Insight
Effective Data Visualization: Building Reports Execs Will Trust

>>Effective Data Visualization: Building Reports Execs Will Trust

Effective data visualization is something that those in healthcare — and other industries — are focused on. All to often, executives don’t trust reports and they question their results. This distrust grows as the dashboards and reporting works its way up the chain and into the c-suite. This distrust grows as the dashboards and reporting works its way up the chain and into the c-suite.

The CFO wants business value and financial performance. The CNO wants care quality metrics like HCAHPS, Press Ganey, or patient feedback.  The CMO wants physician and hospital metrics across quality, productivity, research, and cost. And, well, the COO wants it all, but at a higher level.

Because of this distrust, meetings derail, projects are delayed, and decisions falter when data and dashboards are questionable.

This article covers four reasons leaders don’t trust dashboards and reports, and ways to cure “death by dashboard” through effective techniques, including four tips to design visualizations that convey the right story with data that each type of executive can trust.

Without Effective Data Visualization, we’ll continue to lose trust with execs.

Ineffective data visualizations can distort critical messages, failing to lead to meaningful (or accurate) conclusions while causing the business to take detrimental action.

Generally, organizations struggle with maintaining balance when it comes to building beautiful and meaningful visualizations, meeting the need for speed, and deriving (and using) measures that matter.

Here are three things standing in the way of effective data visualization:

Reason 1:
Not sharing the right data with the right people.

An impressive looking dashboard is not enough. It is important that it measures the right KPIs and includes data that is most important to those who will contribute, manage, and leverage the dashboard.

Reason 2:
Including too much information.

Huge amounts of data are being generated that can offer a wealth of insights, but it is easy to get carried away and utilize non-essential or unrelated data, making a dashboard ineffective.

Reason 3:
Considering the dashboard as an end and not as a means to an end.

“Dashboarding” data is not a solution to a business problem. It can provide a means to reach a solution if the dashboards capture the right measures and help derive effective insights.

Dashboards fail because we don’t know how to act on the insights derived from them.

Effective data visualization builds trust.

Effective data visualizations deliver immediate, actionable, and aesthetically intriguing insights at a glance. Delivering data that various roles can easily recognize, consume, and accept goes a long way towards building trust and avoiding many hours of re-work and rebuilding credibility.

We believe that the path to building trust that’s sustainable has a few steps:

Step 1:
Align Dashboard Design to Mission, Using FAST Goals© Methodology

The SDLC approach involves documenting the program mission and objectives using the FAST Goals© methodology and collaboratively developing a relevant measurement list.

This is followed by designing a balanced dashboard with interactive views and drilldowns for different users, ensuring best practices are used to visually present data that aligns with goal metrics.

A FAST Goals© diagram relates WHAT are the business goals to HOW to achieve them with validation as to WHY every sub goal, or critical process, is being performed. Each is instrumented with success measures. Strategies and tactics are added and instrumented, as well as a dashboard is designed based on this aligned system, providing measures that matter to each type of user.

Step 2:
Choose the Best Charts

The data visualization can answer key questions when it connects to the goal and the metrics that illuminate progress or achievement of that goal. These “measures that matter” naturally emerge from the FAST Goals© methodology.

The most effective data visualizations best reflect those metrics and ensures it answers the “top of mind” questions for the executive or other role/user.  These could be descriptive in nature (operational performance questions) as well as decision-support in nature.   Additionally, every chart should have a clear set of operational definitions to ensure interpretation clarity, documented action and decisions that accelerate progress.

Download our easy reference chart to choose the most effective data visualization for your needs. For example, a Bubble Cloud is useful when showing composition across categories based on size differences.

Fast Goals - Choose Best Data chart
Click image to download chart pdf

Step 3:
Build Dashboards with Best Practices

Building useful (and balanced) dashboards is an art and science. When thinking through how best to package and deliver the above-mentioned visualizations in a way that will meet user needs, we suggest 10 tips to ensure that your dashboards are used and trusted.

Download our dashboard design tips diagram.

