A New Way of Understanding Sports Fans

A lot of sports organizations think about their fan base in terms of subscription tiers. Their business strategy is largely about moving fans up those tiers, converting them to higher levels of monetization. Accordingly, they ask themselves questions like: what would it take for a fan to upgrade to a season ticket holder, an ESPN+ subscriber, or a daily reader of sports news? 

This approach makes sense. After all, a company is in the business of monetization. But to get a fan to upgrade, they must first and foremost be engaged with whatever it is you’re offering — be that a product, a team, or the game itself. To bring them up the tiers is essentially to ask them to increase their level of engagement with you. And if you want fans to engage deeply, you have to deeply understand what it is that they want.  

In other words, the better you can understand the crux of a fan’s engagement — how it is shaped, how it’s maintained, and how it grows (or stagnates…) — the better you can cultivate their inspiration to upgrade. Getting to this level of a fan identity requires an intimacy with their beliefs that goes beyond the details on an account subscription. 

A fan-centric view of sports

As mentioned, while there is clear value in analyzing subscription trends, it is inherently top-down and corporation-centric, placing fan behaviors primarily in relation to their monetary value for the company. If this is the key variable by which segments are sliced and diced, it can limit the organization’s ability to surface the most meaningful characteristics and variations that define their fan base. And, subsequently, limit the organization’s ability to serve such needs, and get the very upgrades they are after.

What many sports companies could use is a complementary bottom-up approach to segmenting and analyzing fans. This approach would start by understanding how and why a fan engages in a sport. Is it all about supporting a particular team? Is it a larger appreciation for the sport? Is it about the culture? Belonging? Nostalgia? Hometown pride? Is it about going to games because all their friends do? Starting on the ground to understand attitudinal and behavioral differences across fans can set organizations up to learn something deeper and more important about that fan than their subscribership status — things that ultimately do more to determine how they can serve each segment. 

Stories from the stadium 

In a previous project, a major sports league asked us to overhaul their mobile app. Their goal was to get fans to spend more time on the app so they could generate more ad revenue. When we started doing research on their fan base, we uncovered surprising trends that ended up influencing the league’s overall engagement strategies. For example, there was a significant portion of the fan base who were what we called “adopted fans.” Instead of inheriting a team from traditional family ties, they adopted a new team when they moved to a new state, or adopted one based on its underdog status. As newer fans, they looked to national news for sports intel. Diehard fans, on the other hand, primarily went straight to their local beat reporters for sports news. This stratification uncovered opportunities for the league to serve each group differently and personalize their experience on the app, increasing the engagement opportunity for each group. 

In another project with a major sports team, the avidity level of a fan turned out to be among the most important characteristics to analyze. We uncovered, for instance, a segment of the population we called “tag-alongs” — those who attended a sporting event because someone had invited them. Many of these tag-alongs didn’t know much about the sport to begin with, but loved the experience of going to a live game and rooting for the team. For this sub-group, the atmosphere and amenities at the stadium made a big impact on their likelihood to return. Once this group was uncovered, the team was able to do more to convert these tag-alongs into fans in their own right. 

Doing the “field” work: meeting fans where they are

Let’s assume your organization has bought into the value of uncovering the unique fan archetypes within their population. What comes next? 

One important way to research fan attitudes is, of course, going to games. Observing fans interacting with their sport or team, and also observing them in community with one another at games, is not to be overlooked. 

But fans are not only fans during sports games. They are also fans when they are reading the news, keeping up with players or stats or the league at large. They are fans when they’re out at a bar with friends and see their favorite player’s jersey on the wall. They are fans during off-season as well, even when there aren’t as many ways to show it. 

Understanding a fan means understanding the rhythm of their fandom, the ebbs and flows in addition to the moments of peak excitement and engagement. How they stay connected to their team or sport when games aren’t going on can be just as informative as how they behave during a game. There are cadences to the experiences of different sports fans, and understanding that richness of detail is key to understanding how their needs can best be met. We’ve found that understanding these harder-to-capture aspects of fandom require different research methodologies — for instance, perhaps you need diary studies to check in on fans during off-season or lulls in action. Perhaps you need to post up in a sports bar and catch people stretching out the emotion of a game by connecting over it. Perhaps you need data points from people as they read sports news throughout the week. 

Sports mean a lot to people. For some, their fandom is a key piece of how they see their own identities. Taking the time to understand these segments with multidimensional attributes with care can pay off greatly for fan satisfaction as well as overall engagement metrics.

Looking to better understand and serve your fans? We’d love to hear from you!

The 2024 Design Forecast

We did it, everyone. We made it through another year. To be perfectly frank, this year was a bit of a weird one, mainly due to a few key elements:

  • A real will they, won’t they dance with a potential recession that constrained the budgets and resources of many organizations this year
  • Generative AI coming front and center and taking over everyone’s conversations and questions about the future in product and business
  • More layoffs across industries, impacting the design & product communities as well as what the remaining team members can accomplish

So let’s look ahead to next year, shake our magic 8 ball, and make our predictions about what trends we’ll see in 2024.

