For any problem tied to big stakes, there are likely several solutions. Some solutions are quick fixes; some solutions expire; some solutions are just too hard to achieve. The bigger the stakes —for example, helping people and businesses thrive—the more our team likes to rely on data to inform our decision-making.
At Search Discovery, we love helping clients build frameworks for data-informed decision making, but this post is a very practical walkthrough of how we helped a large healthcare client choose performance dashboard solutions that save time, access new insights, and empower leaders to act quickly to achieve business goals tied to serving patients better.
Attention to our client’s needs combined with technical expertise led to optimized hospital performance dashboard solutions.
"To pay attention, this is our endless
and proper work."
Multiple solutions within a hospital performance dashboard
Domo is a BI platform that lets users customize the key performance indicator views they most need to focus on. Features such as Filter Views and Interactive Filters allow users to determine the optimal solution given specific project requirements. This flexibility comes in handy when you’re comparing potential solutions with only (apparently) slight differences.
In that case, it’s critical to understand both the project requirements and how end-users will be interacting with the platform. When you know these objectives, you’ll be able to more deeply analyze the options and determine the best solution for the specific situation.
Both cases detailed below share a common theme: Business users needed to be able to see high-level data across large sections of the business and the ability to dive deeply into the data so they could analyze root causes of issues they discovered. Armed with a good understanding of the client’s business requirements and how stakeholders would interact with the solutions, we evaluated the pros and cons of solution choices with an end goal in mind.
Scenario One—Creating a customized hospital performance dashboard from EHR (Electronic Healthcare Record) data with streamlined views for multiple users
Our healthcare client needed a dashboard built in Domo that various teams could leverage. For example, some teams are responsible for performance across an entire region, while others only look at performance for specific hospitals. All teams want to be able to see only the performance for the areas they are responsible for, so they can quickly assess performance and analyze the data that’s relevant to them. There are multiple ways to approach this and create a solution in Domo.
One potential solution was to create multiple dashboards with cards filtered to the specific hospitals and/or regions.
- Pros: This would both make it easy to share specific pages to specific people or groups and easy to understand what was being displayed. Also, PDP (Personalized Data Permissions) would not likely be needed, and each page could be shared to a specific set of users.
- Cons: Multiple copies of the dashboard would need to be made, and card filters would need to be updated to reflect the specific view required. This could be very time-consuming/redundant. Also, if any changes were requested, they would need to be made in every single dashboard/card.
The second potential solution was to create and leverage Filter Cards, which would create filters that sit above the dashboard and allow users to filter across multiple dimensions to zero in on their area of interest.
- Pros: This would not be as time-consuming as creating multiple dashboards.
- Cons: A lot of training would be needed to ensure users understood how to toggle between the filters to accurately see what they want to see.
The third solution and the one we ultimately chose was to use Filter Views to create multiple “dashboards.”
- Pros: We chose this option because the data is drawn into a single dashboard, so any updates made would only need to be made in one place. This solution also made it easier to toggle between views and not go back and forth to multiple pages to see what you want to see. In addition, new filters can be created relatively easily, and users can even create custom filter views that are not shared if they have a specific view they want to see that’s not relevant to other users.
- Cons: There is a bit of end-user training involved with this solution to ensure users understand how the views are set up and how to toggle between views. PDP would still need to be set up in the event any users should not be able to see some aspects of the data since you cannot share views with specific users. Load time can be slow since Filter Views is still in Beta.
Above, you can see the solution we created in Domo. The user can easily toggle between the different Filtered Views set up in the upper left-hand corner. These Filtered Views are all designed to filter specific hospitals and specific regions around those hospitals, and all of the cards on the dashboard will be updated to reflect those particular filter criteria.
This means the client now has a new, single dashboard to easily get all the insights they need without going to nine individual dashboards. In turn, these insights make it easier for the client to clearly understand which hospital groups are performing well and which aren’t, so they can determine how to focus their time and energy and quickly take action to work with hospital leaders in underperforming groups.
The added benefit is that any changes/additions can all be made in a single dashboard rather than on nine separate dashboards, saving a ton of time.
Scenario Two—Creating a healthcare business intelligence performance dashboard from EHR data that allows both high-level and granular views of key performance indicators
Our healthcare client needed to see high-level patient volume, market share, or customer sentiment trends and needed the ability to drill in to see more detail upon observing swings in performance.
The first potential solution was to use Filter Cards.
- Pros: These are relatively easy to set up. It’s also pretty clear what you are looking at when you filter on something.
- Cons: For more extensive dashboards, you would likely need to set up the same card filters multiple times in the various sections to avoid confusion about what you have filtered on.
The next potential solution: Page Filters.
- Pros: Of all the possible solutions, they are the quickest to set up. They’re likely less confusing than Card Filters because you always see the page filters as you scroll down the page.
- Cons: They tend only to work well if the filters can be used across the entire dashboard. Also, the setup can be a bit confusing as to what datasets/cards you want the filter to interact with.
Potential solution number three was to use Drill Paths.
- Pros: From a user perspective, when set up correctly, Drill Paths facilitate a pretty seamless experience.
- Cons: Drill paths are specific to the view you have set up. They’re less flexible than other solutions because the only interaction is in the individual card you are drilling into.
We ultimately chose to employ yet another potential solution: Using the dimensions within the cards on the dashboards to act as filters on the rest of the page.
- Pros: This lets our client see overall trends for an element or a breakdown of other dimensions. A user can filter right where she is on the page, and all cards will be filtered to what she selects. Within a few clicks, the user can quickly identify what areas of the business are contributing to a decrease or increase in performance. This solution is excellent for more extensive dashboards like our client’s because there’s no need to add filter cards that would further increase the dashboard’s size.
- Cons: This solution requires user training because it’s not intuitive the first time around, and users seem to specifically struggle with how to get back to the original view. This solution also does not work with drill-in-place cards.
What this means for the client is that they can quickly hone in on potential problem areas. They can achieve deeper analysis and discover insights faster. This solution helped our client answer the following drill-down questions that they struggled to answer before:
- Are specific service lines underperforming?
- Are they trending back up, or is this an ongoing problem?
- For an underperforming service line, is it isolated to a specific hospital or group of hospitals? Or could it be due to a particular type of insurance?
Now, all of these questions can be answered within seconds. In addition, Search Discovery’s training and capacity-building approach helped alleviate issues and put structures in place to clarify potential confusion with how to get the most out of this type of dashboard.
Clicking on “CARDIOLOGY” updates the trend card above it to show trends specifically for that Service Line, so you can see if the trend is going up or down over time.
This expanded view also shows what the breakdown is by Hospital and Insurance Group, so you can see if the decline is more pronounced in a specific hospital or for one particular insurance group.
Back on the first image, clicking on “Gastroenterology” updates the trend card above it and the Hospital System and County cards. This allows you to see trended performance for “Gastroenterology” and the breakdown by the hospital system and county to understand which hospital systems are getting the majority of Gastroenterology surgeries and which counties have the most opportunity to get new patients.
How Expertise and Attention Work Together
These use cases demonstrate how the right BI solution can do so much more than make data accessible or make it pretty; it makes data more useful.
Our client has adopted our solution and replaced many of their old standardized reports in favor of these dashboards. As a result, information that used to take multiple reports and weeks of effort is now available to them in one view, in near real-time. And the dashboards enable them to compare overall performance with data-deep-dives, allowing them to answer questions they never could before and, ultimately, better serve their patients.
The solutions are robust enough to be adapted to any similar use case that requires a high-level view across multiple divisions, regions, or other key business dimensions, combined with the ability to drill into specific areas.
Looking for more Domo-related content? Check out these posts: