By Seth Snyder, Senior Director of Design

“This platform is truly a breath of fresh air I wish they were all as easy as this :-)”

“This platform is SO EASY TO USE!! I was so excited this morning when I completed my first auth!! This is so much better than what we were using before!! THANK YOU!!”

“This really helps everyone work toward a common goal of helping the patient”

“We LLLLLLLLLLLLLLLove it!!!!”

“That was fun!”

These are the kinds of things our users tell us about the experience we’ve created, and gosh do we love hearing it. I probably don’t have to tell you that healthcare software, especially those that deal with Prior Authorization, typically are not referred to as “a breath of fresh air.” They don’t get described as “easy to use,” or recognized as contributing to what patients need, and they certainly are never thought of as “fun.”

From the beginning, Cohere set out to transform the healthcare system by taking a patient and user-centered approach to utilization management, a system that has historically leveraged denials and delays in care as primary mechanisms of cost reduction. Our software has driven a 38% reduction in provider administrative time and expense. Providers have reported high satisfaction with Cohere’s platform, with 92% indicating they are “satisfied” or “very satisfied,” and 85% indicating that the Cohere platform is “easier” or “much easier” to use than alternative solutions. And this satisfaction means that providers submit their authorization requests in our web-based application – as opposed to phone, email, or fax – 94% of the time, a number unheard of in the healthcare IT sector.

How did we do it?

A design-centric approach.

More specifically, we have built a team of world-class Product Designers and User Researchers and established a collaborative, cross-functional process for solving user problems with elegant, delightful, well-crafted solutions. We know that in order to solve some of healthcare’s gnarliest challenges and overcome some of its strictest constraints, we needed to invest in a process that is more often reserved for consumer products – one which involves deeply understanding our user’s needs, motivations, and behaviors, exploring wildly creative potential solutions to problems, testing prototypes and iterating on ideas with users, supporting the engineering process to ensure every detail is well-crafted, and finally, measuring outcomes to evaluate the efficacy of our solution. It’s a process that marries what is desirable from a human point of view with what is technologically feasible and economically viable.

There are essentially 5 phases to our product development process at Cohere. You can imagine them arranged in a linear way, but in practice, we work in iterative loops weaving back and forth across phases, some taking longer than others.

Discover

In this first phase, our goal is to develop a sharp understanding and articulation of the problem we’re trying to solve. We start by discussing the problem, how it impacts our users and our business, and what success metrics we may use to be confident we’ve addressed it successfully. We’re documenting our open questions and figuring out how we’ll answer them, whether it be looking through past research or conducting new qual or quant research. Outputs from this phase include a current-state user journey map and a written document that captures the refined problem statement, business goals, success metrics, and research findings.

Ideate

While we try to hold off on solutioning until we feel like we sufficiently understand and align on the problem, sometimes it helps to start ideating during the Discover phase. Either way, the goal here is to explore a vast range of ways to solve the problem in the lowest possible fidelity that can give us sufficient confidence to decide on which direction to go. Not everything needs to be figured out, but we should cross the threshold where we feel confident that this is the right solution and we’re not missing out on opportunities to solve it in a better way. During this phase, we’re brainstorming wild ideas, illustrating new user journeys, sketching and wireframing the ideas we’re exploring, evaluating the technical constraints and implications of our concepts on the broader system, gathering feedback from subject matter experts and end users, and eventually, reaching an agreement with the team about which of the proposed approaches to move forward with.

Design

Once everyone is aligned on a solution direction, we dive into the details of the design. Our goal at this point is to figure out and document the user flow, interaction design, and visual design details to enable engineers to start building the first, smallest increment of the solution that we think will deliver value. As we’re building and testing prototypes with users, many times we find that we need to go and do more discovery or ideation work, and we embrace that as part of the process.

Build

In the build phase, our goals are to collaborate with the engineers to finalize any remaining design edge case details, test and refine the experience, and ship it! This is the magical phase of the process where all of the discovery, ideation, and design come to life in a real, working thing. And as with building anything, sometimes we encounter issues that necessitate some iterative loops back through portions of the process.

Measure

Once the solution is live and people are using it, it’s critical to track the impact by understanding what worked and what didn’t with respect to our original success criteria. We use a combination of Pendo and Tableau to measure user behaviors and assess satisfaction with the experience, and report back on how successful the solution was. The outcome of this phase is most often either a team celebration if we achieved our goals with the new solution, or, equally exciting, an actionable plan to learn more about what didn’t work and how we might improve the next iteration of the solution.

Core values that support a design-centric approach

If this all sounds like something you’d never be able to achieve at your organization, I’m here to tell you that you can, and you should. The best thing about this design-centric collaborative process is that it allows people who aren’t trained as designers to use creative tools to address challenges like improving our healthcare system. Whether you are an executive at a large healthcare company, a solo designer at a health-tech startup, or anyone in between, you can leverage a design-centric approach and make healthcare software that people love using, by applying a set of core values and principles to your work.

Collaboration

Everyone at Cohere, from our C-level executives to our product managers to our engineers and clinical writers, participates in the design process, enriching both our final solutions as well as our empathy for our end users. If you’re not already doing cross-functional discovery, ideation, and scoping sessions, start now.

Curiosity

Our teams are rooted in wonder. We usually don’t know the answer when we first approach a problem, and that motivates us. Every day is a new opportunity to ask questions and learn about our users. Avoid getting caught up in your own biases by bringing together internal and external subject matter experts and end users to generate and evaluate concepts and co-design final solutions.

Experimentation

We ship quickly and learn. Our philosophy of experimentation and testing leads us to ship new ideas incrementally so that we can test our hypotheses, observe what happens at scale, and understand the true impact of our solutions. Making sweeping design improvements to your product or experience can be intimidating. Start small and work your way up to your vision.

Quality

We sweat the details because we know they matter. We make experiences that are expertly crafted and pixel-perfect. Even though we have to work fast and ship incrementally, we never sacrifice quality. Even if you’re shipping the smallest of feature updates, leverage design systems, common components, and a diligent QA process to ensure the pixels are where they ought to be. Believe me, if you don’t, the design debt will quickly become insurmountable.

Design principles that help create products people love

As we make more detailed design decisions for a given product, we use a set of design principles to ensure that we’re all aligned on what the Cohere experience feels like. These principles ​​help us design great products in our day-to-day work, while also supporting our vision for the future.

Guide the way forward.

We streamline workflows by guiding users to the best next step for them to accomplish their task. We make hard things easier, tedious things automated, and confusing things delightfully simple. We lead the way, not only through the interface but towards higher quality, evidence-based care.

Put this principle into practice:

• Use simple, human language for UI copy.
• When there are multiple ways to proceed, the options should be easy to understand and if there’s a best option, make it obvious.
• Help the user focus on one job at a time through well-structured information hierarchy.

Earn trust with transparency.

We earn the trust of our users by transparently communicating what will happen and why. We are clear about our intentions, honest in our actions, and free of dark patterns. When we don’t have enough information, we embrace humility by asking questions and adapting.

Put this principle into practice:

• Persistently communicate the status of the user’s work and what’s happening behind the scenes, with access to learn more.
• When the system does something automatically, provide explanatory text with opportunities to dive deeper if the user is interested.
• Help users understand what will happen before it happens via confirmation modals, auto-approval predictions, and simulation tools.

Inject joy into healthcare.

We create experiences that make our users feel joy as they work, proud of their accomplishments, and fulfilled by the greater mission of patient care. Through friendly and familiar interaction patterns and language, easy access to human support, and occasional moments of surprise, we treat our users to an experience they feel competent using and truly enjoy.

Put this principle into practice:

• Use friendly, accessible, and inclusive colors and illustrations to enhance meaning and usability.
• Use animation and interactivity to give users feedback and celebrate their accomplishments.
• Make it clear at all times how users can get help if they feel stuck.

Respect the user’s expertise.

We trust that our users will act honorably and in the best interests of their patients. We embrace the fact that our users have a unique and valuable perspective about what’s right. We give them room to express their expertise and ourselves the room to learn from it.

Put this principle into practice:

• Assume users know more about the patient and medicine than we do.
• When making recommendations, do so respectfully and make it clear if they are optional.
• Help users understand that we trust them to enter accurate information and we will auto-approve based on that.

If these principles make sense for you and your team to use, then great! If not, I suggest creating a set for yourselves that are tailored to your organization and products. Get your design team (or design-first approach advocates) together and brainstorm with prompts like:

• When a user interacts with our products, they should feel _________.
• In the future, our products will be known for being _________.
• Unlike our competitors, our products _________.

The challenges ahead

If there’s one thing that continuously surprises me the most about designing healthcare software, it’s how easy it is to succumb to the vast complexity of it all. What may start off as a simple idea can quickly balloon into an epic spaghetti mess of an interface. Before you know it, you’re feeling overwhelmed by the amount of data or form fields or buttons you need to find a way to present on the screen. When you also consider all of the maddening interoperability constraints, the arcane yet ever-changing regulatory environment, and the frustrations of overworked doctors who despise their software but are reluctant to try anything new for fear of losing efficiency, you have an extremely challenging environment in which to design anything good, let alone lovable. Furthermore, the end user is most often not the economic buyer of the software, making them powerless in influencing which tools are used or how those tools could be improved. All of this, unfortunately, is what has led to the bleak and user-unfriendly ecosystem of legacy healthcare software we find ourselves stuck in. And it is precisely why a design-centric approach is critical to bucking the trend. I hope that by sharing Cohere’s design process, values, and principles with you, together we can create more healthcare software that people actually love <3