When we are building Business Intelligence tools and reports for our healthcare clients, it’s rare for us to hit the limits of the technology we’re using. The challenge is to balance the use of technology so as to offer the care team useful tools for making decisions, taking action and sharing information while avoiding “feature-bloat” that overwhelms or distracts.
Building these types of resources can be especially challenging due to the complexity of patient care, its variability, access to data sources and the human factors that are present; the judgement required on the part of the caregiver: the “art” of healthcare that by necessity has to be paired with the science.
Below are a few of the lessons we’ve learned from our client engagements, especially presenting our work to groups that are often laboring under many projects with unrealized potential, general technology fatigue and more bureaucracy than ever in healthcare.
You Must Solve a Problem
Falling in love with your product is the biggest pitfall for technologists. Most end-users are not interested in the technology you chose or how it was designed except that it has to meet their needs, articulated or otherwise. This is why prototypes and watching users interact with your work is so important and often so humbling. Ask yourself what the problem is that you are solving and consider the perspective of your user. It’s even better if you can have a committed user help you get started.
Are you presenting to providers? If so, show how your approach saves them time. More time is the ultimate gift to providers, more than money. More time means more people that a doctor or nurse can help. Can you get the patients with the highest need into the office and delegate simpler issues to another member of the team? Are you helping decision makers? Ask yourself how your work lends itself to actionable information.
Make it Real
In storytelling, the best advice is “show, don’t tell.” When you let your audience see themselves in the interaction, presentations come to life. Try to hone in on one or two examples of what a user can do with the data. In software development, the idea of personas or typical users, is often used. Consider a few of these users and what they might be trying to specifically achieve. Is this information best used by a doctor, nurse or care manager? Be realistic about who will really interact with the information. Anticipating the actual end-user prevents the cognitive overload for your audiences when you say that they can do “anything and everything” with the product – they don’t want to do everything, they want to do something.
By telling a real story (using properly de-identified data if appropriate), your application or methodology becomes much more concrete. Our minds work in analogues; if we’re shown one way of doing something, we often adapt it for other, similar tasks. Don’t worry that you haven’t thought of every possible way the viewer might look at it – get a strong and viable prototype out for review. In addition to helping you understand your audiences’ business specifics, feedback based on the personas you used can help you prepare training and testing strategies as well as adapt and evolve your product or approach.
Do Your Best to Make it Intuitive, Then Iterate
If you have to explain the joke, the saying goes, it’s not funny. If you have to explain how to interpret a graphic, you probably need to work on making it better. Consider different ages, learning styles and populations when presenting data: most people prefer to engage graphically with complex information and avoid columnar display of large sets of numbers. Many men and some women are color blind which can impact how to best present data. This can be difficult for the color blind person who doesn’t always want to hold back a presentation while they try to make sense of an illustration.
Likewise, if a user has to be shown how to use the menus or other tools, then they likely need to be reorganized. Consider consumer websites like Amazon or Gmail: they are continually refined to create a “frictionless” experience for the user and that is what users of enterprise technology are coming to expect more and more. Put processes in place that encourage continuous improvement.
Get the Numbers Right
This should go without saying, but for most presenters, reliable data is still a challenge and one doubt about the veracity of the data can derail the most compelling presentation and create an unnecessary barrier to adoption, especially with providers. Good data governance is critical to ensuring that everyone trusts the data. Validate and verify with subject matter experts. If it’s not trustworthy, don’t use it. Once you have it right, use the acronym OMAR to help you organize and present:
- Objective – What was your objective as you prepared the analysis?
- Method – Where did the data come from and what did you do to it to achieve your objective?
- Assumptions – What assumptions or trade offs did you have to make? How we choose to look at the inputs of the analysis can influence the outcome – you should be mindful that someone else might get to a different conclusion based on a different set of assumptions.
- Results – What’s the conclusion and how does that translate into action?
It’s considerations like these that seem so obvious yet get missed so often in so many projects, with new client needs and deadlines. You don’t have to channel Edward Tufte in every deliverable but keeping these principles in mind help our teams develop for clients and most importantly, help us partner to make the products better over time.