Generative AI Implementation Begins 

2023 many folks spent learning and hiring for generative AI leadership in their organizations. Certainly both these things will continue into the new year, but orgs are starting to feel ready to “try something” in generative AI and set up their first initiatives and pilots to test out what AI can do for them. 

One unsolicited piece of advice we’ll give from our previous experience: take your time to really define and scope the problem you’re solving, make sure AI is the right solution for your user group, and start small. It’s easy to skip past these steps but most everyone is in a test-and-learn moment right now, so if ever there was a time to understand before blasting a technology everywhere, this is the time.

The Return of Innovation

Businesses at large took a collective pause from innovating to focus on optimizing what already exists. Part of this is burnout from the last few years, part of it was budget slashing and belt-tightening. And to be honest, this will likely continue into the first half of 2024, especially given the geopolitical and economic tensions currently simmering. 

However, our bet is that the second half of 2024 will begin to see a return to pushing innovation, particularly for incumbents who may have newcomers nipping at their heels. It’s always nerve-wracking to feel like you’re losing momentum in the market, and innovation is the way to stay ahead of the pack.

Consumer-Focused Mixed Reality Hits the Shelves

In a prediction that merges both innovation and emerging tech, we’re seeing a trend towards more mainstream utilization of mixed reality and spatial UI design (things like Vision Pro and Quest3), particularly in the enterprise side of things on tasks like training and digital twinning. 

This trend is still looking for its footing on the consumer side, which leaves 2024 wide open to folks looking to lead in that space. We foresee the emergence of tools involving B2C applications in this realm, though it won’t yet be a saturation of the market. Frankly, if the economy continues in an up and down pattern, it will likely remain a luxury novelty in the consumer space until people have the money – or a reason – to invest in it. Likely 2024 will be a learning year for all of us watching this space to see what plays out with consumers.

Research Execution Will Extend Beyond Researchers

While many design and product teams were hit hard with layoffs in late 2022 and throughout 2023, it’s worth noting that research teams seem to have taken the hardest hits. Because of this, needed research is going undone and we’re seeing product managers and UX designers taking up the mantle of executing research – in particular, evaluative research that is necessary for the optimization work that’s been on the forefront this year. 

We see 2024 being a year where research continues to be a deficit and many organizations will look to contractors and agencies to fill the gap – but also to their internal folks with less of a background in how to execute. For this reason, we may also see an uptick in attendance for research conferences, books (we love Steve Portigal’s Interviewing Users and Caroline Jarrett’s Surveys That Work), online courses, and trainings to help support these additional asks on folks. 

Want to talk about your design & research support needs for 2024? Reach out to us!

Great First Impressions: Leveraging UI for Critical Product Moments

Your user interface is your digital first impression. And as we all know, a good first impression can change everything. 

UI design is often thought of as the creation of pleasing aesthetics. While this may be a part of what we focus on, the broader concern of the UI designer is to manage how someone feels while interacting with your content or page. Such feelings will translate, both consciously and subconsciously, into how people feel about you and your brand. Even minor confusion or frustration about what to click or how to input information can erode trust — who’s to say working with your business won’t be similar to using your website?

With ever-higher expectations for our digital world, and such high stakes, UI should not be an afterthought. Grand Studio’s UI team has compiled some tips for any teams taking on a new — or evolving — digital product. 

Tip 1: When everything calls out for attention, nothing calls out for attention

In an effort to get important components noticed, many designers try to crunch a lot of content above the fold (at the topmost part of a page, before needing to scroll down). Unfortunately, this most often ends up backfiring. Even if your user can see all the pieces in one glance, you risk overwhelming them, and generating confusion about what they should pay attention to.

We recommend placing one (or maybe two) areas of focus up top, sending the rest more into the background. Use color, contrast, size, and placement on the page to guide a user cleanly from primary to secondary focus. White space is your friend. 

Tip 2: Your top priority is helping your user know what to do when

It may sound obvious, but it’s far too easy to get caught up in components of the design and lose track of its primary purpose — helping the user take the action they need to take on the page. Consider the user’s paths through the page, and make sure they have everything they need to complete their journey. Never let your user wonder things like, “is this text interactive?” “what does this icon mean?” or “how do I fix the error I’m seeing?” 

At its most basic level, a visual interface is a means to communicate efficiently with your user so you can guide them through what they need as elegantly as possible. You want your UI to be a good communicator.

Tip 3: Everything that can be consistent should be consistent

If we think about a visual interface as a means of communicating with the user, it is critical that we are always using consistent language to do so. Though it may sound minor, using a color to mean one thing (e.g. red button = “remove”) at one point and another thing (e.g. red = “edit!”) at another point can be cognitively taxing on a user. As you establish your visual language, make sure there is always just one meaning attached to any given visual component, whether that’s an icon, a color, a word, or a button. 

These things not only save your user time, they help the user feel better on your page. Smooth sailing = trust generated. 

Tip 4: Consider your breakpoints, and don’t skip testing on the end device!

While of course there may be slight differences between different breakpoints or devices, there should always be visual and functional similarities across them. The idea here is to create an experience that is seamless and consistent for everyone, regardless of the device and its size. 

To create the type of layout that will be clear and usable no matter the size of the window, we recommend designing for breakpoints — points at which a design will “break” if you stretch your browser window wider — to ensure all of the page’s elements will nicely fit the available screen space, no matter the size. Considering breakpoints will naturally optimize content for viewing on different devices (even ones that haven’t been invented yet), ensuring everyone gets a clear and usable view.

That said, it’s still very important to view your design on the end device to be able to visualize it in a realistic way. You can do this by downloading your design and opening it up in your browser or device, like a mobile phone. Seeing your design in context with the device will ensure that elements or text are sized correctly and comfortably.

Tip 5: Accessible designs are better for everyone

Making sure your design can be used by as many types of people as possible not only increases the number of people who can interact with you, it can also help you find and resolve points of confusion that go on to help everyone. Accessibility isn’t icing on the cake; it is a critical component of good design. 

For example, take alt text for images, or text that describes what is happening in an image. Alt text’s primary function is to help those with visual impairments hear through text-to-speech what each image contains. But alt text can also be really helpful for Google searches, and helping people find content on your page. It’s invisible for those who don’t need it, but there for people who do. (For more on this, check out our blog post on accessibility.)

At Grand Studio, our UI team does way more than “pixel push.” Because we see UI as a critical part of a holistic design practice, UI is integrated into product strategy, UX, and even design research. We know that the best user interfaces come out of a deep understanding of not just UI best practices, but an understanding of the particular context and goals of each client we work with. When you’re making a first impression with your users, nothing could be more important. 

Got a user interface project? We’d love to hear from you?

Human-Centered AI: The Successful Business Approach to AI

If AI wasn’t already the belle of the tech ball, the advanced generative AI tools surfacing left and right have certainly secured its title. Organizations are understandably in a rush to get in on the action — not just for AI’s potential utility to their business, but also because, more and more, demonstrating use of AI feels like a marketing imperative for any business that wants to appear “cutting edge,” or even simply “with the times.”

Sometimes, rapid technology integrations can be a boon to the business. But other times, this kind of urgency can lead to poor, short-sighted decision-making around implementation. If the technology doesn’t actually solve a real problem — or sometimes even when it does — many don’t want to change their process and use it. All this to say: a bitter first taste of AI within an organization can also harm its chances of success the next time around, even if the strategy has improved. 

At Grand Studio, we’ve had the privilege of working alongside major organizations taking their first high-stakes steps into AI. We know the positive impact the right kind of AI strategy can have on a business. But we’ve also seen the ways in which pressure to adopt AI can lead to rushed decision-making that leaves organizations worse off. 

Our top-level advice to businesses looking to implement AI: don’t lose sight of human-centered design principles. AI may be among the most sophisticated tools we use, but it is still just that — a tool. As such, it must always operate in service of humans that use it. 

A human lens on artificial intelligence

When implementing AI, it is tempting to start with the technology itself — what can the technology do exceptionally well? Where might its merits be of service to your organization? While these may be helpful brainstorming questions, no AI strategy is complete until it closely analyzes how AI’s merits would operate in conjunction with the humans you rely on, whether it be your employees or your customers.

CASE IN POINT 

In our work supporting a major financial organization, we designed an AI-based tool for bond traders. Originally, they imagined using AI to tag particular bonds with certain characteristics, making them easier for the traders to pull up. It seemed like a great use of technology, and a service that would speed up and optimize the trader’s workflow. But once we got on the ground and started talking to traders, it turned out that pulling up bonds based on tags was not actually their biggest problem. AI may be a golden hammer, but the proposed project wasn’t a nail — it only looked like one from far away. 

As we got more clarity on the true needs of these traders, we realized that what they actually needed was background information to help them make decisions around pricing the bonds. And they wanted the information displayed in a particular way that gave them not just a suggestion, but the data that led them there. In this way, they’d be able to incorporate their own expertise into the AI’s output. 

If we had designed a product based on the original assumptions, it likely would have flopped. To be useful, the AI needed to be particularly configured to the humans at the center of the problem.

The linkage points between human and AI are crucial

We all know that bad blood among employees can spell doom for an organization. Mistrust and negative energy are surefire ways to sink a ship. In many ways, integrating AI can feel a lot like hiring on a slough of new employees. If your existing employees aren’t appropriately trained on what to expect and how to work with the new crowd, it can ruin even the best-laid plans. 

Once you’ve identified where AI fits into your organization, we recommend paying extremely close attention to the linkage points between human and AI. Where must these parties cooperate? What trust needs to be built? What suspicion needs to be mitigated? How can each benefit the other in the best way possible?

CASE IN POINT

Recently, we worked with a financial services technology provider to develop AI that could spot fraud and inaccuracies in trading. We conducted in-depth research into the needs of the surveillance teams who’d be using the software to understand their role and also their expectations for how they’d use such a tool. This allowed us to thoughtfully build a visual interface on top of the AI that could maximally meet the surveillance team’s needs, including helping them with task management.

Taking the time to understand the precise nature of this potential human-AI collaboration helped us use resources wisely and prevent the mistrust and resistance that can cause even the best tools to fail. 

AI integrations require trust and understanding

Your AI also can’t be a “black box.” While not everyone at your organization needs to be an expert on its functionality, simply dropping an unfamiliar tool into a work environment and expecting people to trust whatever it spits out is very likely misguided. This is especially true when AI is helping experts do their jobs better. These roles are defined by the deep training that goes into them — how are they supposed to give an open-arms welcome to a new “employee” whose training they can’t see or understand?

For example, a doctor trained in reviewing mammograms may well benefit from AI software that can review 500 scans and whittle it down to only 20 that need human assessment. But you can imagine a physician’s resistance to simply taking those 20 images without understanding how and why the software weeded out the other 480. They rely on their expertise to save lives, and need to trust that whatever tools are helping them are supported by similar training and values. 

AI has the power to make big change. But if we don’t center humans in our implementations, the change we make may not be the good kind. 

Contemplating your early steps into AI? We’d love to work with you to help make your leap into the future a smart one. 

Designing Products for Healthcare: 5 Important Considerations

In the healthcare space, the design choices you make can quite literally have life-or-death stakes. Getting it right is important. 

But healthcare environments are unique spaces, and what works in other industries might not always carry over. In addition to regulatory considerations like HIPAA, many healthcare organizations have distinct cultures and ways of doing things shaped by decades of caring for people, often in extreme circumstances. 

Grand Studio has had the privilege of working with several healthcare organizations over the years and has come away with some rules of the road when it comes to product design in these specialized spaces. Read on for five key things to keep in mind.

Tip #1: Involve Clinicians & Other Stakeholders from Day 1

For the best chance of success, bring in key stakeholders right from the outset — particularly clinicians. For one, their perspectives will be critical to developing whatever you’re creating. But also their involvement will also help build buy-in and trust. 

Too often, clinical teams get burned by the debut of some new technology that was clearly built without insight into their day-to-day experience and ends up causing more headaches than it eases. Looping these key stakeholders in immediately and keeping them up to date as the design process moves forward will have a two-pronged benefit: you’ll spot potential problems in the design, and you’ll also do a lot to socialize your effort. You build faith that you’re listening to them, working to understand their unique, high-pressure world. 

That said, keep in mind that they are busy literally saving lives. They may not be available for every collaboration you’d want from them, so as part of your Day 1 involvement, settle on a cadence that gets your team the input you need while respecting their often busy schedules.

Tip #2: Onboarding Must Be a Part of Your Design

Take the time up-front to consider the onboarding process. People working in healthcare environments, from doctors to nurses to administrators, almost always have a great deal on their plates, and what’s on their plates is extremely important. Changing a process or asking people to adopt a new product can feel extremely disruptive. Even something people may ultimately find helpful and time-saving might gather proverbial dust if the channels of a routine run too deep — especially if it’s not explicitly clear why or how people should switch things up.

The single most important thing you can do to help onboarding is getting an internal champion — someone who believes in what you are doing and can support you in socializing it from within. Clinicians tend to place a high degree of trust in insiders who know through experience what their day-to-day life is actually like. Finding the key leverage points in the culture of an organization and getting them on your side will be critical to any onboarding/socialization plan. (In our experience, the most powerful shifters of culture are doctors and nurses.)

And of course, onboarding is not a one-and-done thing. Just as the design requires iteration, so too does onboarding. It’s important to continually go back to the front lines and tweak how the value prop and plan are described, paying attention to what yields the best adoption.

Tip #3: Design for Rapid Action

Healthcare providers frequently need to do things quickly. While efficiency is valuable in any situation, there are particularly time-sensitive moments in healthcare — like responding to a patient with a critical condition exacerbation.

In a recent project with one of the largest private healthcare organizations in the US, our task was designing tools for nurses to remotely monitor patients with postpartum hypertension. In our design, we asked ourselves how we could enable nurses to quickly identify which patients required their attention most urgently — digital triage, in essence. If a patient in a life-threatening condition was identified, we also asked ourselves how we could best support the subsequent action that needed to be taken. For patients with dangerously high blood pressure, we worked with our client on a system by which nurses could immediately alert not just the attending physician but also the patient and their circle of care. Nurses could act rapidly on the situation and also keep everyone connected.

Tip #4: Allow for Personalization

In the clinical field, there is a vast diversity in job functions as well as people’s way of doing things. Instead of making a one-size-fits-all solution, you’re better off locating and enabling key points of personalization that allow people to do their jobs in ways that suit their needs. 

We recommend interviewing wide sets of end users and attuning yourself to the subtle differences in process that could inform how you allow for personalization of the product. Some clinicians, for example, only need to view a subset of the patient population in order to do their job, and anything beyond that will be visual clutter. Some clinicians need to filter down by a particular biometric or health status marker. Some clinicians need to respond to issues in different ways than others. Building in personalization helps you meet people where they are and let them practice medicine the way they know best.

Tip #5: Support the Patient-Provider Relationship

The job of technology in healthcare should always be one of making healthcare providers more efficient and effective — not impinging upon or trying to replace a clinician’s relationship with their patients. No matter how “intelligent,” technology is ill-suited to replace this powerful and healing relationship. 

Focus instead on getting technology to solve lower-level problems so that providers can spend time on patients in need of their skill set. Focus on designing tools that enable high-impact interactions and offload low-level ones. Figure out ways to optimize and clear a path for what care providers do best. 

Want to learn how Grand Studio can help with your next healthcare project and build clarity out of complexity?

Drop us a line! We’d love to hear from you. 

Using AI in Enterprise

AI is everywhere these days. There’s no escaping it whether it’s a board room conversation, a conference, or a meme on social media. And with good reason. As Bryan Goodman of Ford Motor Company said recently at the 2023 Reuters Momentum conference on AI, “I can’t imagine anyone being competitive without AI.” The resounding perspective across industries is in agreement with him.

The question amongst many people, particularly those in larger enterprise organizations with less scrappy flexibility and more risk, is how do we use AI in a way that is responsible to our business units, our shareholders, and humanity at large?

Download our whitepaper and read more about the best use cases and challenges to consider for enterprise AI usage.

How to Innovate for Long-Term Vision

Many of us have been in a position where, amid our day-to-day tactical work, there surfaces a nagging suspicion that some long-term organizational vision is needed. You know something major needs to change, that some kind of high-level ethos-tinkering could push your team forward… but nobody quite knows what that thing is.

Even if you manage to set up a brainstorming session to get some fresh ideas out, what often happens is that a few wild ideas will be thrown out, then immediately nixed due to (very real) feasibility or viability concerns. How does anyone get past these blockers to arrive at exciting — yet grounded — innovation?

What does a successful visioning process look like?

This is a common problem we tackle frequently at Grand Studio. While every project will have constraints to take into account, successful visioning typically boils down to 3 main things:

  1. Allow time to think expansively
  2. Keep your ideation separate from your editing
  3. Choose activities that match where your team is at 

We’ll take you through each one of these in a little more detail.

Think expansively 

The purpose of visioning is to let go of constraints and dream up potentially surprising solutions, knowing we can always come back to earth afterwards. The idea is to think big, to make the kinds of lateral, creative connections that generate game-changing ideas.

To do this, we recommend setting expectations and reminders with your team that constraints don’t matter right now. We also recommend helping your team practice expansive thinking with a warm-up. Very often, with our day-to-day jobs, our brains are trained to think pragmatically, and making the cognitive shift into a visioning session can be challenging.

Forced connections can be a great exercise for people new to expansive thinking. With this, you take 2 seemingly unrelated topics or items and force yourself to connect them in some way. You can do this with physical objects in the room you’re in, current events, items in people’s bags… anything at all. 

For example, say your first pairing is an iPhone charger and an apple from someone’s lunch. How could they be related? Maybe there’s a novelty phone charger that uses the energy from an apple to charge an Apple device… kind of like the potato-powered lightbulb of kids’ science experiments. You might start with a few silly pairings to get people in the groove, then start adding in pairings related to your business or user needs. 

Keep your ideation separate from your editing

Wild ideas can always be edited down. But applying too many constraints on imagination can end new ideas before they even begin. We recommend setting aside time for true ideation — even if it’s just a couple of hours — without letting feasibility concerns or fear of the unknown be the leading force in the room. 

To do this, have the group mutually agree upon what the dream-big phase of your ideation session will look like, and put everyone’s mind at ease by also setting expectations for what the editing down phase will look like, complete with agreed-upon deliverables.

Backcasting is a good technique for nailing down specifics that will eventually lead towards your collective vision. Keeping your future vision in mind, start working backwards and identifying the technologies, processes, people, and key milestones that would need to be in place to reach that goal. Keep working backwards in increments until you reach the current day. Now you can use these milestones when developing your roadmap or theory of change.  

Choose your ideation activities wisely

Jumping into the deep end with complex visioning exercises may be a lot for people who’ve never been involved with this kind of work before. We recommend that organizations newer to visioning exercises start off keeping it simple.

For organizations newer to visioning, try some exercises like:

  • Crazy eights: This is a quick sketching exercise that asks people to come up with eight ideas in eight minutes. The speed of this exercise forces people to let go of coming up with a perfect, edited idea, and instead dream up as many possible paths as they can. Looking back at what people shared, you’ll usually see patterns across participants that can illuminate new directions.
  • Exquisite corpse: You may have done this in school. Exquisite corpsing starts with each person creating an idea. You then pass each idea off to another person in the group, who adds on, or ideates a different part of the idea. This can encourage collaborative thinking and riffing off of others’ inspiration, as well as letting go of the impulse to step back and critically evaluate the whole picture.  

