and the corresponding manual burden is great. The productive benefits can be significant if the user is confident in the outcome and not hesitant because of past errors or unpredictability.
- Edit, don't author: Instead of risking writer's block (blank screen) or paralysis of choice (too many options to evaluate and pick from), give a user a single starting point that they can tweak to their liking.
Perhaps a mindset can target a specific emotional state.
- Delight and surprise: This classic and popular experience goal means to exceed a user's expectations in a pleasing way.
- Delight but don't surprise: Instead of surprising, delight a user by actually meeting their expectations directly and openly, but without any frills.
- Experiment risk-free: Enable someone to test drive features or explore possibilities free from unrecoverable error and feelings of blame. E.g. Undo is always there to save the day.
- VIP treatment: This speaks to a club-like experience, found in invite-only websites or membership sites. It's intent is to make people feel special, so they're more likely to engage and evangelize.
Multitude of Experiences
All products, digital or physical, are a pastiche of multiple experiences. People must learn to use a product, use it in context, troubleshoot it when things go wrong, get help if needed, etc. Each can have different mindsets assigned to it, and design process takes over from there armed with valuable, explicit, experiential goals.
And finally, since the phrases used for mindsets, like "understand at a glance," are so simple and familiar that they can even be used to include stakeholders, managers, and other non-designers in the design process as well, opening many possibilities for collaboration and inclusive design process.