Click image to download pdf

Step 4:
Seek Out Successful Real-World Examples

Most executive teams have their go-to reports or data sources that they like (or hate) to use. Their input and preferences can be invaluable in determining the dashboard-reporting style and culture in your department, team, and organization.

There is nothing better than starting with an example that works.


Below, we offer a real-world client example (de-identified and data changed for confidentiality) to demonstrate how we met a client’s unique needs to develop data tools that support decision-making and are reputable with executives.

Here you can see examples taken from the first three steps shared earlier:

Step 1: Establish Mission by Using FAST Goals© Methodology SDLC Partners worked with a large client’s care management program that is focused on managing the health of high-risk, chronically ill patients. Their care management dashboard is a Tableau web application that helps managers track program progress and take data-driven decisions on the organizational program structure, enrollment status, and process. It is extensible for care plans and outcomes.

We created a common mission with the client’s team using the FAST Goals© methodology.

Step 2: Choose the Best Charts

Next, we looked at the charts available to us and decided which data visualization would meet the datum presented, as well as the user roles along with their preferences for consuming analysis. We chose five types of charts and visualizations.

SDLC - Care Management Dashboard
Click image to enlarge

Why these metrics and for whom?

The enrollment/disenrollment/assignment statistics, number of days until first contact, and reasons for not enrolling enable the program executives and managers to track the program progress and to take data-driven decisions on the organizational structure of the program. The Care Management metrics help in assessing the care plan effectiveness and objectives closely associated with providing care to enrolled members.

Snapshot:
The dashboard calculates the current month’s snapshot, which provides an overview to the executives and managers about the organization structure for the month.

Drilldown Map:
Drilldown functionality provides a detailed view of each of the metrics broken down by care manager, practice, and individual patients.

Enrollment Trend Chart:
The Enrollment Trend Chart shows the progress towards reducing the amount of time from patient enrollment until the care managers first contact the patients.

Pie Chart:
The pie chart was chosen to show the distribution of reasons that patients have for not enrolling in the program.

Care Management Trend Chart:
The Care Management Trend charts show a change over time for the care plan goals written in S.M.A.R.T format.

Multi-Series Chart:
The Multi-Series Chart for Care Management displays the performance of care managers in terms of achieving the desired number of interactions with their patients based on pre-set interaction targets determined by the patient’s situation and complexity of care plan.

Step 3: Build Dashboards with Best Practices
Finally, we built the dashboard using the best practices offered above. Out of the 10 best practices we shared earlier, these were the most important for this use case and client:

  1. Use the measures that matter: The FAST© Goals methodology generates a list of the most important metrics needed in the dashboard. That way, the goals tie to the metrics and the dashboard reflects prioritized information and the client can easily maintain focus on the most important information for decision-making and achieving goals. Also, it enables balanced attention to multiple dimensions of measurement.
  2. Keep it relevant: The dashboard is built with user-level security, enabling different views of the data — making it relevant to each end user.
  3. Choose the right chart type: Choosing the wrong visual aid can cause confusion and lead to mistaken data interpretation.
  4. Add interactivity to encourage exploration: The dashboard provides ability to view, drill down, and analyze historical trends.

Value Realized

Our three-step process helped us collaborate with the client in a focused and meaningful way and create a dashboard and data visualizations that are easy-to-use, used, and most importantly, trusted.

Now, the client has a dashboard “story” that aligns with their goals and the operational cadence of the care management program. They have a one-stop destination for all users to track overall program metrics, as well as their own enrollment progress. Their dashboard provides the ability to view, drilldown, and analyze historical trends with a tailored view for different types of users. Lastly, they have a secure way to view and share data among different users they didn’t have previously.

If you’re facing “death by dashboard” or tired of the wasted time and insecurity of distrusted metrics and reporting, consider these reasons and steps to creating more effective data visualizations.

Jeannine Siviy

Insight Author

Jeannine Siviy, Director of Healthcare Solutions

Insight Contributor

Priyanka Kadam,Associate Consultant, BI & Analytics