For organizations more comfortable with visioning and imagination activation, try:

  • Desired state storyboarding: If you had a magic wand, what would the ideal state of your service or product look like? How might people use it? What’s the ideal impact? How has society changed for the better? Create a narrative telling the story of your ideal state. This can be a drawing, a newspaper headline, or a storyboard. This can help everyone align on what the long-term dream would be, and set the stage to figure out the small steps you can take towards that vision.
  • Four Futures Scenarios: Making scenarios is similar to our desired states exercise above, but more in-depth. You can use these scenarios to think through multiple strategies, product ideas, and services to give your organization a wide range of action plans in order to proactively make your own future, instead of operating from a reactive place. Jim Dator at the University of Hawaii was one of the first to develop and popularize 4 key future scenarios that we can use when imagining alternative futures:
    • 1. Continuous Growth: the company grows infinitely and expands market share. (Tip: Stay critical! Think about when growth can be a negative thing for companies.)
    • 2. Discipline: the company faces significant limitations, either imposed internally or from other forces like government regulations.
    • 3. Collapse: the company struggles to survive due to things like market disruptions, economic recession, or other extreme factors. This future can be useful to think through disaster scenarios and work on “risk-proofing.”
    • 4. Transformation: the company changes something fundamental, such as its mission or ownership. This scenario is most exciting when thinking about ways to innovate and find something new. You can use Transformation scenarios to think about radical new directions your company can pursue in order to stay one step ahead of market trends.  

Now take a step back and look at all your futures. See any common themes or patterns? New product lines or directions your company could go that are exciting and generative? Pick one, or a couple, and try backcasting from it to see what steps you would need to take. 

Of course, there are plenty of ways to expand beyond these ideas into some really creative and out-there visioning exercises, but our advice is to start small to get your organization acquainted with what visioning can yield, and also build their faith that creativity doesn’t mean letting go of practicality. Once there is more of a culture around visioning and imagination, you can try exploring larger, more “out there” techniques. 

Bonus: Get an outside perspective

Sometimes, working with an outside organization can help with visioning. While no one will understand your organization better than you, outside eyes can help push past some preconceived notions and generate new ideas. They can also help set the tone for what visioning can look like, shepherding unfamiliar or suspicious parties towards the light of imagination.

Got a visioning project you’d like to collaborate on? We’d love to hear from you. 

What My Cancer Treatment Taught Me About Healthcare Design

Three years ago, I was diagnosed with an aggressive bladder cancer. This type of cancer is very unusual in otherwise healthy 42-year old people, but I caught it early enough that I was able to complete a full course of treatment, and my doctor has put me in the class of patients with the highest likelihood of long-term success.

For my type of cancer, treatment has meant 2 surgeries to remove the cancer and then 17 rounds of immunotherapy over the last three years. I think this immunotherapy is “better” than chemotherapy in terms of side effects, but it is still pretty unpleasant – it involved instilling live botulism bacteria into my bladder to stimulate a local immune response. No fun. I’ve also had an enormous number of tests – CT scans, scopes, a surprise biopsy (cancer-free, thankfully), fluid analysis, etc.

When I’m not dealing with cancer, I run a design studio that does significant work in the healthcare space. When I had my first surgeries, we were literally working on a product that helps improve scheduling of elective surgeries for a major healthcare provider. Last year, we helped another large provider redesign the service experience for patients undergoing cancer care at one of its facilities. It’s been an interesting exercise, and occasionally a useful distraction, to observe my care journey from the perspective of someone who has been responsible for designing care journeys for others. Kind of the ultimate empathy-building exercise (though I don’t recommend it).

A few things I’ve noticed along the way:

Existing online resources for people with cancer were really bad (for me).

When I first got diagnosed, I went to authoritative sites like the Mayo Clinic to learn about my type of cancer. This was a mistake. The best of these sites take the highly appropriate and responsible approach of writing about the disease with a near-clinical tone. They know that people are coming to their material to answer the big question: “Will I die from this disease?” The correct answer from a scientific perspective is to talk about morbidity over timelines across a broad population. This was actively harmful to me – it wasn’t helpful for me to see that an average person with my type of cancer has a significant chance of dying within 5 years. Intellectually, I knew that those populations included people with many comorbidities I do not have, and I knew that I was a couple decades younger than the average patient, but it still was data without comfort.

Balancing medication side effects is hard.

I came into this knowing this to be true – we literally had designed a cancer care app several years back that included medication tracking. Turns out living it gave me a much more nuanced perspective. Over the course of the last three years, I’ve very actively tried medications, observed their side effects, talked to my doctor, and rebalanced. I have the privilege of being a wealthy, educated, cis, white man, and communicating with my doctor was pretty easy for me. However, 80% of patients don’t complete the treatment I received, so I imagine that the same can’t be said for everyone. Without the tools, support or access needed to understand the side effects any one medicine has and to make changes, this process must be incredibly intimidating and discouraging for most people. It was really hard for me, and I was a best-case scenario.

I’ve never felt more like an object than when I was on a hospital bed.

When my car has an issue, I drive over to my mechanic, pull into the garage, and a couple of people descend on the car to poke at it. When I had my surgeries, I was the car. Someone literally wheeled me into an OR where there was a team of people quietly prepping surgical instruments that were about to be put into my body. I didn’t know these people, and these people only knew me as a diagnosis to resolve. It was a miraculous process – I was awake, then I fell asleep, and I woke up without cancer. It was also incredibly alienating. Because of my past work, I’m very aware that the hospital is going to want to do 5 – 7 cases in that OR on the day I had my surgery, and I sure did feel like a broken machine going in for service in a facility run as efficiently as possible.

I come away from this experience with a clearer sense of the design opportunity across this system: healthcare is dispassionate by design, and yet we humans get caught up in it and have human reactions like pain, confusion and fear. Despite my very positive experience with my providers, I didn’t get any support at all for those human reactions. And I’m just one person out of the ~2,000,000 that will have a cancer diagnosis in the US this year, which is a small fraction of the population that will interact with the entire healthcare system. That’s a whole universe of human need. I think it would take a lot of hubris to say that design has a significant role to play in mitigating all of that need, but there are certainly meaningful interventions that digital products and services could make.

Coincidentally, Insider just published this article on a JAMA paper comparing AI-driven responses to questions on Reddit to responses from doctors. It’s worth looking at – responses from the AI were rated by medical professionals as having higher quality and being more empathetic. The empathy piece is particularly relevant to me. I sought that out several times in the course of my treatment, and found it to be in really short supply.

I think what’s also clearer to me is that the scale of our healthcare situation, and its associated service problems, is so large that it’s impossible to imagine we will have enough humans to truly care for all the patients. There’s an enormous gap waiting to be filled by innovative products and design in allowing patients to better serve themselves. That Insider article above discusses the value of self-service in terms of diagnosis and triage. There’s also a huge need in supporting people with chronic conditions who may only see their doctor four times per year, but are dealing with their disease conditions every day.

I, for one, am looking forward to getting started working on these tools that will provide meaningful support between the doctor visits. Especially now that I’m done with my fucking cancer treatments.

Best Practices for Qualitative UX Research

The better the research, the better your plan. Combining the hard-and-fast numbers-driven data points of quantitative research with the broader “why” and “how” questions of qualitative research sets you up to make well-informed decisions that truly speak to the problem at hand. 

That said, to some, qualitative research can feel squishy and daunting. Protocols as rigid as a survey will prevent truly illuminating findings from surfacing, but also, going in with too loose a plan can lead to time-wasting or significant gaps in findings. 

At Grand Studio, answering juicy qualitative research questions is among our favorite things to do. We start by working with clients to understand the decisions they’re trying to make, then co-craft a plan to help answer those questions. While every research plan is uniquely informed by the context and questions surrounding it, we have a few best practices to share. 

Best Practice #1: (Really, Really) Know Your Research Questions

We know this one sounds obvious, but this is a common problem we run into with clients. They’ll have a general sense of what they want to learn about but haven’t yet distilled these into concrete research questions that can keep teams on track in the field. 

Stakeholders who’ve been steeped in a problem for some time can probably rattle off the most critical research questions on the fly. But when they aren’t concretized into a clear list, things can get lost in translation when more people get involved. Researchers constantly encounter twists and turns out in the wild, and their clear understanding of the research goals can be the difference between a smart on-the-fly adjustment and a wasted day. 

We’ve also seen situations where the research goals are made clear at one point, but as folks get wrapped up in the work and adapt to surprises, the initial need can get lost in the mix. As pedantic as it may be, we recommend putting the research questions at the center of every research check-in and synthesis meeting. “How are we progressing along our goal of learning X, Y, and Z?” Impossibly simple, but the benefit cannot be overstated. 

Best Practice #2: Be Wary of Shortcuts

Life gets busy, and shortcuts can be tempting, especially when your research questions are so vast. Every day, there is slick new technology released, promising to enhance and streamline your qualitative research. We’ve looked at every piece of technology that’s come our way and tried out a lot of the new tools, Ray-Ban Stories and dScout’s diary study tools. While some are helpful augmentation tools, nothing we’ve found is a viable replacement for on-the-ground research. 

The bread and butter of qualitative research is talking to humans, and there’s a reason for that. No matter how crunched your timeline, be suspicious of things that try to steer you away from that. 

Best Practice #3: Choose Your Methodology Wisely

There is an art to choosing a research methodology that is nearly as critical as identifying your research questions. Once you’ve got a clear understanding of what you want to learn, you need to figure out the best way to learn it, and with whom. Say, for example, you’re doing a study on how people make decisions while grocery shopping. An interview may be able to tell you general things about how people shop. But when you’re on the ground, observing or doing shop-along, you’ll get so many details people would never have thought to tell you in an interview. Plus, the way people explain the why and how of what they did in the past can often be very different from what they say in the moment.

Perhaps the most tricky part of this is responding to inevitable restrictions that come up. Usually, there will be at least one or two feasibility or legal concerns that keep you from using your “dream” population. This requires you to get creative about how you can optimally answer your research questions using the tools you have. You don’t have to be in an exact real-life situation to get highly usable data.

One last point on methodology is to consider the reciprocity of the research. Sometimes, research questions may be sensitive, or catch people in a difficult moment in their lives. You don’t want to skip these perspectives, but you do want to be sensitive to them. For example, at one point, we collaborated with a client on a project to better serve the needs of cancer patients. With their support, we set up a table where patients could get free warm socks in exchange for their thoughts. 

Best Practice #4: Go Wide, Wider Than You Think

One of our favorite parts of qualitative research is the surprise involved in answering such large-scale questions. But it’s also the trickiest because, by nature, one cannot entirely plan for surprises.

Knowing that, at the beginning of the research, you don’t know what you don’t know, we recommend casting a wide net. Talk to more types of people and roles than you think you need, especially if these people are easy to access. You don’t need to go overboard — just one or two sessions with a new type of participant can do the trick. We’ve made critical discoveries by taking the time to talk to someone slightly adjacent to the primary field of research who ended up having a great deal of insight into it. Go small at first, just to see if there are new stones worth turning over. 

Best Practice #5: Always Stay Aware of Bias in Synthesis

There is no way around the fact that humans are biased, even researchers who try very hard to be objective. 

We’ve found that the best way to introduce objectivity is not to push individual brains into objectivity, but to stack many ideas on top of one another. Try to use at least a few researchers, and gather their own unique perspectives on what they are seeing. Ideally, people from your organization should also review their findings and point out their own conclusions. Individually, no one’s perspective is objective. But when we see which trends emerge collectively, it’s more likely that they’ll be rooted in the actual research.

As is often the case with hard things, they frequently turn out to be the most worthwhile. Don’t be tempted to skip qualitative research if it sounds time-consuming, just make sure the research plan is well-matched to your timeline and budget. And, of course, we’re always here to help.

Got a qualitative research project you’d like to work together on? We’d love to hear from you.

When Technology Meets Healthcare: Practical Considerations From the Field

While every environment is unique in its own way, healthcare might take the cake when it comes to a need for specialized design considerations. 

There’s a trifecta of challenges: legal requirements to contend with, complex funding pathways, and the fact that what’s at stake might be life or death. How do you build technology poised to make a meaningful difference without endangering what’s already working? 

Grand Studio had the privilege of working alongside several clients on the front lines of healthcare and learned a few lessons.

Find a sustainable funding source, and check the CPT codes

With the interplay of public payers, private payers, and providers, the flow of funding in healthcare can be complicated. Whenever possible, aligning your intervention with a CPT billing code can be a smart way to position your technology of sustainable financing.

In the past several years, there have been more and more codes released that allow providers to bill insurance companies for digital health interventions like remote patient monitoring and patient engagement, and still, more are on the way. It’s a win-win-win for all involved: providers get reimbursed for investing in technology that improves health outcomes, patients get more tools that help them get and stay well, and the insurance companies spend less on patients who need less care.

Once you’ve found CPT codes that may be applicable, dig into the specific requirements to ensure your technology can meet them, and — critically — make sure your technology is designed to facilitate easy billing. 

While recently working with one of the largest US private healthcare systems, we co-designed an intervention for pregnant people with pre-eclampsia. We created a digital triage system to catch early signs of a problem, improving patient safety and staving off costly ED visits. While the avoidance of an ED visit was a saving unto itself for the healthcare system (and for the patient), we also designed the program to align with the CPT code’s billing requirement, which stated that providers could only bill for the remote monitoring if they saw the data and acted on it. As designers, this kind of requirement is frustrating in how much it limits the service experience. We saw enormous opportunities for automated alerts and interventions. Still, it was critical for us to remember that the service could not be funded and, therefore, couldn’t exist if it did not live within the payer requirements defined by CMS. Understanding our real-world limitations allowed us to build something viable and desirable.  

When patients are involved, study their mindsets carefully

We are perpetually in favor of doing robust research on end users, and healthcare is no exception. 

Whether leaving the hospital, adjusting to a new diagnosis, or even coming in for a check-up, people’s experiences in and around a doctor’s office can be highly emotionally charged. You may be dealing with people who are exhausted, people who are sick, people who are nervous about their loved ones, people who are confused but afraid to tell their doctor or people who can’t process any information at all because they are still in shock. Standard “design conventions” that may work in other environments could crumble in healthcare if they don’t address the specific emotional valence of the patient experience.

For instance, we once worked with a nonprofit health system to redesign how cancer patients checked into the hospital. In most contexts, efficiency is priority number one — the fastest solution is the best solution. In this setting, however, it was also essential to prioritize the slower face-to-face interactions that let these patients feel like humans, not numbers being churned through a system. Moments of warmth and connection were precious for people dealing with the magnitude of doctor’s appointments that come with a cancer diagnosis. (It’s worth mentioning that helping people feel more connected to their healthcare providers doesn’t just make them feel better — it can improve health.)

Understanding the patient’s mindset is also crucial when your goal is health behavior change. As healthcare providers know all too well, simply having the solution to a healthcare problem doesn’t mean it will be solved. Patients often do not follow their provider’s recommendations to take their medications or implement a lifestyle change. It usually isn’t because they don’t want to — it’s simply because there are too many barriers. Often, these barriers are emotional. It requires thorough research to stand a chance at cracking the code on what actually moves the needle on patient behavior. 

Despite the challenges of healthcare, it’s a gratifying field to work in, and we’re eager to collaborate with more clients trying to support people where it matters most.

Got a healthcare technology project you’d like to work together on? We’d love to hear from you.